Removal of Lead(II) by Adsorption Using Treated Granular Activated Carbon: Batch and Column Studies

Abstract

In the present study, a deeper understanding of adsorption behavior of Pb(II) from aqueous systems onto activated carbon and treated activated carbon has been attempted via static and column mode studies under various conditions. It probes mainly two adsorbents that is, activated carbon (AC) and modified activated carbon (AC-S). Characterization of both the adsorbents was one of the key focal areas of the present study. This has shown a clear change or demarcation in the various physical and chemical properties of the modified adsorbent from its precursor activated carbon. Both the adsorbents are subjected to static mode adsorption studies and then after a comparison based on isotherm analysis; more efficient adsorbent is screened for column mode adsorption studies. The lead removal increased for sample of treated carbon. The extent of Pb(II) removal was found to be higher in the treated activated carbon. The aim of carrying out the continuous-flow studies was to assess the effect of various process variables, viz., of bed height, hydraulic loading rate and initial feed concentration on breakthrough time and adsorption capacity. This has helped in ascertaining the practical applicability of the adsorbent. Breakthrough curves were plotted for the adsorption of lead on the adsorbent using continuous-flow column operation by varying different operating parameters like hydraulic loading rate (3.0-10.5 m3/(hm2)), bed height (0.3-0.5 m) and feed concentrations (2.0-6.0 mg/l). At the end, an attempt has also been made to model the data generated from column studies using the empirical relationship based on Bohart-Adams model. This model has provided an objective framework to the subjective interpretation of the adsorption system and the model constant obtained here can be used to achieve the ultimate objective of our study that is, up scaling and designing of adsorption process at the pilot plant scale level. AC-S column regeneration using 0.5 and 1.0M concentration of HNO3 has been investigated. It has shown a regeneration efficiency of 52.0% with 0.5 M HNO3.

Full text

Journal of Hazardous Materials B125 (2005) 211–220
Removal of lead(II) by adsorption using treated granular activated
carbon: Batch and column studies
Jyotsna Goela,b, Krishna Kadirvelua,∗, Chitra Rajagopala, Vinod Kumar Gargb
aCentre for Fire, Explosives and Environment Safety (CFEES), Defence R&D Organisation (DRDO),
Brig. S.K. Majumdar Marg, Timarpur, Delhi 110054, India
bDepartment of Environmental Science and Engineering, Guru Jambheshwar University, Hisar, Haryana 125001, India
Received 11 January 2005; received in revised form 19 May 2005; accepted 24 May 2005
Available online 12 July 2005
Abstract
In the present study, a deeper understanding of adsorption behavior of Pb(II) from aqueous systems onto activated carbon and treated
activated carbon has been attempted via static and column mode studies under various conditions. It probes mainly two adsorbents that is,
activated carbon (AC) and modified activated carbon (AC–S). Characterization of both the adsorbents was one of the key focal areas of the
present study. This has shown a clear change or demarcation in the various physical and chemical properties of the modified adsorbent from its
precursor activated carbon. Both the adsorbents are subjected to static mode adsorption studies and then after a comparison based on isotherm
analysis; more efficient adsorbent is screened for column mode adsorption studies. The lead removal increased for sample of treated carbon.
The extent of Pb(II) removal was found to be higher in the treated activated carbon. The aim of carrying out the continuous-flow studies was
to assess the effect of various process variables, viz., of bed height, hydraulic loading rate and initial feed concentration on breakthrough time
and adsorption capacity. This has helped in ascertaining the practical applicability of the adsorbent. Breakthrough curves were plotted for the
adsorption of lead on the adsorbent using continuous-flow column operation by varying different operating parameters like hydraulic loading
rate (3.0–10.5 m3/(hm2)), bed height (0.3–0.5m) and feed concentrations (2.0–6.0 mg/l). At the end, an attempt has also been made to model
the data generated from column studies using the empirical relationship based on Bohart–Adams model. This model has provided an objective
framework to the subjective interpretation of the adsorption system and the model constant obtained here can be used to achieve the ultimate
objective of our study that is, up scaling and designing of adsorption process at the pilot plant scale level. AC–S column regeneration using
0.5 and 1.0 M concentration of HNO3has been investigated. It has shown a regeneration efficiency of 52.0% with 0.5 M HNO3.
© 2005 Published by Elsevier B.V.
Keywords: Lead; Adsorption; Treated activated carbon (AC–S); Static; Column; Bohart–Adams Model; Regeneration
1. Introduction
The importance of heavy metal pollution control has
increased significantly in last decades. Environmentalists are
primarily concerned with the presence of heavy metals due
their high toxicity and impact on human health and environ-
ment.
Therefore,therehavebeen tremendous effortsonreducing
the concentration of heavy metals in the effluent wastewaters
∗Corresponding author. Tel.: +91 11 23907278; fax: +91 11 23819547.
E-mail address: kadirvelu@lycos.com (K. Kadirvelu).
in view of ethical obligations as well as to meet stringent
permissible discharge levels, as set by the various pollution
control and regulatory authorities in various countries. One
of the toxic metals among the potentially toxic heavy metal
list is lead [1].
Lead poisoning in human causes severe damage to the
kidney, nervous system, reproductive system, liver and brain.
Severe exposure to lead has been associated with sterility,
abortion, stillbirths and neo-natal deaths [2,3].
