Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

Contents lists available at SciVerse ScienceDirect

Journal of Industrial and Engineering Chemistry

journal homepage: www.elsevier.com/locate/jiec

Preparation, modication and industrial application of activated

carbon from almond shell
Abdessalem Omri a,*, Mourad Benzina a, Najwa Ammar b
a
b

Laboratory of Water-Energy-Environment (LR3E), Code: AD-10-02, National School of Engineers of Sfax, University of Sfax, BP W, 3038 Sfax, Tunisia
Tunisian Chemical Group, Sfax Tunisia, BP 393/3018, Tunisia

A R T I C L E I N F O

Article history:
Received 10 November 2012
Accepted 25 March 2013
Available online 31 March 2013
Keywords:
Almond shell
Activated carbon
Characterization
Ammoxidation
Total organic carbon

A B S T R A C T

Almond shell was used to prepare activated carbon using physical activation method, consisted of
carbon dioxide (CO2) gasication. The effects of the preparation variables which were activation
temperature, activation time and carbon dioxide ow rate on the adsorption capacity of iodine and
methylene blue solution were investigated. The optimal activated carbon was obtained by these
conditions as follows: 800 8C activation temperature, 100 cm3/min carbon dioxide ow rate and 120 min
activation time. The characterization of carbon materials is performed by scanning electron microscopy
(SEM), X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FTIR), 13C (CP/MAS and MAS)
solid-state NMR, nitrogen adsorption (BET) and Boehms titration method. For an industrial application,
the optimal activated carbon was ammoxidated to improve its adsorption capacity toward total organic
carbon from Tunisian industrial phosphoric acid. The inuence of experimental parameters such as
specic consumption, initial concentration, contact time, agitation speed and temperature on TOC
removal was studied.
2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights
reserved.

1. Introduction
Phosphoric acid is the second most produced acid in the world,
after sulfuric acid. It was used as raw materials for the production
of detergents, food products, and alimentary supplies for cattle,
toothpastes and fertilizers. There are two methods of producing
phosphoric acid: the wet and the thermal process. In the wet
process, sulfuric acid is used to digest the calcium phosphate rock
producing a calcium sulfate slurry and phosphoric acid, which is
separated by ltration [1,2]. The thermal process consists of
burning phosphorus vapor and treating the resulting oxide with
steam [3]. The crude acid produced by the wet process is heavily
entrained with both mineral (arsenic, cadmium, uoride, sulfate,
copper, etc.) and organic impurities (di-n-bytul phthalate, humic
acid and fulvic acid) [4,5]. Depending on the industrial phosphoric
acid origin and quality, organic matter content is generally
between 300 and 700 ppm [1], these organic impurities affect
the acid grade.
Among the methods used for decreasing the organic matter
concentration, adsorption by activated carbon can be mentioned.
This method is particularly used in the pretreatment of industrial

* Corresponding author. Tel.: +216 96803179; fax: +216 74 275 595.

E-mail address: omriabdesslem@yahoo.fr (A. Omri).

phosphoric acid to obtain, by the processes based on liquidliquid

extraction, a high purity acid [6,7]. The choice of the adsorption
treatment should be made in terms of economy, efciency of the
process, and ease of control.
Activated carbon can be considered as a good adsorbent for the
purication of gases, and the removal of organic pollutants from
water (i.e., purication of drinking water, and wastewater) [816].
Their textural and adsorptive properties depend on the precursor,
the activation process used, chemical or physical, the number of
steps, the activating agents, the time of gasication and the
oxidation process [1721]. A variety of sources may be considered
for activated carbon generation; coal, wood char, and agricultural
by-products [22]. Unlike coal and to some extent wood,
agricultural by-products are abundant, low cost and renewable.
Studies have reported that almond shell-based material could be
activated by various methods to prepare activated carbons [23
27]. However, none of these studies focused on ammoxidation of
almond shell activated carbons prepared at optimal conditions.
In the present study, carbon dioxide was selected as the
activation gas: CO2 is clean and easy to handle. Moreover, CO2 has
been a preferred choice as the activation gas on the laboratory
scale: it facilitates control of the activation process due to the slow
reaction rate at temperatures around 800 8C [28,29]. Gonzalez et al.
[30] conducted their investigation of CO2 activation with both
vertical and horizontal furnaces. They have concluded that a

1226-086X/$ see front matter 2013 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jiec.2013.03.025

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

horizontal furnace is advantageous for micropore development.

