boron removal and recovery

8/19/2019 Boron Removal and Recovery

1/11

Boron removal by electrocoagulation and recovery

Mohamed Hasnain Isa a, Ezerie Henry Ezechi a, Zubair Ahmed b,*,Saleh Faraj Magram b, Shamsul Rahman Mohamed Kutty a

a Civil Engineering Department, Universiti Teknologi PETRONAS, 31750 Tronoh, Perak, Malaysiab Department of Civil Engineering, King Abdulaziz University, Jeddah, Saudi Arabia

a r t i c l e i n f o

Article history:

Received 4 October 2013

Received in revised form

9 December 2013

Accepted 16 December 2013

Available online 27 December 2013

Keywords:

Adsorption kinetics

Boron

Electrocoagulation

Hydrothermal mineralization

Produced water

Response surface methodology

Thermodynamics

a b s t r a c t

This work investigated the removal of boron from wastewater and its recovery by elec-

trocoagulation and hydrothermal mineralization methods respectively. The experimental

design was developed using Box-Behnken Model. An initial study was performed based on

four preselected variables (pH, current density, concentration and time) using synthetic

wastewater. Response surface methodology (RSM) was used to evaluate the effect of pro-

cess variables and their interaction on boron removal. The optimum conditions were ob-

tained as pH 6.3, current density 17.4 mA/cm2, and time 89 min. At these applied optimum

conditions, 99.7% boron removal from an initial concentration of 10.4 mg/L was achieved.

The process was effectively optimized by RSM with a desirability value of 1.0. The results

showed that boron removal efficiency enhanced with increase in current density and

treatment time. Removal efficiency also increased when pH was increased from 4 to 7 and

subsequently decreased at pH 10. Adsorption kinetics study revealed that the reaction

followed pseudo second order kinetic model; evidenced by high correlation and goodness

of fit. Thermodynamics study showed that mechanism of boron adsorption was chemi-

sorption and the reaction was endothermic in nature. Furthermore, the adsorption process

was spontaneous as indicated by negative values of the adsorption free energy. Treatment

of real produced water using electrocoagulation resulted in 98% boron removal. The hy-

drothermal mineralization study showed that borate minerals (Inyoite, Takadaite and

Nifontovite) can be recovered as recyclable precipitate from electrocoagulation flocs of

produced water.

ª 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Boron is an essential compound for the manufacture of

different products. Boron compounds are widely used in the

manufacture of glass, ceramics, high quality steel, catalysts,

cosmetics and flame retardants (Yilmaz et al., 2008a). Boron is

alsoan essential micronutrient for plants and is readily present

inthe formof boric acid(H3BO3). Boron exists as undissociated

boric acid and borate ions in aquatic environment. The func-tions of boron in plants include degradation of carbohydrates,

sugar translocation, and hormonal action. Boron deficiency

causes stunted growth, yield loss and even death of plant

(Yilmaz et al., 2008b). High boron concentration in irrigation

water, however, can cause severe environmental problem

because boron compounds form complexes with heavy metals

present in soil and increase the potential toxicity of these

* Corresponding author. Tel.: þ966 (0)2 6402000x68239; fax: þ966 (0)2 6952179.E-mail addresses: [email protected], [email protected] (Z. Ahmed).

Available online at www.sciencedirect.com

ScienceDirect

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / w a t r e s

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3

0043-1354/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved.

http://dx.doi.org/10.1016/j.watres.2013.12.024

https://www.researchgate.net/publication/229092678_An_empirical_model_for_kinetics_of_boron_removal_from_boroncontaining_wastewaters_by_the_electrocoagulation_method_in_a_batch_reactor?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/229092678_An_empirical_model_for_kinetics_of_boron_removal_from_boroncontaining_wastewaters_by_the_electrocoagulation_method_in_a_batch_reactor?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/229092678_An_empirical_model_for_kinetics_of_boron_removal_from_boroncontaining_wastewaters_by_the_electrocoagulation_method_in_a_batch_reactor?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==mailto:[email protected]:[email protected]://www.sciencedirect.com/science/journal/00431354http://www.elsevier.com/locate/watreshttp://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024https://www.researchgate.net/publication/229092678_An_empirical_model_for_kinetics_of_boron_removal_from_boroncontaining_wastewaters_by_the_electrocoagulation_method_in_a_batch_reactor?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://www.elsevier.com/locate/watreshttp://www.sciencedirect.com/science/journal/00431354http://crossmark.crossref.org/dialog/?doi=10.1016/j.watres.2013.12.024&domain=pdfmailto:[email protected]:[email protected]


2/11

complexes when passedto groundwater(Sekietal.,2006). High

boron concentrationin surface water makes thewaterunfit for

consumption because boron has been shown to induce male

reproductive impediments in laboratory animals orally

exposed to boric acid and borax (Linder et al., 1990). The World

Health Organization (WHO, 2011) has set a guideline value for

boron concentration in drinking water as 2.4 mg/L.

Presence of boron in various industrial wastewaters, suchas produced water, can cause problems in wastewater recla-

mation and reuse. Produced water is water trapped in un-

dergroundstratawhich is brought to the surface together with

oil and gas during drilling. Produced water is reportedly the

largest waste-stream of oil and gas exploration with an esti-

mated 250 million barrels per day compared with about 80

million barrels per day of oil worldwide (Fakhruâul-Razi et al.,

2009). The composition of produced water differs from other

wastewater because produced water has beenconfined within

underground formations for a very long time (Ezechi et al.,

2012a). Produced water is being considered as a supplement

to limited freshwater resource especially in arid areas because

of its large production volume. One of the impediments to thisusage is the presence of boron at higher than permissible

concentrations. The large volumetric generation of produced

water would also suggest the potential for high amount of

boron recovery.

Boron removal from wastewater presents several chal-

lenges. Membrane process is a widely acceptable method for

wastewater treatment. However, studies have shown that

boron can diffuse through membranes in a non-ionic way,

similar to that of carbonic acid or water (Hou et al., 2010). The

use of selective ion exchange chelating resins has been shown

to be effective in boron removal (Kabay et al.,2004). Disposal of

the subsequently generated sludge and periodic regeneration

of resin, however, remain as major challenges. On the otherhand, conventional biological process only removes a small

amount of boron from wastewater due to its antiseptic nature

(Malakootian and Yousefi, 2009).

Electrocoagulation as a treatment process has been used in

the removal of various water contaminants. Process versa-

tility, sludge reduction, minimal operator attention and ease

of operation are some of its advantages. The major action of

electrocoagulation depends on the ability of water particles to

respond to strong electric field in a redox reaction ( Ezechi

et al., 2010b). Electrocoagulation involves three major mech-

anisms; formulation of coagulants by electrolytic oxidation of

sacrificial anodes, destabilization of the contaminants and

particulate suspension, breaking of emulsions and aggrega-tion of the destabilized phases to form a floc (Babu et al., 2007).