Process industries, such as acid battery manufacturing,
metal plating and finishing, ammunition, tetraethyl lead
manufacturing, ceramic and glass industries and environ-
mental cleanup services treat and disposal of lead con-
0304-3894/$ – see front matter © 2005 Published by Elsevier B.V.
doi:10.1016/j.jhazmat.2005.05.032

212 J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220
taminated water are the major sources of lead pollution
[4].
The permissible level for lead in drinking water is
0.05 mg/l. The permissible limit (mg/l) for Pb(II) in wastew-
ater, given by Environmental Protection Agency (EPA), is
0.05mg/l and that of Bureau of Indian Standards (BIS) is
0.1mg/l [5].
Forlead in wastewater,current abatement and remediation
proceduresincludepHadjustment with lime or alkali hydrox-
ides, coagulation–sedimentation, reverse osmosis and ion
exchange.The current study is based on exploring its removal
methodbasedon adsorption technique. Adsorption compared
with other method appears to be an attractive process in view
of its efficiency and ease with which it can be applied in the
treatment of wastewater containing heavy metals [6].
In the present batch mode adsorption study, two types of
adsorbent are explored one is commercial activated carbon
as such and other is alkali sulphide treated activated carbon.
More efficient adsorbent out of the two is opted for column
studies after assessing the adsorption potential of two adsor-
bents with respect to the lead removal.
The idea following enriching the sulphur content of the
commercial activated carbon is that heavy metals have more
affinity for sulphides then other anions as can also be seen
from the natural occurrence of most of the metals in their
sulphide form [7]. The chemical affinity of the lead towards
sulphur groups is higher. This is also consistent with the fact,
as reported by Pearson [8] and Gomez-Serrano et al. [9], that
molecules where donor atom is F, O or N are all very hard,
whereasforthesimilarmoleculeswherethedonor atom is Cl,
S or P there is always a large drop in absolute hardness. Since,
lead chemical species show a high affinity toward sulphur
(also reported by Ochiai, 1985) [10], a tentative method of
enhancing the adsorption capacity of activated carbon could
bebased on theintroductionof surface sulphurin the material.
Moreover, the chemical treatment was essentially required to
introduce suitable functional groups on its surface for the
improvement in the adsorption affinity and efficiency of the
AC.
Several studies are available which are in covenant with
the above statement [9,11,12]. Krishnan and Annirudhan has
modified activated carbon prepared from bagasse pith using
SO2and H2S gases [12].
The majority of adsorption investigations were conducted
in the batch mode. The search for an effective and econom-
ical removal of Pb(II) from wastewater resulted in the use
of various low-cost adsorbents like baggase pith sulphurised
activated carbon, blast furnace sludge, biogas residual slurry,
olive mill products and peanut hull carbon [13–17]. Although
batch laboratory adsorption studies provide useful informa-
tionontheapplication of adsorption to the removalofspecific
waste constituents, continuous column studies provide the
most practical application of this process in wastewater treat-
ment.Thereasonforthisisthatthe high adsorption capacities
in equilibrium with the influent concentration rather than the
effluent concentration can be achieved [18]. In static mode
adsorption studies, the same solution remains in contact with
a given quantity of the adsorbent. The adsorption process
continues, however, till equilibrium between the solute con-
centration in solution, and the solute adsorbed per unit weight
of the adsorbent is reached. This equilibrium established is
static in nature, as it does not change further with time. In
dynamiccolumn adsorption, solutioncontinuouslyenters and
leaves the column, so that the complete equilibrium is never
established at any stage between the solute in solution and the
amount adsorbed. Equilibrium has to be continuously estab-
lished, as each time, it meets the fresh concentrations, hence,
equilibrium in column mode is termed as dynamic equilib-
rium.
Additional information on the efficiency of the treated
adsorbent in the column mode has been gathered in order
to ascertain the practical applicability of the adsorbent for
real industrial wastewaters.
2. Experimental methods
2.1. Adsorbent
Coconutshell based granulated activatedcarbon (AC)pro-
duced from Active Carbon Ltd.; Hyderabad, India was used
as precursor carbon in the study. The carbon was washed with
distilled water to remove fines and impurities, oven dried at
110 ◦C for 6 h and stored in plastic containers for further use.
This adsorbent is termed as AC.
To enrich the sulphur percentage, 99.5 g of AC is
immersed for 24h in minimum quantity of distilled water
containing 0.5g of Na2S. The mixture was heated almost
to dryness and then dried in an oven for 4h at 110 ◦C. The
dried sample was washed with distilled water several times
till it gave nil concentration of sulphide. The washed sam-
ple was again dried at 110 ◦C for 4h cooled in the dessicator
and stored in airtight plastic container for further use. The
physico-chemical properties of both the adsorbents are sum-
marized in the Table 1. The adsorbents were characterized
with the aim of assessing its various physical and chemical
properties, so that a better interpretation of the mechanism
involved during the adsorption process can be provided. Car-
bon was analyzed for various parameters like pH, conduc-
tivity, moisture content, volatile matter and ash content by
using standard methods [19,20]. Determination of C, H, N, S
and O (by difference) was performed with a GmBH Elemen-
talAnalyzer System of from ELEMENTAR Analysesysteme,
Germany. Thesurfacearea (BET) measurements oftheadsor-
bents were obtained using BET method with nitrogen gas at
77 K using Micrometrics surface area analyser (Model ASP-
2010, USA). The surface functional groups on the adsorbent
were measured using Boehm Titration Method [21]. The sur-
face morphological changes in the AC after treatment were
visualized by using a scanning electron microscopy (SEM)
with JSM-840 JEOL microscope of JEOL Techniques LTD,
Japan at 2500×magnification.