Due to acidic role of CO2 (weak Lewis acid), it is expected that the
introduction of Lewis bases onto the activated carbon surfaces may
favor the adsorption performance of these materials. One of the
popular ways used for the preparation of activated carbon with
increased basicity is to remove or neutralize the acidic functionalities, and other way is to replace acidic groups with proper basic
groups (e.g., basic nitrogen functionalities). Nitrogen containing
functionalities can be introduced through either reaction with
nitrogen containing reagents (such as NH3, nitric acid, and amines)
or activation with nitrogen containing precursors [3135]. In this
context, we proposed the ammoxidation process to modify the
surface of our prepared activated carbons.
In this paper, we describe the performance of activated carbons
made via steam activation of almond shells.
In this study, Almond shell obtained from the local company in
Sfax (Tunisia) has been used for preparing activated carbon. To
optimize the preparation method, the effect of the main process
parameters (activation temperature, carbon dioxide ow rate,
activation time) on the performances of the prepared activated
carbons (expressed in terms of iodine and methylene blue
numbers) was studied. The surface characteristics of the activated
carbon prepared under optimized condition were examined by
several techniques. Adsorption characteristics of this sorbent were
also determined.
To enhance the adsorption capacity of this carbon for the
organic compounds, a modication of the chemical characteristics
of the sorbent surface was performed, using ammoxidation
treatment. The efciency of this treatment was evaluated
considering the elimination of total organic carbon contained in
Tunisian industrial phosphoric acid manufactured from phosphate
rock by the SIAPE society using wet process.
2. Experimental
2.1. Preparation conditions of activated carbons
Almond shell was rstly washed and subsequently dried at
105 8C for 24 h to remove the moisture content. The dried samples
were ground and sieved to the size of 12 mm. Carbonization of the
raw shells and activation of the resulting chars were both carried
out in a vertical stainless-steel reactor, which was placed in an
electrical furnace Nabertherm. During the carbonization process,
about 15 g of raw material was used to prepare the chars. Nitrogen
gas at a ow rate of 150 cm3/min was passed through the reactor
right from the beginning of the carbonization process. The furnace
temperature was increased at a rate of 5 8C/min from room
temperature to 450 8C and held at this temperature for 1 h. After
pyrolysis, the furnace was cooled down to room temperature with
N2 ushing through the sample. The resulting chars were then
activated by a CO2 ow to prepare for the nal product at xed
heating rate at 10 8C/min. For the activation process, the effects of
activation temperature, activation time and carbon dioxide ow
rate were studied to obtain the optimum conditions for the
process.
2.2. Optimum conditions for the production of activated carbon
Some parameters which had effect on the activated carbon
namely impregnation ratio of activation temperature (600
900 8C), activation time (30120 min) and carbon dioxide ow
rate (50200 cm3/min) were studied to determine the optimum
conditions for producing this activated carbon. In this work,
efciency and quality of the activated carbon were preliminarily
characterized by measuring both iodine number and methylene
blue number. Iodine number can be used for estimation of the

2093

relative surface area and measurement of porosity, the pores size

greater than 1.0 nm in diameter. Therefore, the iodine number was
measured to evaluate the adsorptive capacity of the produced
activated carbon. In case of methylene blue number, it is also one of
the most widely recognized probe molecules for assessing the
removal capacity of the specic carbon for moderate-size pollutant
molecules (1.5 nm) [36].
2.2.1. Adsorption of iodine
The iodine number of activated carbon was obtained on the
basis of the Standard Test Method, ASTM Designation D4607-86 by
titration with sodium thiosulphate (ASTM, D 4607-86). The
concentration of iodine solution was thus calculated from total
volume of sodium thiosulphate used and volume dilution factor.
2.2.2. Adsorption of methylene blue
Methylene blue adsorption was conducted by mixing 0.2 g of
the prepared activated carbon with 100 mL of 1000 mg/L methylene blue solution [37]. After agitation during 24 h, the suspension
was ltered and the methylene blue residual concentration was
measured at 660 nm, using an UV/vis spectrophotometer (Visible
spectrophotometer 1011, CECIL). A previously established linear
BeerLambert relationship was used for the concentration
determination.
2.3. Characterization of the prepared activated carbon
2.3.1. Textural characterization
Textural characteristics were determined by nitrogen adsorption at 196 8C with an automatic adsorption instrument (ASAP
2010, Micromeritics). Prior to the measurements, the samples
were out gassed at 300 8C under nitrogen for at least 24 h. The
surface area of the sample was calculated by BrunauerEmmet
Teller (BET) method in relative pressure (P/P0) range of 0.050.30
at 196 8C.
The activated carbon porosity was given by pycnometry. In
order to determine the real density (rs), water was chosen since it
can penetrate in porous space [38]. Whereas, for the apparent
density (rp), mercury was chosen since it does not penetrate in the
porous network. The porosity (x) can be calculated by following
equation:
x

rs  r p
 100
rs

(1)