The mechanism of aluminium oxidation during electro-

coagulation is shown below (Balasubramanian et al., 2009).

Anode:

Al ðsÞ/Al3þ ðaqÞ þ 3e (1)

Cathode:

2H2O ðlÞ þ 2e / H2 ðgÞ þ 2OH ðaqÞ (2)

Aluminum forms polymeric speciesduring oxidation of the

sacrificial anode. These polymeric species Al6(OH)153þ,

Al7(OH)174þ, Al8(OH)20

4þ, Al13O4(OH)247þ, Al13(OH)34

5þ, etc. transform

finally into Al(OH)3(s) according to the following simplified

equation (Ghosh et al., 2008).

Al3þ ðaqÞ þ 3H2O ðlÞ/AlðOHÞ3 ðsÞ þ 3Hþ ðaqÞ (3)

The formed Al(OH)3 (s) appears as sweep flocs with large

surface area which increases its adsorption capacity and aids

in boron removal from solution. The formed flocs are sepa-

rated from aqueous medium by sedimentation or flotation.

Considering that many landfill sites are filled up and

finding new landfill sites is difficult. Recovery of boron will not

only mitigate the adverse effect of boron in the environment

but also provide a means of producing boron compounds for

industrial use.

This study focuses on the use of electrocoagulation (EC) for

boron removal from aqueous solution and its recovery. The

specific objectives are: (a) to optimize EC removal of boron

based on significant operating parameters using response

surface methodology (RSM), (b) to study the boron adsorption

kinetics and thermodynamics, and (c) to determine the poten-

tial recovery of boron by hydrothermal mineralization (HM).

2. Experimental method

2.1. Characteristics of wastewater

A preliminary study was conducted with synthetic waste-

water prepared with appropriate amount of boric acid (H 3BO3)

and dissolved in 1 L distilled water to yield varying boron

concentrations of 10, 20 and 30 mg/L. The produced water was

collected from a local Crude Oil Terminal in Malaysia and was

characterized with atomic absorption spectrometry (AAS) and

ion chromatography (IC). A pH meter (Hach Sension 2 pHmeter) and a conductivity meter (Myron L conductivity meter)

were used to measure the pH and conductivity of the sample

respectively. The produced water characteristics are shown in

Table 1.

Table 1 e Produced water characteristics.

Parameter Concentrationa Parameter Concentration

Boron 15 Bromine 31.2

pH 7.84 Total

phosphate

12

TSS 136 COD 1560

TDS 15,829 BOD 883Conductivity 30,000 mS/cm Nitrite 0.03

Turbidity 72 NTU Copper 2.98

Aluminum 0.65 Ammonia

nitrogen

16.5

Iron 1.66 TKN 60.7

Chloride 7546 Sulphate 168

Sodium 3952 Nitrate 1.9

Calcium 357 Sulphide 0.21

Magnesium 600 Phenol 15

Sulphate 168 Total alkalinity 1546

Potassium 284 Zinc 0.04

Hardness 957 Fluoride 0.61

a All concentrations are expressed in mg/L unless stated

otherwise.

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3114

https://www.researchgate.net/publication/6998762_Removal_of_boron_from_aqueous_solution_by_adsorption_on_Al2O3_based_materials_using_full_factorial_design?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/6998762_Removal_of_boron_from_aqueous_solution_by_adsorption_on_Al2O3_based_materials_using_full_factorial_design?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/6998762_Removal_of_boron_from_aqueous_solution_by_adsorption_on_Al2O3_based_materials_using_full_factorial_design?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/20941985_Effect_of_acute_exposure_to_boric_acid_on_the_male_reproductive_system_of_the_rat?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/20941985_Effect_of_acute_exposure_to_boric_acid_on_the_male_reproductive_system_of_the_rat?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/20941985_Effect_of_acute_exposure_to_boric_acid_on_the_male_reproductive_system_of_the_rat?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/41040003_Boron_removal_from_aqueous_solution_by_direct_contact_membrane_distillation?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/41040003_Boron_removal_from_aqueous_solution_by_direct_contact_membrane_distillation?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/41040003_Boron_removal_from_aqueous_solution_by_direct_contact_membrane_distillation?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/223930858_Removal_and_recovery_of_boron_from_geothermal_wastewater_by_selective_ion_exchange_resins_I_Laboratory_tests?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/223930858_Removal_and_recovery_of_boron_from_geothermal_wastewater_by_selective_ion_exchange_resins_I_Laboratory_tests?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/223930858_Removal_and_recovery_of_boron_from_geothermal_wastewater_by_selective_ion_exchange_resins_I_Laboratory_tests?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==http://-/?-http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024https://www.researchgate.net/publication/41040003_Boron_removal_from_aqueous_solution_by_direct_contact_membrane_distillation?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/223930858_Removal_and_recovery_of_boron_from_geothermal_wastewater_by_selective_ion_exchange_resins_I_Laboratory_tests?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/20941985_Effect_of_acute_exposure_to_boric_acid_on_the_male_reproductive_system_of_the_rat?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/6998762_Removal_of_boron_from_aqueous_solution_by_adsorption_on_Al2O3_based_materials_using_full_factorial_design?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://-/?-


3/11

2.2. Electrocoagulation setup

The electrocoagulation setup consisted of a 500 mL beaker

with six aluminium plate electrodes of size

10 cm 1 cm 0.3 cm. Effects of different operating param-

eters (pH, current density, initial boron concentration and

treatment time) were investigated. The electrodes were con-

nected to a digital DC supply characterized by the ranges of 0e3 A for current and 0e30 V for voltage. A digital ammeter-

voltmeter was used to regulate the current and voltage.

After each run of the experiments, the used aluminium elec-

trodes were dipped in acetone solution for 10 min and rinsed

with deionized water and dried for 10 min at 105 C to remove

surface impurities before reuse. Samples were let stand for 2 h

to allow concentration of flocs prior to analysis for boron.

Boron concentration was analysed according to standard

method using carmine reagent. All results are average of three

analyses. Chemicals and reagents used were analytical grade

(Merck).

2.3. Adsorption kinetics

Boron concentration for the adsorption study was varied in

the range 10e30 mg/L using a 500 mL beaker. Current density

12.5 mA/cm2, pH 7, temperature 308 K, and inter-electrode

spacing 0.5 cm were kept constant. Supernatant was

collected at different times and analyzed for residual boron

concentration. The amount of boron adsorbed at equilibrium

(qe) was calculated using the following equations:

qe ¼ ðC0 CeÞ V

W (4)

W ¼ ITMZF (5)

where C0 is initial boron concentration (mg/L), Ce is equilib-

rium boron concentration (mg/L), V is volume of sample (L), W

is mass of adsorbent (g), I is current (A), T is time (s), M is molar

mass of electrode, Z is number of electrons involved in the

redox reaction, and F is Faraday’s constant (C/mol).