J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220 213
Table 1
Main characteristics of the adsorbent
Parameters Value
AC AC–S
Raw material Coconut shell Coconut shell
Surface area (m2g−1) (based on BET) 1000 900
Bulk density (g ml−1) 0.5 0.5
Particle density (gml−1) 0.85 0.98
Ash content (on dry basis) (%) 3.59 2.00
Moisture content (%) 5.0 7.0
pHZPC 5.3 4.5
Conductivity (␮Scm−1) 94.0 110.0
Particle size 8–20 mesh 8–20 mesh
Elemental analysis (%)
C 66.0 63.0
H 3.5 1.75
N 0.331 0.281
O 29.6 27.109
S 0.56 7.86
Acidic surface functional groups (meq g−1)
Carboxylic 0.105 0.119
Phenolic 0.006 0.026
Lactonic 0.003 0.03
Carbonyl Nil Nil
2.2. Adsorbate
Stock solution of Pb(II) was prepared (1000mg/l) by dis-
solving required amount of, Pb(NO3)2in acidified double
distilled water. The stock solution was diluted with dis-
tilled water to obtain desired concentration ranging from 5
to 70 mg/l. All the chemicals used were of analytical reagent
grade and obtained from SD fine chemicals, Mumbai, unless
and otherwise mentioned.
2.3. Analytical method
Lead concentration was analyzed using atomic absorption
spectrophotometer (AAS; model GBC 935) at 217nm.
2.4. Batch mode adsorption studies
Stock solution of the adsorbate was diluted as required to
obtain standard solutions containing 5–70mg/l of Pb(II). A
100 mg of adsorbent was added to 50ml of Pb(II) solution of
a desired concentration at pH 5.0 in 125ml reagent bottles
and were agitated at 120rpm for 20 h at room temperature
(37±2◦C) in a mechanical shaker. At the end of agitation,
suspensions were separated by filteration and analyzed for
their Pb(II) content. From Pb(II) concentration measured
before and after adsorption (Ciand Ce, respectively) dry
weight of the adsorbent (W) and the volume of aqueous solu-
tion (Vin l), the amount of equilibrium adsorption of metals
(qe) was calculated using the Eq. (1a):
qe(mg/g) =(Ci−Ce)V
W(1a)
The removal percentage (R%) is defined as the ratio of
difference in metal concentration before and after adsorption
(Ci−Ce) to the initial concentration of Pb(II) in the aqueous
solution (Ci) was calculated using the Eq. (1b):
R(%) =Ci−Ce
Ci×100 (1b)
The batch adsorption study was replicated thrice for each of
the adsorbents (AC and AC–S).
2.5. Dynamic column study experimental details
Experimental arrangement used for the dynamic column
studies was same as used by Rajagopal and Kapoor [22] in
their study for adsorptive removal of explosive contaminated
wastewater using granulated activated carbon. The experi-
mental arrangement is shown in Fig. 1 and it consists of
column of borosilicate glass of 20mm diameter. Sampling
points before and after the column were also provided for
drawing the effluent samples at regular intervals. Effluent
after passing though the columns was discharged into a sump
below the column. Previously wetted and degassed treated
activated carbon is packed up to desire bed height (m) in
water filled column of 20mm (internal diameter) and was
kept submerged throughout the runs to avoid air entrapment
in the bed. Columns are mounted vertically and the AC–S bed
is supported on perforated plate. The operation is in down
flow plug mode.
A control valve to regulate the flow and a rotameter to
monitor hydraulic loading rate is incorporated in the feed line
of the column. Effluent and influent samples are collected
after a regular interval. All the sorption experiments were
carried out atthe room temperature of37 ±2◦C and initialpH
of 5.0. The residual concentration of lead in aqueous sample
was determined using atomic adsorption spectrophotometer
as before.
3. Result and discussion
3.1. Physical and chemical characterization of the
adsorbents
3.1.1. Scanning electron microscopy
The SEM enables the direct observation of the changes
in the surface microstructures of the carbons due to the
modifications. Studies are available which have reported the
utilization of the scanning electron microscopy analysis for
showing the surface modification changes in the developed
adsorbent [23]. As can be observed from Fig. 2(a) [AC] and
Fig. 2(b) [AC–S], there is clear demarcation in the surface
morphology of AC after treatment.
3.1.2. Fourier transform infrared spectroscopy
Fourier transform infrared spectroscopy (FTIR) was used
todetermine the vibrationfrequencychanges inthefunctional

214 J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220
Fig. 1. Experimental setup for dynamic column studies.