In addition, a scanning electron microscopy (Philips XL30) was

used to visualize morphology of the raw material (almond shell)
and the activated carbon prepared in the optimal conditions.
2.3.2. Chemical characterization
The surface organic functional groups and structure were
studied by FTIR spectroscopy. The FTIR spectra of the raw material,
optimal activated carbon and ammoxidated sample were recorded
between 500 and 4000 cm1 in a NICOET spectrometer.
The well-known Boehms method allows modeling the principal acidic oxygenated functions of the activated carbon such as
carboxylic acids, lactones and phenols using bases of increasing
strength as NaHCO3, Na2CO3 and NaOH, respectively. Then, the
total basicity is given by titration by HCl. More details are given in
Refs. [39,40].
The pH of the activated carbon was determined as follows: 5.0 g
of carbon is weighed into a 250 ml beaker and 100 ml of water is
added. The beaker is heated to boiling temperature for 5 min. The
decanted portion is cooled to room temperature and pH value is
measured.
The NMR spectrum was recorded in a Bruker (300 MHz)
spectrometer operating at a carbon resonance frequency of

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

75.467 MHz and at 20 8C. 13C cross-polarization magic-angle

spinning (CP/MAS) and magic-angle spinning (MAS) were used
to characterize the optimal activated carbon.
The surface of inorganic components were analyzed by an XRay diffractometer a Philips1 PW 1710 diffractometer (Cu Ka,
40 kV/40 mA, scanning rate of 2u per min).
2.3.3. TOC removal test
The industrial acid solution used was a 54% P2O5 (about 9 M
H3PO4) solution containing an initial concentration of total organic
carbon (COT) equalizes with 578 ppm. The removal of TOC was
carried out in the following manner: a known amount of prepared
activated carbon was placed in Erlenmeyer asks of 250 ml
capacity in contact with 100 g of industrial phosphoric acid
solutions at a certain temperature (4060 8C) under variable
agitation (200600 rpm) and a very acidic value pH. A reux
refrigerant is placed on each Erlenmeyer to avoid water evaporation. After contact desired time, carbon particles and the
supernatant acid were separated by ltration on an organic
membrane of convenient porosity. TOC was determined by a
Shimadzu TOC-5050 apparatus.
The removal efciency of TOC by prepared activated carbon was
calculated as follows:
% Removal of TOC

TOC0  TOCr
 100
TOC0

(2)

where TOC0 and TOCr are the concentrations of total organic

carbon initial and residual, respectively.
3. Results and discussion
3.1. Optimum conditions for the production of activated carbon

30
25
20
15
10
5
0
500

1000

Iodine number

200

0
500

1000

800

600

400

200

Methylene blue number

900

3.1.2. Activation time

The effect of activation time on iodine and methylene blue
number was studied by varying the time from 30 to 120 min at
constant activation temperature of 800 8C and carbon dioxide ow
rate about to 100 cm3/min. Fig. 3 shows that the increasing of the
activation time results in the increase of the iodine and methylene
blue numbers. The results present that 120 min is the suitable
period of time for activation because it gave the highest iodine and

1000

400

800

numbers at constant activation time of 2 h and carbon dioxide ow

rate equal to 100 cm3/min.
Additionally, the yield of activated carbon dened as the ratio of
the mass of activated carbon produced upon the mass of the raw
material used, is decreased for the activation temperature
increases from 600 to 900 8C (Fig. 2). This is attributed to the
removal of volatile matters resulting from the decomposition of
major compounds of almond shell i.e. cellulose (long polymer of
glucose without branches) and hemicellulose (constituted of
various branched saccharides) [43].
Above 800 8C, the yield becomes constant because all cellulose
and hemicellulose are decomposed. On the other hand, due to its
more difcult decomposition, the third component of almond
shell, lignin, remains stable. Accordingly, 800 8C may be chosen as
the optimum activation temperature.

1200

600

700

Fig. 2. Effect of activation temperature on the yield of activated carbon.