The suitability of both pseudo first order and pseudo sec-

ond order kinetic models was further evaluated using the chi-

square (c2) represented as follows (Sundaram et al., 2008):

c2 ¼

qexpe qcale

2

qcale(6)

where qexpe is experimental adsorption capacity at equilibrium(mg/g), and qcale is calculated adsorption capacity at equilib-

rium (mg/g).

The c2 test measures the goodness of fit between the

experimental equilibrium adsorption capacity and the calcu-

lated equilibrium adsorption capacity. The value of c2 for the

applicable model should be lowest. A good correlation coeffi-

cient and a low c2 indicate that the model is applicable.

2.4. Adsorption thermodynamics

Thermodynamic parameters which include free energy

change (DG0), enthalpy of reaction (DH0) and entropy change

(DS0) can be used to deduce the mechanism of a reaction.

Observations were made at four different temperatures,

controlled using a thermostatic warm water bath, to deter-

mine these parameters. The thermodynamic constant was

evaluated using the following equations (Shen et al., 2009):

DG0 ¼ RTlnKc (7)

ln Kc ¼

DS0

R

DH0

RT (8)

where Kc is distribution coefficient, R is thermodynamic gas

constant (8.314 J/mol.K), and T is temperature (K).

2.5. Boron recovery with hydrothermal mineralisation(HM)

The flocs produced during electrocoagulation were collected

after settling. They were transferred into an evaporating dish

and placed in the oven at room temperature for 24 h. Thedried

flocs were kept in the desiccator for 20 min before grounded

and 2 g of the flocs were accurately weighed to recover boron.

40 mL of 3 M HNO3 was used to dissolve the flocs. The solutionwas placed in an orbital shaker for one hour at 150 rpm to

enable complete dissolution. The pH of the solution was

adjusted to pH 10 with 1 M NaOH and 0.3 g calcium hydroxide

Ca(OH)2 was used as the mineralizer. Hydrothermal mineral-

ization was conducted with a conventional oven for 2 h at

120 C. Thereafter, the solvent was collected for boron anal-

ysis and the precipitate was analysed using Simens diffrac-

tometer (Model D5000) with graphite monochromated Cu Ka

source operated at 40 kV and 40 mA. The XRD spectrum was

obtained at scanning angles (2q) ranging from 5 to 150 and at

scanning speed of 0.04 per second. The microstructure

properties were analyzed with scanning electron microscope

(SEM).

3. Statistical methods and data analysis

Response surface methodology was employed to determine

the optimum levels of process parameters. RSM uses a

collection of mathematical and statistical techniques to

analyse the effects of several independent variables on the

response. It is often used in process design, improvement and

optimization. The methodology is practical as it employs

experimental data and thus includes the interactive effects of

variables on the overall process performance. Box-Behnken

design was established with the help of the Design Expert6.0.7 software for statistical design of experiment and data

analysis. The four significant process variables (Yilmaz et al.,

2008b) considered in this study were: pH (A), current density

(B), initial boron concentration (C) and time (D) as shown in

Table 2. Synthetic wastewater was used to determine the

optimum treatment conditions. The total number of experi-

ments in this study was 29 including five replicates at the

centre point for the estimation of error. A second-order

polynomial model (Equation (8)), using Design Expert soft-

ware, was fitted to the experimental data obtained according

to the Box-Behnken design; where Y is the response, Xi and X jare variables, bo is a constant coefficient, b j, b jj and bij are

interaction coefficients of linear, quadratic and the second

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3 115

https://www.researchgate.net/publication/23445024_Kinetics_and_thermodynamics_of_sorption_of_nitroaromatic_compounds_to_as-grown_and_oxidized_multiwalled_carbon_nanotubes?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/23445024_Kinetics_and_thermodynamics_of_sorption_of_nitroaromatic_compounds_to_as-grown_and_oxidized_multiwalled_carbon_nanotubes?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==https://www.researchgate.net/publication/23445024_Kinetics_and_thermodynamics_of_sorption_of_nitroaromatic_compounds_to_as-grown_and_oxidized_multiwalled_carbon_nanotubes?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024https://www.researchgate.net/publication/23445024_Kinetics_and_thermodynamics_of_sorption_of_nitroaromatic_compounds_to_as-grown_and_oxidized_multiwalled_carbon_nanotubes?el=1_x_8&enrichId=rgreq-5082a3a3-ef35-4aaf-8b14-6a979260b59e&enrichSource=Y292ZXJQYWdlOzI1OTY5ODY2ODtBUzoxOTYzNjM2Nzg2ODcyNDVAMTQyMzgyODE1Nzc0OA==http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024


4/11

order terms respectively, k is the number of studied factors

and ei is the error. The coded values of the process parameters

in Eq. (9) could be determined by Eq. (10) where Xi is the

dimensionless coded value of the ith independent variable, xiis the uncoded value of the ith independent variable, x0 is the

uncoded value of the ith independent variable at the center

point and Dx is the step change value between low level (1)

and high level (þ1) (Zhang et al., 2011).

Y ¼ b0 þXk

j¼1

b jX j þXk

j¼1

b jjX2 j þ

XX

i F < 0.0001) indicate that the

model is significant. Values of Prob > F less than 0.05 indicate

that the model terms are significant. ANOVA results for the

response surface quadratic model are summarized in Table 5.Adequate precision compares the range of the predicted

values at the design points to the mean prediction error. Its

value greater than 4 is desirable and confirms the applicability

of the model for navigation of the design space (Zinatizadeh

et al., 2007). The adequate precision of 23.088, in the present

case, shows that the model is acceptable. The R2 value of

0.9769 is in reasonable agreement with the model adjusted R2

value of 0.9538 and predicted R2 value of 0.8740. The agree-

ment is desirable for a good fit of a model (Mohajeri et al.,

2010). The R2 value shows that the process can explain about

97% of the model output. A significant lack of fit suggests that

there may be some systematic variation unaccounted for in

the hypothesised model. The lack of fit in this study is not

significant which is good for the model. In this study, A, B, C,D, A2, C2 are significant model terms. Insignificant model

terms have limited influence on the model and were excluded.

Based on the results, the response surface model constructed

in this study for predicting boron removal efficiency was

considered reasonable. The final regression model (second-

order polynomial equation) in terms of coded factors is

expressed in Equation (11):

B removal% ¼ 72:90 þ 4:35A þ 12:54B 6:49C þ 9:85D

33:53A2 þ 4:32C2 (11)

The suitability of the selected model to provide adequate

approximation of the real system is also confirmed by the

diagnosticplots. Such plots include normalprobabilityplots of

the studentized residuals and the predicted versus actual

Table 2 e Independent variables of the Box-Behnkendesign.

Level pH Current density(mA/cm2)

Initial boronconcentration (mg/L)

Treatmenttime (min)

1 4 6.25 10 30

0 7 12.5 20 60

þ1 10 18.75 30 90

Table 3 e Response (boron removal) values for different experimental conditions.