Table 2
Some fundamental FTIR frequencies of activated carbon (AC) and the sul-
phurised activated carbon (AC–S)
Band position (cm−1) Possible assignments
AC AC–S
3550 3450 O H stretching (intermolecular dimeric)
1725 1749 C O stretching (aldehydic)
3040 3010 C H stretching (alkenes)
– 1190 C S stretching
– 1380 S O stretching (sulphonates)
– 460 S S stretching
groups in the carbons. The spectra of carbons were measured
by an FTIR spectrometer (Buckner, Germany) within the
range of 400–4000cm−1wave number. Spectra were plot-
ted for both the adsorbents that is, AC and AC–S using the
same scale on the transmittance axis. Table 2 presents the
fundamental frequencies of AC and AC–S, and their respec-
tive possible band frequencies in the FTIR spectrum. As can
be inferred from FTIR analysis, there is presence of sulphur
functional groups like C S, S O and S S on the AC S
surface.
3.1.3. Elemental analysis
Elemental analysis provided the complete elemental com-
position of both the adsorbents as shown in Table 1. Presence
of the sulphur functional groups, as interpreted from the
FTIR spectrum of AC–S is again substantiated by increases
in sulphur composition to 7.86% in AC–S from 0.56% in
AC.
3.2. Pb(II) equilibrium adsorption isotherms
Adsorption isotherm data for Pb(II) adsorption were plot-
ted and presented in Fig. 3. Equilibrium data obtained for
the two adsorbents were fitted to the Langmuir and Fre-
undlich isotherms. The following expressions of a straight
line were used, found by means of mathematical transforma-
tion of isotherms:
For Langmuir isotherm:
Ce
qe=1
Q0b+1
Q0Ce(2)
where Ceis the equilibrium metal ion concentration (mg/l),
qethe amount of lead adsorbed at equilibrium (mg/g) and
Q0(mg/g) and b(l/mg) are Langmuir constants related to
adsorption capacity and energy of adsorption, respectively
[24].
For Freundlich isotherm:
ln qe=ln Kf+1
nln Ce(3)
where Ceis the equilibrium metal ion concentration (mg/l), x
theamountofleadremoved (mg), mweight of adsorbent used
(g) and Kfand nare the Freundlich constants incorporating
all the factors effecting adsorption capacity, an indication of
favouraility of metal adsorption onto adsorbent [25].
Linear plots of Ce/qeversus Ceand ln(qe) versus lnCe
show that the adsorption follows Langmuir and Freundlich
isotherm models. As presented in the Table 3, a high value
of coefficient of correlation, R2for both the adsorbents indi-

J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220 215
Table 3
Adsorption constants for Langmuir and Freundlich isotherm models
Adsorbents Langmuir constants Freundlich constants
Q0(mg/g) b(l/mg) R2Kf(mg1−1/n l1/n/g) 1/nR
2
AC 21.88 3.51 0.9971 11.4937 0.409 0.858
AC–S 29.44 0.67 0.9313 12.44166 0.373 0.956
cates good agreement between experimental and predicted
data using Langmuir equation.
Triplicate runs for batch mode adsorption experiments
were made for each adsorbent to determine the relative devi-
ation of the experiments. The adsorption of metal ions on the
adsorbentmaterialremainedalmostconstantwiththerelative
deviation of the order of ±2%.
Maximum adsorption capacity [Q0(mg/g)] was found to
be 29.44mg of Pb(II) per g of AC–S as adsorbent. Values
of adsorption capacity Q0(mg/g) of the other adsorbents are
given in Table 4 for comparison.
Pb(II) adsorption onto AC–S shows about 35% increase
over the adsorption onto AC, as can be seen by comparing
Q0values for both the adsorbents. Hence, justify the validity
of the modification process for activated carbon.
Fig. 2. (a) Scanning electron micrograph (SEM) of AC. (b) Scanning elec-
tron micrograph (SEM) of AC–S.
Table 4
Comparison of adsorption capacity of other adsorbents for Pb(II)
Adsorbent Adsorption capacity (mg/g) Reference
Olive mill product 21.56 [16]
GAC saturated with bacteria 26.40 [26]
Eicchornia carbon 16.61 [27]
Carbon aerogel 34.72 [28]
Carbon nano tubes 12.41 [29]
Sphagnum moss peat 19.90 [30]
Red mud 64.79 [31]
Bentonite 15.38 [32]
Bituminous coal 8.89 [33]
AC–S 29.44 This work
3.3. Adsorption dynamic column studies
3.3.1. Column adsorption capacity
As the adsorbate solution passes through column, the
adsorption zone (where the bulk of adsorption takes place)
startsmovingout of the columnandthe effluent concentration
start rising with time. This is termed as break point. The time
taken for the effluent concentration to reach a specific break-
through concentration of interest is called the break though
time. The breakthrough time (tb) for each of the columns
operation was defined as the time when the effluent concen-
tration (Ce) of Pb(II) reached 50% of the feed concentration
(Cf). Breakthrough curve were plotted-giving ratio of efflu-
ent and feed (influent) concentrations (Ce/Cf) and time (h)
for varying operating conditions.
The approach of Treybal [34] has been adopted for calcu-
lating the column capacity for the removal of Pb(II). Break-
through capacity Q0.5 (at 50% or Ce/Cf= 0.5) expressed in
Fig. 3. Break through curve for different hydraulic loading rate at constant
bed height of 0.4 m and feed concentration of 6mg/l.