1200

800

600

Activation temperature (C)

Adsorption capacity (mg/g)

Adsorption capacity (mg/g)

3.1.1. Activation temperature

Activation temperature was argued to have remarkable effects
on both surface area and the nature of porosity. The increase in
temperature would intensify the dehydration and elimination
reaction and also increase the CCO2 reaction increasing hence the
released volatiles [41,42]. Activation temperature signicantly
enhances both micro-and mesopore developments, resulting in the
increase of both iodine and methylene blue adsorptive capacities of
the activated carbon. As shown in Fig. 1, starting from 800 8C, the
prepared activated carbon has high iodine and methylene blue

35

Activated carbon yield (%)

2094

Iodine number
Methylene blue number

600

700

800

900

1000

Activation temperature (C)

Fig. 1. Effect of activation temperature on iodine and methylene blue numbers of
optimal activated carbon (activation time = 120 min; carbon dioxide ow
rate = 100 cm3/min).

30

60

90

120

150

Activation time (min)

Fig. 3. Effect of activation time on iodine and methylene blue numbers of optimal
activated carbon (activation temperature = 800 8C; carbon dioxide ow
rate = 100 cm3/min).

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

2095

1200

Adsorption capacity (mg/g)

1000

800

600

400

200

Iodine number
Methylene blue number

50

100

150

200

250

Carbon dioxyde flow (cm3/min)

Fig. 4. Effect of carbon dioxide ow rate on iodine and methylene blue numbers of
optimal
activated
carbon
(activation
temperature = 800 8C;
activation
time = 120 min).

methylene blue numbers of product with a maximum value of

1060.43 mg/g and 414.78 mg/g, respectively. This trend is
expected because some volatile compounds which were in the
inner part of particle could be released more with a long activation
because the heat penetrated deeper inside of particles than at short
time. As a result, the porosity development during long time was
higher than during short time resulting in the increase of iodine
number [44].
3.1.3. Carbon dioxide ow rate
The effect of carbon dioxide ow rate on the iodine and
methylene blue number is given in Fig. 4. Increasing the CO2 ow
rate from 50 to 100 cm3/min increases the adsorption capacity of
the activated carbons for the molecules of iodine and methylene
blue. This means that, at a ow rate of 50 cm3/min, the carbon
burnoff due to carbonCO2 reaction is not sufciently high to
maximize the BET surface area in the activated carbons. At 50 cm3/
min, it is also possible that the volatiles released from the samples
might not be completely entrained in the low-velocity CO2 ow
but be deposited on the sample surface again, thereby hindering
pore development [45]. At a ow rate of 100 cm3/min, the iodine
and methylene blue numbers are maximum values. For ow rates

Fig. 6. SEM image of resulting activated carbon.

of 150 cm3/min and greater, the higher ow velocities above that

for 100 cm3/min reduce the contact time between the CO2
molecules and the sample surface and thereby decrease the
diffusion of the molecules into the pore structures and hence
reduced adsorption capacity was obtained.
Note that all experiments were conducted in optimal conditions
of activation temperature and contact time.
3.2. Characterization of the optimal activated carbon
According to Section 3.1, optimized conditions were selected as
activation temperature at 800 8C, a carbon dioxide ow rate of
100 cm3/min and an activation time equal to 120 min. Fig. 5 shows
nitrogen adsorption/desorption isotherm curve onto the activated
carbon prepared at optimized conditions at 196 8C. The obtained
isotherm curve belongs to a mixed type in the IUPAC classication,
type I at low pressure and type IV at intermediate and high P/P0. In
the initial part, they are type I, with an important uptake at low
relative pressures, the characteristic of microporous materials.
However, the open knee is presented, no clear plateau is attained
and certain slope can be observed at intermediate and high relative
pressures. All these facts indicate the transition from microporosity to mesoporosity (type IV). Additionally, a desorption hysteresis
loop due to adsorbate condensation in the mesopores is presented.
The adsorption isotherm of this material showed a good agreement
with those reported in the literature [46].
Microstructure of the resulting activated carbon prepared in the
optimal conditions in Fig. 6 shows that the adsorbent gives roughly
texture with heterogeneous surface and a variety of randomly
distributed pore size. Furthermore, it contains an irregular and
highly porous surface, indicating relatively high porosity and
surface area. This observation can be supported by the structure
parameters of the prepared activated carbon as illustrated in
Table 1.

Table 1
Porous structure parameters of the activated carbons.
Parameters

Fig. 5. Nitrogen adsorptiondesorption isotherms of optimal and ammoxidated

activated carbons.

BET surface area, SBET (m2/g)

Pore volume (cm3/g)
Pore diameter (nm)
Porosity, x (%)

Sample
Optimal
activated
carbon

Optimal activated
carbon ammoxidated

1310.134
0.621
1.896
74

1398
0.84
2.4
77

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

2096

Table 2
Concentrations (mequiv/g) of acidic and basic groups of the surface in activated
carbons.
Type activated carbon

Optimal activated carbon

Optimal activated carbon ammoxidated
Norit
Chemviron F400

Fig. 7. X-ray diffraction prole of optimal activated carbon.