Run no. pH Currentdensity

(mA/cm2)

Initial boronconc. (mg/L)

Treatmenttime (min)

Boronremoval (%)

1 10 6.25 20 60 26

2 4 12.5 10 60 50

3 7 12.5 30 30 55

4 7 12.5 20 60 76

5 7 18.75 20 90 94

6 4 12.5 20 30 37.5

7 7 12.5 20 60 74

8 7 12.5 20 60 73

9 7 18.75 20 30 68

10 7 18.75 10 60 97

11 7 12.5 30 90 82.7

12 7 6.25 20 30 49.5

13 4 12.5 20 90 50

14 7 12.5 20 60 71.5

15 10 12.5 20 30 26.5

16 7 18.75 30 60 87.5

17 7 6.25 30 60 52

18 10 12.5 30 60 33.6

19 10 12.5 20 90 44

20 4 18.75 20 60 53

21 7 12.5 10 90 98

22 4 6.25 20 60 34

23 7 6.25 10 60 67

24 10 18.75 20 60 43.5

25 7 6.25 20 90 64

26 7 12.5 10 30 78

27 7 12.5 20 60 70

28 10 12.5 10 60 41

29 4 12.5 30 60 42.3

Table 4 e Analysis of variance result for significant modelterms.

Sum of squares

DF Mean square F value Prob > F

Boron

removal

A 227.07 1 227.07 11.13


5/11

value plot. These plots are used to judge the adequacy of a

model. Fig. 1 shows the normal probability plot for the stu-

dentized residuals for boron removal. Studentized residuals

represent normal probability plots where the residuals follow

a normal distribution in which case the points will follow a

straight line. Some scattering is expected even with normal

data. It can be deduced from Fig. 1 that the data is evenly

distributed. As shown in Fig. 2, the predicted and actual values

are in good agreement.

4.2. Boron removal efficiency

The three dimensional (3D) response surface plots (Fig. 3) of

the quadratic model were generated by the Design Expert 6.0.7

software and utilized to assess the interactive effect of the

independent variables on the response. In Fig. 3(a), the 3D

response surface plot was developed as a function of initial pH

and current density at initial boron concentration 10 mg/L and

reaction time 90 min. From the plot, removal efficiency

increased from pH 4 to 7 and decreased towards pH 10. Be-

tween pH 7e8, removal efficiency was near constant. As

mentioned by (Bayramoglu et al., 2004), at pH 4e8, Al3þ and

OH ions generated by electrodes react to form various

monomeric and polymeric species that finally transform intoinsoluble amorphous Al(OH)3 (s) through complex polymeri-

sation. Above pH 10, the highly soluble monomeric AlðOHÞ4

anion concentration increases at the expense of Al(OH)3 (s).

However, the solubility of aluminum hydroxide is less at pH

6e8 (Emamjomeh and Sivakumar, 2009). In the present work,

the removal efficiency was highest (98%) at pH 7. Electro-

coagulation acted as a pH neutralizer at alkaline pH in this

study.At pH10, the finalpH (8.8) was observed tobe lower than

the initial pH. However, at initial pH 4, the final pH (4.4) was

observed to be higher than the initial pH. The increase in final

pH at acidic condition has been attributed to the increase in

hydrogen evolution at the cathode while the decrease in final

pH at alkaline condition has been attributed to the generationof an alkalinity consumer (AlOH)4 (Vik et al., 1984).

The collision between particles, release of coagulants and

amount of coagulants generated at the electrode are

controlled by the electric current. From Fig. 3(a), removal ef-

ficiency increased when current density was increased. In-

crease in current density reduced treatment time. However,

ohmic heating at high current density increases sample

temperature, therefore it may not be feasible to increase

current density beyond 12.5 mA/cm2 in this study. In addition,

energy consumption increases with increased potential. The

optimum current density was observed as 12.5 mA/cm 2.

In Fig. 3(b), the 3D response surface plot was developed as a

function of initial boron concentration and time at currentdensity 10 mA/cm2 and initial pH 8. Increase in initial boron

concentration was found to cause a decline in the removal

efficiency. This can be attributedto thefactthatthe amount of

Table 5 e ANOVA result for response surface quadratic model.

Sum of squares DF Mean square F value Prob > F

Boron removal Model 12069.28 14 862.09 42.25


6/11

metal ions generated at the same current density for a low

boron concentration was insufficient for solutions of higher

boron concentration. It can be seen from Fig. 3(b) that increase

in time resulted in increase in removal efficiency, due to

prolonged interaction between the coagulant (Al(OH)3)andthetarget constituent (boron).

4.3. Optimization

Numerical conditions optimization for boron removal was

carried out using the Design Expert software. The desired goal

for each operational condition (pH, initial boron concentra-

tion, treatment time and current density) was chosen within

the range while the response (boron removal) was defined as

maximum to achieve the highest performance. The pro-

gramme combines the individual desirabilities into a single

number, and then searches to maximize this function. The

model predicted boron removal efficiency of 99.7%

corresponding to pH 6.3, current density 17.4 mA/cm2, initial

boron concentration 10.4 mg/L, and reaction time 89 min;

whereas a removal efficiency of 98% was obtained from the

experiment. It is, therefore, evident that the model is

adequate for prediction of boron removal using

electrocoagulation.

4.4. Adsorption kinetics

To understand the adsorption kinetics of boron using elec-

trocoagulation, four different adsorption models were

evaluated.

4.4.1. Largegren pseudo first order kinetics

The linearized form of pseudo first order equation is shown

below (Boparai et al., 2011).

log

qe qt

¼ log

qe

k1t

2:303 (12)

where qe and qt are the amount of boron adsorbed at equilib-

rium (mg/g) on Al(OH)3 and at any time (t) respectively, and k1(min1) is the calculated pseudo first order rate constant of

adsorption.

If the adsorption follows the pseudo first order kinetics, a

plot of log (qe qt) versus t should be linear. qe and k1 are

calculated from the intercept and slope of the plot of log

(qe qt) versus t respectively. As shown in Fig. 4, the data did

not completely conform to a linear plot. The points deviated

from the straight line. Though the correlation coefficient was

high, the calculated equilibrium adsorption capacity ðqcale Þ

deviated from the experimental equilibrium adsorption ca-

pacity ðqexpe Þ and the chi-square was high. This implies that

the adsorption did not completely follow the pseudo first

order kinetics. The kinetic constants of pseudo first order ki-netics is shown in Table 6.

4.4.2. Pseudo second order kinetics

The linearized form of the pseudo second order equation is

represented below (Ho and McKay, 1998).

t

qt¼

1k2q2e

þ t

qe(13)

Fig. 3 e Three dimensional surface plots of boron removal.

Fig. 4 e First order kinetic plot of different concentrations.

pH 7, current density 12.5 mA/cm2, Inter-electrode spacing

0.5 cm.