216 J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220
Table 5
Column adsorption capacity, Q0.5 at various operating conditions at 50%
break through concentrations
Concentration
(mg/l) Breakthrough
time, 50% (h) Hydraulic
loading rate
(m3/(h m2))
Bed height
(m) Adsorption
column capacity
(mg/g)
2207.5 0.5 1.75
3177.5 0.5 2.23
4157.5 0.5 2.63
6213.0 0.5 2.21
6164.5 0.5 2.52
6117.5 0.5 2.89
6710.5 0.5 2.57
667.5 0.4 1.91
612.57.5 0.6 2.79
mg of lead(II) adsorbed per gram of adsorbent was calculated
using Eq. (4):
Breakthroughcapacity,Q
0.5
=metaladsorbedon adsorbent bed(mg)
massof adsorbent in bed(g)
=
breakthrough time(at 50%) ×flowrate
×feedconcentration
massof adsorbent in bed (4)
Table 5 presents the column adsorption capacity for lead
onto the adsorbent for varying operating variables that is bed
height,flowrate and feed concentration. Thecolumncapacity
for Pb(II) adsorption for the bed height of 0.4 m, hydraulic
loading rate of 7.5m3/(h m2) and the feed concentration of
6mg/l for 50% breakthrough concentration were found to
be 2.89 mg/g. From comparison of adsorption capacity from
Langmuir isotherm and column experiments, we can see that
the less-stirred property in column mode reduced the lead(II)
adsorption capacity on AC–S.
3.3.2. Effect of hydraulic loading rate
The experiments were conducted at bed height of 0.4 m, at
constant feed concentration of 6 mg/l, with hydraulic loading
rate ranging from 3.0 to 10.5 m3/(hm2). Results of the exper-
iments on effect of hydraulic loading rate (HLR) are plotted
in Fig. 4, which shows that breakthrough time decreases from
21 to 7h, as HLR increases from 3.0 to 10.5 m3/(h m2). The
adsorption capacity, calculated using Eq. (4), accordingly
shows a maximum for 7.5m3/(h m2) as shown in Table 5.
The variation in the slope of the breakthrough curve and
adsorption capacity may be explained on the basis of mass
transfer fundamentals. Increase in the hydraulic loading rate
causes increase in zone speed, resulting in decrease in the
time required to achieve breakthrough [35].
3.3.3. Effect of bed height (adsorbent mass)
Break through experiments were conducted at bed heights
of 0.3, 0.4 and 0.5 m at constant feed concentration of 6mg/l
and at the hydraulic loading rate of 7.5 m3/(h m2) optimized
Fig. 4. Break through curve for different feed concentration at constant bed
height of 0.4 m and hydraulic loading rate of 7.5m3/(h m2).
above(as showingthe maximum adsorptioncapacity). Exper-
iments on effect of bed height showed a decrease in minimum
effluent concentration with bed height keeping other param-
eter constant. Minimum effluent concentration is defined as
the average concentration of the metal ion at the column out-
let (or effluent) in initial constant phase. As the bed height
increases, the length of the bed through which the effluent
passes increases. The increase in the total adsorptive capacity
of the bed results in a decrease in the solute concentration in
theeffluent.Beyond a bed height of 0.4 m, the curve shows no
appreciable change in minimum effluent concentration with
further increase in bed height, hence 0.4m has been chosen
as optimized bed height for rest of the experiments [36].
3.3.4. Effect of feed concentration
The change in the initial metal ion concentration has a sig-
nificant effect on breakthrough curve as illustrated in Fig. 5.
The larger the initial feed concentration, the steeper is the
slope of break through curve and smaller is the breakthrough
time. These results demonstrate that the change of concen-
tration gradient affects the saturation rate and breakthrough
time, or in other words, the diffusion process is concentration
Fig. 5. Isoremoval lines for 20, 30 and 60% breakthrough for different bed
height at constant feed concentration of 6 mg/l and hydraulic loading rate of
7.5m3/(h m2): Bohart–Adams modeling at mini-column studies.

J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220 217
dependent.As the feedconcentrationincreases, metal loading
rate increases, but so does the driving force for mass transfer,
which in a decrease in the adsorption zone length. The net
effect is an appreciable increase in adsorption capacity [37]
as presented in Table 4.
3.4. Modeling of column study results: Bohart–Adams
model
Bohart–Adams model based on the surface reaction the-
ory [18] and it assumes that equilibrium is not instantaneous;
therefore, the rate of the sorption is proportional to the frac-
tion of sorption capacity still remains on the sorbent [38,39].
According to the Bohart–Adams model, the following equa-
tion to predicts the performance of continuous adsorption
columns:
t=N0X
C0v−ln C0
Cb−1×1
[C0K](5)
where, tis the time to break point (h); C0the influent concen-
tration (mg/l); Cbthe concentration at break through (mg/l);
N0the adsorptive capacity of the adsorbent (mg adsorbed per
litre of solution); Xthe bed depth of column (m); vthe linear
flow rate (m/h); Kis the rate constant (l/(mgh)).
A simplified form of the Bohart–Adams Model is:
t=aX +b(6)
where
a=N0
C0V(7)
and bin Eq. (6) is given by:
b=ln C0
Cb−1×1
[C0K](8)
From iso-removal lines, i.e. the plots of time versus bed
height (Fig. 5) regarding column operation under con-
stant experimental conditions (except for bed height), the
main parameters of the Bohart–Adams model can be cal-
culated. Iso-removal lines were plotted for linear flow rate
of 7.5m3/(h m2) for three different bed heights that is, 0.3,
0.4 and 0.5m with the influent concentration of 6.0 mg/l.