For this activated carbon prepared under optimized condition,

the X-ray diffraction patterns shown in Fig. 7 exhibit broad peaks
and absence of a sharp peak that revealed predominantly
amorphous structure, which is an advantageous property for
well-dened adsorbents. However, the occurrence of broad peaks
around 258 and 438 showed signs of formation of a crystalline
carbonaceous structure, resulting in better layer alignment [47].
Similar result reported was proposed by Pechyen et al. [48]. The
analysis of surface functions covered the carbons allows the
identication of the whole of the acid and basic groupings. Some
examples of oxygen containing functionalities detected on the
carbon surface include the following: carboxylic, lactone, phenol,
carbonyl, pyrone, chromene, quinone, and ether groups [49].
The results of analysis the surface oxygen functional groups of
our activated carbon product (Table 2) shows that the quantity of
the acid functions is much more important than those of the basic
functions. Thus, it may be concluded that the activated carbons are
of acid types.
The solid-state 13C NMR spectrum of produced activated carbon
shows that the lignocellulosic structures were completely
transformed into a polycyclic material with the preponderance
of heteroaromatic structures, as indicated by the signal observed in
the 110130 ppm region (Fig. 8). Aromatic groups present in the
lignin favor re-ordering of elemental carbon upon a possible
thermal treatment. In addition, the signal in at 125 ppm indicates
the presence of benzene groups associated with these basal planes.
Additionally information about the chemical structure and
functional groups initially found on the surface or provided by the
activation, infrared spectroscopy of the raw material and optimal

Fig. 8. Solid-state

13

C NMR spectrum of produced activated carbon.

Concentrations (mequiv/g)
Acid groups

Basic groups

2.606
1.756
0.408
1.125

1.598
2.251
0.610
2.125

activated carbon is given in Fig. 9. According to the literatures, it

was found that almond shell presents the IR band around
3390 cm1 assigned to OH vibrational stretching of hydroxyl
groups. Peaks occurring at 1411 and 1355 cm1 are all ascribed
to oxygen functionalities such as highly conjugated C5
5O
stretching, CO stretching in carboxylic groups, and carboxylate
moieties [50]. The relatively intense band at 1048 cm1 can be
assigned to alcohol groups (R-OH) [50]. Moreover, FT-IR spectra
of optimal activated carbon shows a relatively low intensity of
OH band, a disappearance of intensity of aliphatic bands around
2919 and 2850 cm1. In addition, the relative intensity of the
bands (1411 and 1355 cm1) decreased in the optimal activated
carbon compared with that of raw material. The above-IR
analysis shows the absence of nitrogen groups in the almond
shell and optimal adsorbent structure, whereas we note the
presence of different oxygen groups, mainly: carbonyl, alcohol,
ethers and esters.

4. Application of optimal activated carbon to the removal of

total organic carbon contained in Tunisian industrial
phosphoric acid
Activated carbon prepared from almond shell at optimum
conditions is characterized by an acidic surface. Considering the
acid type of the majority of organic matter contained in industrial
phosphoric acid, it is necessary to modify surface of the activated
carbon to transform it more alkaline.

Fig. 9. FT-IR spectra of almond shell, optimal activated carbon and ammoxidated
activated carbon.

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

100

4.1. Modication of the surface chemistry of carbons

4.1.2. Effect of ammoxidation on the surface functions

The concentrations of acid groups and the basic group are
detailed in Table 2. One can note rst of all, for the ammoxidated
samples, an expected increase in the total basicity occurs.
Secondly, the phenols and the lactones undergo a decrease in
their concentration. The amount of carboxylic acids decreases also
but more slightly. Lactones are cyclic esters, thus they could react
easily with ammonia to give amide groups. The carboxylic acids
could also react with ammonia giving the same nitrogenated
functions according to the following reaction [51]:
H2 O

H2 O

(3)

Moreover, some dihydroxy-phenols became quinones by oth

effects of the oxygen and the high temperature [51]:
OH NH3 ! NH2 H2 O

(4)

Also, modication of surface functions was determined by

analyzing the infrared spectrum of the ammoxidated activated
carbon (Fig. 9). Comparing this spectrum with that optimal
activated carbon, we can essentially notice: appearance band at
2220 cm1 revealed the presence of nitrile groups [52]. The band at
1720 cm1 disappears and the ammoxidated sample shows C5
5O
band centered 1670 cm1 which corresponds to amide functions.
The peak at 1110 cm1 for the modied activated carbon was more
evident than that for the virgin carbon, which corresponded to the
CO bond stretching of CH2OH units [53].
The porous properties of the ammoxidated activated carbon can
be seen in Table 1. The BET specic surface area, total pore volume,
average pore diameter and porosity of the modied sample
increased as compared to the optimal activated carbon.