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3118

http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024


7/11

where qe and qt are the amount of boron adsorbed at equilib-

rium (mg/g) on Al(OH)3 and at any time (t) respectively, and k2is the calculated pseudo second order rate constant of

adsorption.

If the adsorption kinetics follows the pseudo second order,

the plot of t /qt versus t should be linear. The qe and k2 can be

calculated from the slope and intercept of plot of t /qt versus t.

As shown in Fig. 5, the plot of t /qt versus t was linear for all

concentrations studied. The correlation coefficient was high,

the chi-square was very low and the calculated equilibrium

adsorption capacity ðqcale Þ was in agreement with the experi-

mental equilibrium adsorption capacity ðqexpe Þ. This implies

that the adsorption of boron followed a second order kinetics

as shown in Table 6.

4.4.3. Intra-particle diffusion model

The linearized form of intra-particle diffusion equation is

represented below (Morris and Weber, 1963).

qt ¼ Kidt0:5 þ Ci (14)

where Kid (mg/g min0.5) is a measure of the diffusion coeffi-

cient and Ci is the intra-particle diffusion constant (mg/g).

If intra particle diffusion is the rate limiting step, a plot of

fraction of solute adsorbed against the squareroot of time(t0.5)

should be linear, passing through the origin. Kid and Ci areobtained from the slope and intercept of the graph respec-

tively. Ci is directly proportional to the boundary layer thick-

ness. From the result (Figure not shown), the plot of qt against

(t0.5) was observed to be linear but did not pass through the

origin. This could be due to the boundary layer effect. Addi-

tionally, the values of C i (mg/g) for all boron concentrations

studied were positive. Kid (mg/g min0.5) and Ci (mg/g) of plot of

qt vs t0.5 were found to increase when the initial concentration

of boron was increased. This implies that intra particle

diffusion may not be the only transport mechanism, rather

there could be more than one mechanism involved in the

transport of boron. The adsorption constants obtained for

intra-particle diffusion are shown in Table 7.

4.4.4. Elovich model

The modified form of Elovich equation is represented as

(Chien and Clayton, 1980).

qt ¼ 1b

lnðabÞ þ 1b

lnðtÞ (15)

where a is the initial adsorption rate (mg/g min) and b is the

desorption constant (g/mg) during any experiment.

The Elovich model does not predict anyprecise mechanism

but it is helpful in explaining predominantly, chemical

adsorption on highly heterogeneous adsorbents (Gupta and

Bhattacharyya, 2006). A plot of qt vs ln (t) should be linear.

The adsorption and desorption constants are calculated from

the slope and intercept of the plot of qt vsln(t). When data was

fitted into the Elovich equation, the plot of qt vs ln (t) gave a

straight line (Figure not shown). The initial rate of adsorption

as calculated from the slope was found to increase anddesorption constant as calculated from the intercept was

observed to decrease as boron concentration was increased.

The correlation coefficient was also high. Increase in rate of

adsorption and decrease in desorption constant as concen-

tration increases implies that the process is chemisorption.

The adsorption constant forElovich model is shown in Table 7.

4.5. Adsorption thermodynamics

The enthalpy change (DH0) and entropy (DS0) were calculated

from the slope and intercept of the plot of ln Kc versus 1/T

(Fig. 6). The thermodynamics constants obtained from the plot

of ln Kc versus 1/T is shown in Table 8. As shown in the Table,the free energies were increasingly negative as temperature

was increased. Negative free energies (DG0) indicate that the

Table 6 e Adsorption constants for first and second order kinetics.

Co (mg/L) qexpe (mg/g) Pseudo-first order Pseudo-second order

k1 (min 1) qcale (mg/g) R

2c

2 k2 (g/mg min) qcale (mg.g) R2

c2

10 6.09 0.078 4.45 0.918 0.62 0.017 6.80 0.998 0.08

20 11.22 0.029 6.56 0.898 3.31 0.084 11.90 0.99 0.04

30 16.68 0.037 14.06 0.957 0.49 0.037 17.86 0.989 0.08

Fig. 5 e Second order kinetic plot of different

concentrations. pH 7, current density 12.5 mA/cm2, inter-

electrode spacing 0.5 cm.

Table 7 e Adsorption constants for Intra-ParticleDiffusion and Elovich model.

Co(mg/L)

Intra-particle diffusion Elovich model

ki(mg/g min0.5)

Ci(mg/g)

R2 a (mg/g min) b (g/mg) R2

10 0.36 3.132 0.97 1.06 1.55 0.98

20 0.59 4.809 0.98 1.72 0.99 0.94

30 1.28 5.017 0.98 3.77 0.499 0.97

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3 119



8/11

adsorption was spontaneous for the temperature range eval-

uated and the degree of spontaneity of the reaction increases

with increasing temperature (Li et al., 2010). The positive value

of the standard enthalpy change (DH0 44.8 kJ/mol) implies that

the adsorption process is endothermic. Values of enthalpy

change less than 40 kJ/mol indicates that the process is

physiosorption while values above 40 kJ/mol indicate a

chemisorption process (Ma et al., 2011). Accordingly, in thepresent case, the process is chemisorption. The positive value

of entropy change (DS0 125.04 J/mol) indicates increased

randomness of the solution interface during the adsorption of

boron on the electrocoagulant Al(OH)3.

4.6. Process advantages and operational cost

Process evaluation and operational cost are two important

indices which determine the implementation of a technique.

In analysing these two indices, factors such as other treat-

ment methods, environmental and health implications, elec-

trical energy consumption, electrode consumption, use of

chemicals and sludge disposal are considered. One compari-son of interest to this study is chemical coagulation which

follows the same pollutant removal mechanism as electro-

coagulation. Their major difference is the mode of introduc-

tion of coagulants. Whereas coagulants are continuously

generated over an extended area of the anode material in

electrocoagulation, point addition of coagulants is done with

chemical coagulation. The freshly precipitated flocs generated

in electrocoagulation are more effectively dispersed resulting

in increased adsorptive removal of pollutants (Zhu et al.,

2005). In terms of floc separation, chemical coagulation is

associated with settling and electrocoagulation is charac-

terised by both settling and flotation due to air bubbles

released at the cathode (Holt et al., 2002). During electro-coagulation, the smallest colloidal particles have a higher

probability of coagulation because the electric field sets them

in motion and produces a relatively low amount of sludge

compared to chemical coagulation (Pouet and Grasmick,

1995). Additionally, secondary pollution is mitigated by elec-

trocoagulation because the rate of generation of coagulants is

regulated by the voltage and current while secondary pollu-

tion may occur at high chemical addition using chemical

coagulation (Yildiz et al., 2008). Electrocoagulation sludge is

readily settleable and easy to de-water since it is composed of

mainly metallic oxides/hydroxide and its flocs tend to be

much larger, contain less bound water, have acid resistant

capacity, are more stable and are easily separated by filtration

(Avsar et al., 2007). Chemical coagulation is highly sensitive to

pH change with effective coagulation at pH 6e7 while elec-

trocoagulation has a pH neutralization effect in a much wide

pH range (4e9). In literature, comparative reports favour

electrocoagulation over chemical coagulation (Avsar et al.,

2007; Zhu et al., 2005).The overall cost was investigated to determine the feasi-

bility of electrocoagulation compared to other processes. In

evaluating the overall cost, the following equation was used

(Olmez-Hanci et al.).