The breakthrough time at desired breakthrough concen-
trations exhibit linearity with bed depth. From the slope
(a) and intercept (b) of the respective lines the adsorption
capacity (N0) and the rate of constant of adsorption (K),
respectively. The calculated constants for the Bohart–Adams
model for the adsorption of Pb(II) on to the AC–S are pre-
sented in Table 6 and will be of use in the design of the
column.
From the respective linear equation, the necessary bed
heightfor a pre-selectedtimeperiod can bedirectlycalculated
until a defined breakthrough concentration [39].
The slope constant for a different flow rate can be directly
calculated by multiplying the original slope (a) by the ratio
Table 6
Thecalculated constantsof Bohart–Adams modelfor theadsorption ofPb(II)
onto AC–S
Iso-removal
percentage (%) a(h/m) b(h) N0(mg/l) K(l/(hmg)) R2
20 12.5 −0.7 187.5 0.346 0.9423
30 16.0 −0.5 240.0 0.282 0.9046
60 35.0 −3.0 525.0 −0.022 0.9868
between the original (V0) and the new flow rates (Vn).
Accordingly,
anew =aold V0
Vn(9)
Similarlythe equation can bedevelopedfor oneconcentra-
tioncanbemodified to be applying for another concentration:
anew =aold C0
Cn(10)
bnew =bold C0
Cnln(Cn−1)
ln(C0−1)(11)
where C0and Cnare the original and the new feed concen-
trations.
Based on Eqs. (9)–(11), breakthrough time was predicted
for a new feed concentration and flow rate using the calcu-
lated Bohart–Adams model constants and obtained results
are presented in Tables 7a and 7b, respectively. Good corre-
lation prediction has been found for the case of changed feed
concentration and flow rate. This is proved by the low value
of standard deviation. Thus, developed model and the con-
stants evaluated can be employed for the design of adsorption
columns over a range of feasible flow rates and concentra-
tions.
3.5. Column regeneration studies
Regeneration of the adsorbent material is of crucial impor-
tance in the economic development. The aim is to remove the
loaded metal from the column in the smallest possible vol-
ume of an eluting solution. Regeneration must produce small
volumeof metalconcentratessuitable for metal-recoverypro-
cess, without damaging the capacity of the adsorbent, mak-
ing it reusable in several adsorptions and desorption cycles.
Regeneration should also ensure that eluted solution is not
posing any disposal problem waste in terms of high acidity.
In the present study, elution of the lead (adsorbate) from
aqueous solution was done using two concentrations of
HNO3that is, 1.0 and 0.5 M under identical conditions of
hydraulic loading rate of 4.5 m3/(hm2), adsorbent bed height
of 0.5m as used for preloading of adsorbent with 6 mg/l of
lead(II) as feed concentration. The results obtained with the
two concentrations of HNO3are shown in Fig. 6.
From comparison of the elution histograms, the use of
1.0M HNO3appeared to be more effective than 0.5M
HNO3. It can be inferred from Fig. 6 that the total eluant

218 J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220
Table 7a
Predicted breakthrough time based on the Bohart–Adams constants for a new feed concentration
Break point (%) aold bold C0CnC0/Cnanew bnew Bed height (m) Predicted time (h) Observed time (h) Ea
60 35 −3 6 4 1.5 52.5 −3.07 0.4 17.92 18 0.03
30 16 −0.5 6 4 1.5 24.0 −0.51 0.4 9.08 10
20 12.5 −0.7 6 4 1.5 18.8 −0.68 0.4 6.81 6
Table 7b
Predicted breakthrough time based on the Bohart–Adams constants for a new flow rate
Break point (%) aold bold V0VnV0/Vnanew Bed height (m) Predicted time (h) Observed time (h) Ea
60 35 −3 7.5 4.5 1.67 58.3 0.4 20.33 17.5 0.04
30 16 −0.5 7.5 4.5 1.67 26.7 0.4 10.16 11.5
20 12.5 −0.6 7.5 4.5 1.67 20.8 0.4 7.66 8.00
aStandard deviation (E): E=
N

j=1(qe)experimental−(qe)predicted
(qe)experimental 
2
.
volume amounted to be 60ml for 1.0 M HNO3and 80 ml
for 0.5 M HNO3, respectively, with corresponding total
amounts of lead(II) desorbed being 1.66 and 1.37mg/g of
lead(II), respectively. The reason for this behaviour could be
that 1.0 M HNO3provided more exchangeable H+ion with
metal ion than in case of 0.5M HNO3.
However, in terms of the quantity of eluant used (as dif-
ference between eluant’s volumes used in two cases are not
much significant) in the actual elution process as well as seri-
ous problems caused by disposal of waste with a higher acid
content, it could be that 0.5M HNO3may be more suitable
as eluant.
After the metal ion was recovered, column regenerated
with 0.5 M HNO3, was washed with distilled water and again
loaded with lead(II) concentration of 6mg/l under the iden-
tical conditions as used for regeneration. Regeneration effi-
ciency was calculated using Eq. (12) was found to be 52.0%
for lead(II) [40]:
RE% =Ar
A0×100 (12)
Fig. 6. Regeneration of AC–S adsorbent column with 1.0 and 0.5M HNO3
for desorbing lead(II).