80

% Removal of TOC

4.1.1. Ammoxidation of the resulting activated carbon

The ammoxidation was performed on a glass sintered inside a
glass tube reactor where a ow of air and ammonia circulates. A
mass of activated carbon equal to 10 g is used to avoid large
differences in terms of ammoxidation degree. The ows of the air
and the ammonia are 6000 and 1200 ml/h, respectively, controlled
with two rotameters. This apparatus is then introduced in a
circular furnace for 6 h at a xed temperature of 400 8C. The power
of heating is controlled with a thermocouple that is in contact with
the surface of the sample. After reaction, the removed sample is
mixed and placed in a dessicator under vacuum at 110 8C. To
complete the elimination of ammonia, the activated carbon is
washed with 2000 ml of distilled water under agitation for 24 h
and then dried at 110 8C.

COO NH4 ! CONH2 ! C  N

2097

60

40

Optimal activated carbon ammoxidated

20

Optimal activated carbon

0
0

30

60

90

120

150

180

210

Contact time (min)

Fig. 10. Effect of specic consumption on the removal of TOC contained in Tunisian
industrial phosphoric acid (initial TOC concentration = 578 ppm; contact
time = 150 min; temperature = 60 8C; agitation speed = 400 rpm).

activated carbon is expected, because of the increased adsorbent

surface area and availability of more adsorption sites [54,55]. After
a certain dosage, the adsorption efciency was not increased
signicantly. It can be seen that the percentage of removal of TOC
on the modied activated carbon is much higher than the virgin
sample. The high adsorption capacity of these samples can be
related to a strong attraction between basic compounds of the
surface carbon (nitrogenous groups) and the organic matter
contained in industrial phosphoric acid.
For further adsorption experiments the maximum specic
consumption of optimal and ammoxidated activated carbon were
selected as 16 and 13.5 g AC/kg P2O5, respectively.
4.2.2. Effect of initial TOC concentration
To study the effect of the initial concentration, we used another
acid phosphoric industrial (28%) with a concentration of TOC is
equal to 365 ppm. The results presented in Fig. 11 show that the
percentage removal of TOC increased with the initial concentration
for the two adsorbents studied. The higher elimination at lower
concentration may be due to the presence of more available sites
on the adsorbent than the number of organic compounds which
are available in the solution. However, at higher concentration the
number of compounds is relatively higher than available sites for
adsorption. For the comparison, the modied activated carbon has

4.2. TOC removal tests

4.2.1. Effect of specic consumption
Fig. 10 shows the effects of specic consumption on the removal
of TOC contained in industrial phosphoric acid (54% P2O5) at 60 8C
onto virgin activated carbon and modied activated carbon. The
specic consumption was calculated by using the following
equation,
Specific consumption

mAC
mP2 O5

(5)

where mAC is the mass of the activated carbon (g), mP2 O5 is the mass
of P2O5 contained in industrial phosphoric acid (kg).
The specic consumption is an important parameter because
this parameter determines the capacity of adsorbent for a given
TOC concentration and also determines sorbentsorbate equilibrium of the system. The increased removal at high consumption of

Fig. 11. Effect of initial TOC concentration on TOC removal (contact time = 150 min;
temperature = 60 8C; agitation speed = 400 rpm).

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

100

100

80

80

% Removal of TOC

% Removal of TOC

2098

60

40

Optimal activated carbon ammoxidated

20

60

40

20

Optimal activated carbon ammoxidated

Optimal activated carbon

0

30

60

90

120

Optimal activated carbon

150

180

210

100

200

Contact time (min)

300

400

500

600

700

Agitation speed (rpm)

Fig. 12. Effect of contact time on TOC removal (initial TOC concentration = 578 ppm;
temperature = 60 8C; agitation speed = 400 rpm).

Fig. 14. Effect of agitation speed on TOC removal (initial

concentration = 578 ppm; contact time = 150 min; temperature = 60 8C).

the highest percentage elimination of TOC. Similar observation was

reported by other studies [56].

due to the decrease in the viscosity of the solution [57]. The

maximum percentage removal of TOC reaches 78% and 69% for
optimal activated carbon and modied sample, respectively at
60 8C.