OPCost ¼ a Cenergy þ b Celectrode þ c Csludge (16)

where

Cenergy and Celectrode are amount of consumed electricity

(kWh/m3) and amount of consumed electrode material (kg

electrode/m3) respectively. Csludge (kg/m3) is the amount of

sludge generated during electrocoagulation. Unit prices used

in this study (for Malaysia market) were expressed as (a) unit

prices for electrical energy 35 cents Malaysia Ringgits (MYR)/

kWh (Tenaga National Malaysia, 2013), (b) unit price for elec-

trode material (Bayramoglu et al., 2004) ($1.8; current ex-

change in MYR) RM 5.8/kg (c) Cost associated with sludge

handling and disposal 50 cents MYR/kg (MIDA, 2013).

The electrical energy consumption (Cenergy) is an important

parameter which defines the energy usedup fora process. The

electrical energy consumptions at a constant potential of 2 V

was 1.2 kWh/m3, 2.4 kWh/m3 and 3.6 kWh/m3 for 6.25 mA/

cm2, 12.5 mA/cm2 and 18.75 mA/cm2 respectively as calcu-

lated using the equation below. Increase in current density

increased the electrical energy consumption and also

increased the overall cost.

ECC ¼IUT

V (17)

where

ECC ¼ energy consumption (kWh/m3); I ¼ current (A);

U ¼ voltage (V); T ¼ time (h); V ¼ volume (L)

The aluminum electrode consumption (AI consumption)

with a unit of (g Al/g of Boron) removed was investigated using

the Faraday law as expressed below. The electrode con-

sumption increasedfrom 0.4 g Al/g at 6.25 mA/cm2 to0.81 g Al/

g at 12.5 mA/cm2. At 18.75 mA/cm2, there was further increase

in aluminum electrode consumption to 1.22 g Al/g making

current density and the corresponding voltages important

controls of operational cost.

Fig. 6 e ln Kc vs 1/ T (K); pH 7, initial boron concentration

20 mg/L, current density 12.5 mA/cm2, inter-electrode

spacing 0.5 cm.

Table 8 e Thermodynamic constants for boronadsorption.

Temperature (K) Kc DG

(kJ/mol)DH

(kJ/mol)DS (J/mol.K)

298 3.60 3.18

308 6.16 4.66 44.8 125.04

318 8.75 5.73

328 12.86 6.96

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3120



9/11

Al consumption ¼ ItM

ZFW (18)

where

I ¼ current in amperes (A); t ¼ time (s); M ¼ molecular

weight of the electrode material (Aluminum 26.98);

Z ¼ number of electrons involved in the redox reaction

(zAl ¼ 3); F ¼ Faraday constant (96,500 C/mol); W ¼ weight of

treated wastewater (g).

The amount of sludge generated at 6.25 mA/cm 2, 12.5 mA/

cm2 and 18.75 mA/cm2 was 0.037 kg, 0.029 kg and 0.021 kg

respectively. Sludge generation is influenced by electric

charge. The unit cost of NaOH (50%) and H 2SO4 (98%) used in

this study was about 0.018 cents MYR/kg. 0.3e0.5 ml of 1 M

NaOH and H2SO4 was used to control the pH of each experi-

ment. The operational cost for the removal of boron in this

study at optimum conditions is RM 0.88/m3 as depicted in

Table 9. Electrical energy consumption was the largest

contributor to the total operational cost for boron removal.

The unit cost for the electrodialytic treatment of boron

containing wastewater was estimated at $1.27/m3 for a two-

step process removing boron and salinity (Turek et al., 2007)

and MYR 6.33/m3 for adsorption-flocculation (Chong et al.,

2009). With an operational cost of MYR 0.88/m3 at optimum

condition, electrocoagulation has comparative advantage in

terms of cost over some boron removal processes.

In large installations, the operation is feasible using an

electrocoagulation reactor equipped with a secondary sedi-

mentation tank. Installation of parallel perforated rhombus

shaped electrode materials into the reactor will enhance

movement of metal ions, improve coagulation, reduce the

number of electrodes, improve hydrogen bubble generation

and increase floatation of formed flocs. However, residual

aluminum concentration should be controlled by operating at

low electric potential and current. This could increase treat-

ment time but will not affect operational cost because the

increased time is compensated by the low electrical energy

consumption (kWh/m3) and electrode consumption.

4.7. Boron recovery

Real produced water containing 15 mg/L boron was used for

this phase of the study. Electrocoagulation treatment of the

produced water at pH 7, current density 12.5 mA/cm 2, and

reaction time 90 min yielded a boron removal efficiency of

98%. The flocs thus produced were studied for boron recovery.

The electrocoagulation flocs were characterized with X-rayfluorescence (XRF) and analysed with X-ray diffractogramme

(XRD) and Scanning Electron Microscope (SEM) during the

recovery of boron as a recyclable precipitate. The chemical

composition of the flocs was obtained to be 2.9% B2O3, 1.3%

Al2O3, 1.09% Fe2O3, 14.3% CaO, 19.2% MgO, 28.4% NaO, 6.4%

SiO2, 2.4% K2O, and 24.01% loss on ignition. After hydrother-

mal mineralization of EC flocs, the results show about 1.1 mg/

L residue boron concentration indicating about 91.4% recov-

ery. The SEM analysis of the obtained precipitates showed a

fibrous root like structure on an irregular beam shape base-

ment as depicted in Fig. 7.

XRD investigation of the fibrous root like structure on an

irregular beam shape basement is shown in Fig. 8. The graph

showed a bragg reflections possessing broad humps and low

intensity which indicates that the analyzed phase is a shortrange i.e. more amorphous and little crystalline. The chemical

speciation of this amorphous phase can be aluminum hy-

droxide or aluminum oxyhydroxide. This is suspected

because crystallization of Al hydroxides or oxyhydroxides is a

very slow process. It is reported that most Al hydroxides and

Al oxyhydroxides are found to be either poorly crystalline or

amorphous (Dixon and Weed, 1989).

Using Eva DiffraPlus indexing software in combination with

ICDD (International Center for Diffraction Data) database,

Nifontovite (Ca3B6O6(OH)12$2(H2O), Inyoite

(CaB3O3(OH)5$4H2O) and Takedaite (Ca3B2O6$2H2O) were

identified in the flocs. The three identified compounds are

hydrated calcium borate minerals. From the XRD shown inFig. 8, it can be concluded that Nifontovite could be the fibrous

root like structure identified from the SEM result while Inyoite

and Takadaite could be the irregular beam shape basement.