In the above equation, Aris the adsorptive capacity of the
regenerated column and A0is the original capacity of the
virgin adsorbent.
3.6. Approach to the adsorption mechanism: pH profile
at pilot scale level
The most important factor influencing the adsorption rate
andcapacity is theadsorptionpH. All thecolumn experiments
were carried out at initial pH 5.3 and the pH profile of exit
solution for the sorption of Pb(II) onto AC–S is examined at
the pilot plant level. At pilot plant operating conditions were
maintained at the hydraulic loading rate of 4.93 m3/(hm2),
bed height of 80 cm, column diameter of 10cm, initial pH of
5.3 and initial feed concentration of 6mg/l.
The pH profile of exit solution for the sorption of Pb(II)
onto AC–S in the pilot scale column is shown in the Fig. 7.
From pH profiles it has been demonstrated that at the begin-
ning of the experiment the exit pH initially increased from
6.4 to 7.34 for the first 10h, indicating the release of alkali
Fig.7. Break through curve at pilotstudies at 4.93 m3/(h m2)and the pH pro-
file at the exit of the column at pilot studies, Conditions: bed height =80 cm,
column diameter =10 cm, initial pH 5.3 and feed concentration of 6 mg/l.

J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220 219
ions from activated carbon surface. The very low liquid-
phase Pb concentration during initial stage of run suggests
possibility of some Pb(OH)2precipitation on the adsorbent
surface in the bed [37]. Zhan and Zhao has also reported
increase in the final pH value due to the precipitation of
lead(II) in adsorbent synthesized from natural condensed tan-
nin [41]. Subsequently, a drop in pH profile of outlet solution
was observed, which indicates release of cationic species
like H+in the effluent during the adsorption of free Pb(II)
ions onto AC–S. This suggests that adsorption of Pb(II) by
AC–S is initially by surface precipitation followed by ion
exchange and surface complexation of Pb(II) at carbon sur-
face. Above mechanisms are very much in agreement with
Fig. 7.
4. Conclusions
The objective of this work was to study the dependence of
adsorption on adsorbent and adsorbate (lead) characteristics
by means of both batch and column studies. Modeling of
results of column mode studies can be used for further up-
scaling process. Conclusions from the present study are as
follows:
1. Characterization has shown a clear demarcation in the
physico-chemical properties of the two adsorbents.
2. Pb(II) adsorption onto treated activated carbon (AC–S)
shows about 35.0% increase over the adsorption onto AC,
as can be seen by comparing Q0values for both the adsor-
bents. Hence the more efficient adsorbent, AC–S is further
investigated by column mode studies.
3. Adsorption of Pb(II) on AC–S followed both Langmuir
and Freundlich adsorption isotherm models.
4. Adsorption capacity for 6mg/l feed concentration of
Pb(II) at hydraulic loading rate of 7.5 m3/(hm2) and 0.4 m
bed height is found to be 2.89mg/g, which indicates that
practically adsorption capacity of AC–S is far less than
batch mode results. This may be due to the potential
irreversibility of the sorption process and the different
approaches to adsorption equilibrium in different sys-
tems, i.e. the solution-phase concentration is continuously
decreasing in the adsorption isotherm and in the batch sys-
tems while that concentration is continuously increasing
in the column system.
5. Removal of Pb(II) onto adsorbent depends on adsorbent
concentration, and hydraulic loading rate, feed concentra-
tion and adsorbent bed height.
6. Column studies data has shown a good agreement
with the predicted results obtained by application of
Bohart–Adams model as can be concluded from low
value of standard deviation. The constants evaluated from
Bohart–Adams model can be employed for the designing
of adsorption columns over a range of feasible flow rates
and concentrations.
7. Theadsorbed lead(II) ion canbe effectivelyeluted with the
use of mineral acid such as HNO3of 0.5 M concentrations
with the regeneration efficiency of 52%.
8. Theinvestigationof the pH profile studies at thepilotscale
level elucidates the adsorption mechanism. The mecha-
nism includes ion exchange/adsorption, surface complex-
ation and surface precipitation.
Similar work is in progress for investigating the efficacy
of ACS for other heavy metals. Initial results imply that ACS
has good adsorption capacity for other heavy metals and
hence ACS can be used for wastewater applications in multi-
componentsystem due to itsuniversality andeasy preparation
method. On the basis of bench scale column studies and an
adsorption model developed to predict breakthrough curve,
a pilot plant for the treatment of Pb(II) effluent wastewater is
now in operations.
Acknowledgements
The authors would like to thank Director, Centre for Fire,
Explosives and Environment Safety (CFEES) for provid-
ing all facilities and encouragement to carry out this work.
Thanksare also due toSh. Satish KumarandSh. Ashok Rawat
ofCFEES,forcarryingoutanalysisofthe samples by Atomic
Absorption Spectrophotometer (AAS).
References
[1] M. Sitting, Handbook of Toxic and Hazardous Chemicals, Noyes
Publications, Park Ridge, NJ, USA, 1981.
[2] S. Manahan, Environmental Chemistry, Brooks/Colei, CA, USA,
1984.
[3] R.A. Goyer, I.J. Chisolon, Lead: In Metallic Contamination and
Human Health, Academic Press, New York/London, 1972.