4.2.3. Effect of contact time

Contact time is an important parameter to determine the
equilibrium time of adsorption process. The characteristics of
activated carbon and its available sorption sites affected the time
needed to reach the equilibrium. The experimental results for the
determination of equilibrium time are given in Fig. 12 and it is
obviously seen that removal capacity increased with the contact
time. Large amounts of TOC were removed in the rst 90 min and
the equilibrium was reached in 150 min for both virgin and
ammoxidated activated carbon.
4.2.4. Effect of temperature
Fig. 13 shows the experimental results obtained from TOC
removal when temperature was varied from 30 to 60 8C. The
removal efciency increased with an increase in temperature. The
increase in removal efciency indicates an endothermic process. At
higher temperature, more organic molecules have sufcient
energy to undergo an interaction with active sites at the surface.
In addition, increasing the temperature was known to increase the
rate of diffusion of the adsorbate molecules across the external
boundary layer and in the internal pores of the adsorbent particle

TOC

4.2.5. Effect of agitation speed

The effect of agitation speed on TOC removal is shown in Fig. 14.
As the agitation rate increased from 200 to 400 rpm, the adsorption
capacity increased for the two adsorbents. The decrease in
elimination efciency at a higher shaking rate is probably due
to the fact that more particles are broken for agitation above
400 rpm, thereby enhancing the desorption process. Another
explanation is that the suspension is not homogeneous [6,56,58].
4.3. Comparison to commercially available adsorbent materials
Fig. 15 provides a comparison of removal capabilities for
optimal activated carbon and oxidized sample relative to
commercial activated carbons designed for TOC removal. The
commercial activated carbons, Norit and chemviron 400, have
surface area of 941 and 1200 respectively. The surface functional
groups with oxygen of these activated carbons are presented in
Table 2.
700

100

Norit
Optimal activated carbon
Chemviron F400
Optimal activated carbon ammoxidated

600
500

TOC residual (ppm)

% Removal of TOC

80

60

40

20

400
300
200
100

Optimal activated carbon ammoxidated

Optimal activated carbon

0
20

30

40

50

60

70

50

100

150

200

Contact time (min)

Temperature (C)
Fig.
13.
Effect
of
temperature
on
TOC
removal
(initial
TOC
concentration = 578 ppm; contact time = 150 min; agitation speed = 400 rpm).

Fig. 15. A comparison of optimal activated carbon oxidized and commercial

activated carbons (initial TOC concentration = 578 ppm; temperature = 60 8C;
agitation speed = 400 rpm).

A. Omri et al. / Journal of Industrial and Engineering Chemistry 19 (2013) 20922099

It can be observed that the tested commercial carbons have lower

performance compared to the activated carbon oxidized. The removal
of TOC onto four types of the activated carbon increases with time and
then attains equilibrium value at a time of about 150 min.
4.4. pH variation of the activated carbons
The results of determination of pH of the investigated samples
were as follows: 5.52, 8.04 and 7.59 for optimal activated carbon,
optimal activated carbon ammoxidated and modied activated
carbon after adsorption, respectively. As it is clear from the data,
the pH of virgin sample was acidic due to have carboxylic groups
and other acidic oxygen functionalities; while the pH of aminated
sample was larger than 7 (in the basic range). Ammonia treatment
of the activated carbon through the reaction of NH3 with these
oxygen functional groups causes to formation of nitrogen
functional groups like amine, nitrile and pyridonic groups that
are naturally basic. This leads to produce basic adsorbents with the
pH greater than 7.
After the TOC removal tests, the surface of the activated carbon
was load of a signicant quantity of acids. The pH of this sample
was measured as 7.59.
5. Conclusion
From the above results it may be concluded that, using almond
shell, the CO2-based method of physical activation is very effective
to prepare activated carbon ranging widely in its textural, surface
properties and with well-developed porosity. The most efcient
activated carbon is that obtained under the following optimal
conditions: an activation temperature equal to 800 8C, an
activation time of 120 min, and a carbon dioxide ow rate of
100 cm3/min. Using ammoxidation treatment, chemical modication of optimal activated carbon was performed to improve its
surface characteristics for the removal of total organic carbon from
Tunisian industrial phosphoric acid. The results revealed that the
BET surface area increased from 1310 m2/g to 1398 m2/g after
chemical treatment. The FTIR showed that the chemically treated
activated carbon has peaks at 2220 cm1 and 1670 cm1 indicating
the presence of nitrile and amide functions respectively. Due to the
enhanced surface properties, the removal efciency of the
modied activated carbon was found to be 78% of TOC. The
adsorption characteristic has been examined with the variations in
the parameters of specic consumption, contact time, initial TOC
concentration, agitation speed and temperature.
According to these results, the ammoxidation seems to be an
effective treatment for modifying the chemical structure of
activated carbon. These studies strengthen the prospects for
quantifying surface functions by other techniques and use these
activated carbons in different industrial applications.
Acknowledgements
The authors gratefully acknowledge nancial support from the
Tunisian Chemical Group. They also wish to express their gratitude
to Mr. A. Char, Mr. L. Fourati, K. Foued, W. Trabelsi and Mme N.
Ammar for their help and support. Thanks are due to Mr. Z.
Fakhfakh from Faculty of Science of Sfax for the assistance in MEB.
Also, I thank Mr. Houcine SEFI for this linguistic assistance in this
manuscript.
References
[1] P. Becker, Phosphates and Phosphoric Acid, Raw Materials, Technology and
Economics of the Wet Process, 2nd ed., Marcel Dekker, Inc., New York, 1989.
[2] Sulfuric/Phosphoric Acid Plant Operation, American Institute of Chemical Engineers, New York, 1982.