Table 9 e Operational cost analysis for electrocoagulation.

Item Operational parameter values Cost (MYR/m3)

6.25 mA/cm2 12.5 mA/cm2 18.75 mA/cm2 6.25 mA/cm2 12.5 mA/cm2 18.75 mA/cm2

Energy consumption 1.2 kWh/m3 2.4 kWh/m3 3.6 kWh/m3 0.42 0.84 1.26

Electrode plate consumption 0.4 g Al/g 0.81 g Al/g 1.22 g Al/g 0.0024 0.0049 0.0073

Chemicals 10 ml 10 ml 10 ml 0.018 0.018 0.018

Sludge disposal 0.037 kg 0.029 kg 0.021 kg 0.006 0.014 0.019

Total 0.45 0.88 1.3

Fig. 7 e SEM result for floc precipitate after hydrothermal

mineralization at 120 C, 2 h, 0.3 g Ca (OH)2, 2 g

electrochemical coagulation flocs, pH 10.

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3 121



10/11

The amorphous and poor crystalline nature of aluminiumflocs has been reported in literature. It is reported that

aluminum produces poor crystalline precipitates in the pres-

ence of natural organic matter (Masion et al., 1994). This could

be the reason why the XRD result is more amorphous since

produced water contains high concentration of organic

compounds.

5. Conclusions

Presence of boron in produced water is one of the reasons that

hinder its reuse for purposes such as irrigation and drinking.

The electrochemical technique employed in this study wasable to reduce boron concentration to below the WHO

permissible level of 2.4 mg/L. Response surface methodology

was successfully applied to optimise boron removal from

aqueous solution. At optimal operating conditions of pH 7,

current density 12.5 mA/cm2, inter-electrode spacing 0.5 cm,

and treatment time 90 min, 98% removal of boron from pro-

duced water was achieved. The adsorption of boron followed

the pseudo second order rate kinetics. The thermodynamics

study revealed that the adsorption is chemisorption and

endothermic. The adsorption process showed an increased

dispersal of particles in the solution and the adsorption was

spontaneous. Attempt to recover boron as a recyclable pre-

cipitate from electrocoagulation flocs revealed that Inyoite,Takedaite and Nifontovite can be recovered through hydro-

thermal mineralization of the electrocoagulation flocs. How-

ever, selective recovery of individual boron compounds from a

mixture of produced water electrocoagulation flocs, duration

of flocs settling, and flocs homogeneity is still a subject of

further investigation.

Acknowledgement

This work was funded by Deanship of Scientific Research

(DSR), King Abdulaziz University (KAU), Jeddah, under

research grant (no. 135-006-D1434) and Universiti Teknologi

PETRONAS (UTP), for graduate assistantship to the second

author. The authors, therefore, acknowledge with thanks for

the technical and financial support of DSR, KAU and UTP.

r e f e r e n c e s

Avsar, Y., Kurt, U., Gonullu, T., 2007. Comparison of classicalchemical and electrochemical processes for treating roseprocessing wastewater. J. Hazar. Mater 148 (1e2), 340e345.

Babu, R.R., Bhadrinarayana, N.S., Sheriffa Begum, K.M.M.,Anantharaman, N., 2007. Treatment of tannery wastewater byelectrocoagulation. J. Environ. Sci. (China) 19 (12), 1409e1415.

Balasubramanian, N., Kojima, T., Srinivasakannan, C., 2009.Arsenic removal through electrocoagulation: kinetic andstatistical modeling. Chem. Eng. J. 155 (1e2), 76e82.

Bayramoglu, M., Kobya, M., Can, O.T., Sozbir, M., 2004. Operating cost analysis of electrocoagulation of textile dye wastewater.Separ. Purif. Technol. 37 (2), 117e125.

Boparai, H.K., Joseph, M., O’Carroll, D.M., 2011. Kinetics andthermodynamics of cadmium ion removal by adsorption

onto nano zerovalent iron particles. J. Hazar Mat. 186 (1),458e465.

Chien, S.H., Clayton, W.R., 1980. Application of Elovich equationto the kinetics of phosphate release and sorption in Soils1. SoilSci. Soc. Am. J. 44 (2), 265e268.

Chong, M.F., Lee, K.P., Chieng, H.J., Syazwani Binti Ramli, I.I., 2009.Removal of boron from ceramic industry wastewater byadsorption “flocculation mechanism using palm oil mill boiler(POMB) bottom ash and polymer”. Water Res. 43 (13),3326e3334.

Dixon, J.B., Weed, S.B., 1989. Soil Science Society of America Inc.(SSSA).

Emamjomeh, M.M., Sivakumar, M., 2009. Review of pollutantsremoved by electrocoagulation and electrocoagulation/flotation processes. J. Environ. Manag. 90 (5), 1663e1679.

Ezechi, E.H., Isa, M.H., Kutty, S.R.M., 2012a. Boron in producedwater: challenges and improvements: a comprehensivereview. J. Appl. Sci. 12 (5), 402e415.

Ezechi, E.H., Isa, M.H., Kutty, S.R.M., 2010b. Removal of boronfrom produced water by electrocoagulation. In: 10th WSEASInternational Conference on Environment, Ecosystems andDevelopment (EED’12), Switzerland.

Fakhruâul-Razi, A., Pendashteh, A., Abdullah, L.C., Biak, D.R.A.,Madaeni, S.S., Abidin, Z.Z., 2009. Review of technologies for oiland gas produced water treatment. J. Hazar. Mater 170 (2â V“3),530e551.

Ghosh, D., Solanki, H., Purkait, M.K., 2008. Removal of Fe(II) fromtap water by electrocoagulation technique. J. Hazard Mater 155(1), 135e143.

Gupta, S.S., Bhattacharyya, K.G., 2006. Adsorption of Ni(II) on

clays. J. Colloid Interface Sci. 295 (1), 21e32.Ho, Y.S., McKay, G., 1998. A comparison of chemisorption kinetic

models applied to pollutant removal on various sorbents. Proc.Saf. Environ. Prot. 76 (4), 332e340.

Holt, P.K., Barton, G.W., Wark, M., Mitchell, C.A., 2002. Aquantitative comparison between chemical dosing andelectrocoagulation. Coll. Surf. Physico. Eng. Asp. 211 (2e3),233e248.

Hou, D., Wang, J., Sun, X., Luan, Z., Zhao, C., Ren, X., 2010. Boronremoval from aqueous solution by direct contact membranedistillation. J. Hazard Mater 177 (1e3), 613e619.

Kabay, N., Yilmaz, I., Yamac, S., Samatya, S., Yuksel, M.,Yuksel, U., Arda, M., Saglam, M., Iwanaga, T., Hirowatari, K.,2004. Removal and recovery of boron from geothermalwastewater by selective ion exchange resins. I. Laboratory

tests. React. Funct. Poly. 60 (0), 163e170.