[4] K.S. Subramanian, J.W. Cooner, J. Environ. Sci. Health: Part A 54
(1991) 29–33.
[5] BIS, Tolerance limits for industrial effluents prescribed by Bureau
of Indian Standards, IS 2490 (Part I), New Delhi, 1981.
[6] K.P. Yadav, B.S. Tyagi, V.N. Singh, J. Chem. Technol. Biotechnol.
51 (1991) 47–60.
[7] G. Bitton, Wastewater Microbiology, Wiley, Europe, 1999.
[8] R.G. Pearson, Absolute electro negativity and hardness: application
to, inorganic chemistry, Inorg. Chem. 27 (1988) 734–740.
[9] V. Gomez-Serrano, A. Macias-Garcia, A. Espinosa-Mansilla, C.
Valenzuella-Calahorro, Wat. Res. 32 (1998) 1–4.
[10] E. Ochiai, Quimica Bioinorganica, Editorial Reverte, Barcelona,
1985.
[11] C. Raji, T.S. Anirudhan, Indian J Chem. Techno. 3 (1996) 49–52.
[12] K.A. Krishnan, T.S. Anirudhan, Indian. J. Chem. Techno. 9 (2002)
32–40.
[13] K.A. Krishnan, T.S. Anirudhan, J. Hazard. Mater. B92 (2002)
161–183.
[14] D. Lopez, C. Perez, F.A. Lopez, Wat. Res. 32 (1998) 989–996.
[15] C. Namasivayam, R.T. Yamuna, Bioresour. Techno. 52 (1995)
125–131.
[16] S.H. Gharaibeh, W.Y. Abu-El-Shar, M.M. Al-kofahi, Wat. Res. 32
(1998) 498–502.

220 J. Goel et al. / Journal of Hazardous Materials B125 (2005) 211–220
[17] K. Periasamy, C. Namasivayam, Sep. Sci. Technol. 3 (1995)
2223–2237.
[18] W.W. Eckenfelder Jr., Industrial Water Pollution Control, McGraw
Hill publication, USA, 1989, pp. 273–283.
[19] ISI, Methods of Sampling and Tests for Activated Carbon Used for
Decolorizing Vegetable Oils and Sugar Solutions, Bureau of Indian
standards, New Delhi, India, 1977, pp. 877.
[20] K. Kadirvelu, Characterization of coir pith carbon and its utilization
in the treatment of metal bearing wastewater, Ph.D. Thesis, Bharthiar
University, Coimbatore, India, 1998.
[21] H.P. Boehm, High Temp. High Press. 22 (1990) 275–288.
[22] C. Rajagopal, J.C. Kapoor, J. Hazard. Mater. B 87 (2001) 73–98.
[23] W. Qiao, Y. Korai, I. Mochida, Y. Hori, T. Maeda, Carbon 39 (2001)
2355–2368.
[24] I. Langmuir, J. Amer. Chem. Soc. 40 (1918) 1361–1403.
[25] H. Freundlich, W.J. Helle, J. Amer. Chem. Soc. 61 (1939) 2–28.
[26] J. Rivera-Utrilla, I. Bautista-Toledo, M.A. Ferro-Garcia, C. Moreno-
Castilla, Carbon 41 (2003) 323–330.
[27] P. Shekinath, K. Kadirvelu, P. Kanmani, P. Senthilkumar, V. Subbu-
ram, J. Chem. Tech. Biotech. 77 (2002) 1–7.
[28] J. Goel, K. Kadirvelu, C. Rajagopal, V.K. Garg, Ind. Eng. Chem.
Res. 44 (2005) 1987–1994.
[29] L. Yan-Hui, D. Zechao, J. Ding, D. Wub, Z. Luan, Y. Zhua, Wat.
Res. 39 (2005) 605–609.
[30] J.K. Mclellan, C.A. Rock, Water Air Soil Pollut. 37 (1988) 203–
215.
[31] V.K. Gupta, M. Gupta, S. Sharma, Wat. Res. 35 (2001) 1125–1134.
[32] G. Bereket, A.Z. Aroguz, M.Z. Ozel, J. Colloid Interface Sci. 183
(1997) 338–343.
[33] D. Singh, N.S. Rawat, Ind. J. Chem. Tech. 4 (1995) 49–50.
[34] R.E. Treybal, Mass transfer operations, third ed., McGraw Hill, New
York, USA, 1980, pp. 447–522.
[35] R.A. Perry, C.H. Chilton, Chemical Engineering Handbook, fifth ed.,
McGraw Hill, New York, USA, 1973, pp. 16–23.
[36] U.S.EPA: Process Design Manual for Carbon Adsorption, Technol-
ogy Transfer, USA, 1973.
[37] D.C.K. Ko, J.F. Porter, G. McKay, Trans. I Chem. E, Part B 81
(2003) 78–86.
[38] T.R. Muraleedharan, L. Philip, L. Iyenger, Bioresour. Technol. 49
(1994) 179–186.
[39] M. Lehman, A.I. Zouboulis, K.A. Matis, Environ. Pollut. 113 (2001)
121–128.
[40] V.K. Gupta, S. Sharma, Environ. Sci. Technol. 36 (2002) 3612–3617.
[41] Z.-M. Zhan, X. Zhao, Wat. Res. 37 (2003) 3905–3912.