Fig. 8 e XRD pattern after hydrothermal mineralization at

120 C, 2 h, 0.3 g Ca (OH)2, 2 g electrochemical coagulation

flocs, pH 10.

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3122

http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref17http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref16http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref15http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref14http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref13http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref12http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref11http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref10http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref9http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref8http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref7http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref6http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref5http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref4http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref3http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref2http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1http://refhub.elsevier.com/S0043-1354(13)01023-3/sref1


11/11

Li, Q., Yue, Q.-Y., Su, Y., Gao, B.-Y., Sun, H.-J., 2010. Equilibrium,thermodynamics and process design to minimize adsorbentamount for the adsorption of acid dyes onto cationic polymer-loaded bentonite. Chem. Eng. J. 158 (3), 489e497.

Linder, R.E., Strader, L.F., Rehnberg, G.L., 1990. Effect of acuteexposure to boric acid on the male reproductive system of therat. J. Toxicol. Environ. Health 31 (2), 133e146.

Ma, J., Jia, Y., Jing, Y., Yao, Y., Sun, J., 2011. Kinetics and

thermodynamics of methylene blue adsorption by cobalt-hectorite composite. Dyes Pigments 93 (1e3), 1441e1446.

Malakootian, M., Yousefi, N., 2009. The efficiency of electrocoagulation process using aluminum electrodes inremoval of hardness from water. Iran. J. Environ. Health Sci.Eng. 6 (2), 131e136.

Masion, A., Thomas, F., Tchoubar, D., Bottero, J.Y., Tekely, P.,1994. Chemistry and structure of Al(OH)/Organic precipitates.A small-Angle x-ray scattering study. 3. Depolymerization of the Al13 polycation by organic ligands. Langmuir 10 (11),4353e4356.

Malaysian Investment Development Authority (MIDA), 2013.Scheduled Waste Treatment Rates. Available: http://www.mida.gov.my/env3/index.php?page¼scheduled-waste-treatment-rates.

Mohajeri, L., Aziz, H.A., Isa, M.H., Zahed, M.A., 2010. A statisticalexperiment design approach for optimizing biodegradation of weathered crude oil in coastal sediments. Biores. Technol. 101(3), 893e900.

Morris, W., Weber, W., 1963. Kinetics of adsorption on carbonfrom solution. J. Sanit. Eng. Div. 89 (2), 31e60.

Olmez-Hanci, T., Kartal, Z., Arslan-Alaton, I. Electrocoagulation of commercial naphthalene sulfonates: process optimizationand assessment of implementation potential. J. Environ.Manag. 99(0), 44e51.

Pouet, M.F., Grasmick, A., 1995. Urban wastewater treatment byelectrocoagulation and flotation. Water Sci. Technol. 31 (3e4),275e283.

Seki, Y., Seyhan, S., Yurdakoc, M., 2006. Removal of boron fromaqueous solution by adsorption on Al2O3 based materials

using full factorial design. J. Hazard Mater 138 (1), 60e66.Shen, X.E., Shan, X.Q., Dong, D.M., Hua, X.Y., Owens, G., 2009.

Kinetics and thermodynamics of sorption of nitroaromatic

compounds to as-grown and oxidized multiwalled carbonnanotubes. J. Colloid Interface Sci. 330 (1), 1e8.

Sundaram, C.S., Viswanathan, N., Meenakshi, S., 2008.Defluoridation chemistry of synthetic hydroxyapatite at nanoscale: equilibrium and kinetic studies. J. Hazard Mater 155(1e2), 206e215.

Tenaga National Malaysia, 2013. Pricing and Tariff for Electricity.Turek, M., Dydo, P., Trojanowska, J., Bandura, B., 2007.

Electrodialytic treatment of boron-containing wastewater.Desalination 205 (1e3), 185e191.

Vik, E.A., Carlson, D.A., Eikum, A.S., Gjessing, E.T., 1984.Electrocoagulation of potable water. Water Res. 18 (11),1355e1360.

World Health Organization (WHO), 2011. Guidelines for Drinking-water Quality. Available: http://whqlibdoc.who.int/publications/2011/9789241548151_eng.pdf .

Yildiz, Y.I., Koparal, A.S., Keskinler, B., 2008. Effect of initial pHand supporting electrolyte on the treatment of watercontaining high concentration of humic substances byelectrocoagulation. Chem. Eng. J. 138 (1e3), 63e72.

Yilmaz, A.E., Boncukcuog lu, R., KocakerUm, M.M.,Kocadag istan, E., 2008a. An empirical model for kinetics of boron removal from boron containing wastewaters by the

electrocoagulation method in a batch reactor. Desalination230 (1e3), 288e297.

Yilmaz, A.E., Boncukcuoglu, R., Kocakerim, M.M., Yilmaz, M.T.,Paluluoglu, C., 2008b. Boron removal from geothermal watersby electrocoagulation. J. Hazard Mater 153 (1e2), 146e151.

Zhang, H., Ran, X., Wu, X., Zhang, D., 2011. Evaluation of electro-oxidation of biologically treated landfill leachate using response surface methodology. J. Hazard Mater 188 (1e3),261e268.

Zhu, B., Clifford, D.A., Chellam, S., 2005. Comparison of electrocoagulation and chemical coagulation pretreatment forenhanced virus removal using microfiltration membranes.Water Res. 39 (13), 3098e3108.

Zinatizadeh, A.A.L., Mohamed, A.R., Mashitah, M.D.,Abdullah, A.Z., Isa, M.H., 2007. Optimization of pre-treated

palm oil mill effluent digestion in an up-flow anaerobic sludgefixed film bioreactor: a comparative study. Biochem. Eng. J. 35(2), 226e237.

w a t e r r e s e a r c h 5 1 ( 2 0 1 4 ) 1 1 3 e1 2 3 123

All i t t f d li d i bl li k d t bli ti R hG t l tti d d th i di t l
http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref30http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://whqlibdoc.who.int/publications/2011/9789241548151_eng.pdfhttp://whqlibdoc.who.int/publications/2011/9789241548151_eng.pdfhttp://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://dx.doi.org/10.1016/j.watres.2013.12.024http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref39http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref38http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref37http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref36http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref35http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://refhub.elsevier.com/S0043-1354(13)01023-3/sref34http://whqlibdoc.who.int/publications/2011/9789241548151_eng.pdfhttp://whqlibdoc.who.int/publications/2011/9789241548151_eng.pdfhttp://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref32http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref31http://refhub.elsevier.com/S0043-1354(13)01023-3/sref30http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref29http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref28http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref27http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref26http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref25http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://refhub.elsevier.com/S0043-1354(13)01023-3/sref24http://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://www.mida.gov.my/env3/index.php?page=scheduled-waste-treatment-rateshttp://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref22http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref21http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref20http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref19http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18http://refhub.elsevier.com/S0043-1354(13)01023-3/sref18

boron removal and recovery

Documents