2099

[3] Atmospheric Emissions from Thermal Process Phosphoric Acid Manufacture, AP-48,
US Environmental Protection Agency, Research Triangle Park, NC, October 1968.
[4] B. Nasr, B. Hedi, G. Abdellatif, M. Rodrigo, Chemical Engineering and Technology
28 (2005).
[5] A. Mellah, A. Silem, R. Kada, A. Boualia, Canadian Journal of Applied Spectroscopy
36 (1991) 94.
[6] A. Silem, A. Boualia, A. Mellah, R. Kada, Canadian Journal of Chemical Engineering
70 (1992) 491.
[7] H. Belfadhel, A. Ratel, A. Ouederni, S. Besr, J.C. Mora, Science & Engineering 17
(1995) 637.
[8] I. Ali, Separation and Purication Technology 39 (2010) 95.
[9] I. Ali, V.K. Gupta, Nature Protocols 1 (2007) 2661.
[10] I. Ali, Separation and Purication Technology (2013), http://dx.doi.org/10.1080/
15422119.2012.748671.
[11] M. Berrios, M. Angeles Martn, A. Martn, Journal of Industrial and Engineering
Chemistry 18 (2012) 780.
[12] I. Ali, M. Asim, T.A. Khan, Journal of Environment Management 113 (2012) 170.
[13] I. Ali, Chemical Reviews 112 (2012) 5073.
[14] A. Omri, M. Benzina, Desalination and Water Treatment 97 (iFirst) (2012) 1.
[15] A. Omri, M. Benzina, Alexandria Engineering Journal 51 (2012) 343.
[16] A. Omri, A. Wali, M. Benzina, Arabian Journal of Chemistry (2012), http://
dx.doi.org/10.1016/j.arabjc.2012.04.047.
[17] F.C. Wu, R.L. Tseng, C.C. Hu, C.C. Wang, Journal of Power Sources 144 (2005) 302.
[18] W.T. Tsai, C.Y. Chang, S.Y. Wang, C.F. Chang, S.F. Chien, H.F. Sun, Resources,
Conservation and Recycling 32 (2001) 43.
[19] R.S. Alejandro, A.E. Alfredo, L.T.S. Luis de, Carbon 39 (2001) 1367.
[20] H. Benaddi, T.Z. Bandosz, J. Jagiello, J.A. Schwarz, J.N. Rouzaud, Carbon 38 (2000)
669.
[21] C.S. Ju, H.J. Park, K.S. Jeong, Journal of Environmental Sciences 12 (2003) 1005.
[22] W. Heschel, E. Klose, Fuel 74 (1995) 1786.
[23] K. Thomas Klassona, C.A. Ledbetterb, L.H. Wartellea, S.E. Lingle, Industrial Crops
and Products 31 (2010) 261.
[24] J.M.V. Nabais, C.E.C. Laginhas, P.J.M. Carrott, M.M.L.R. Carrott, Fuel Processing
Technology 92 (2011) 234.
[25] C.A. Toles, W.E. Marshall, M.M. Johns, L.H. Wartelle, A. McAloon, Bioresource
Technology 71 (2000) 87.
[26] M.T. Izquierdo, A.M. de Yuso, B. Rubio, M.R. Pino, Biomass and Bioenergy 35
(2011) 1235.
[27] E. Demirbas, M. Kobya, A.E.S. Konukman, Journal of Hazardous Materials 154
(2008) 787.
[28] F. Rodrguez-Reinoso, Controlled gasication of carbon and pore structure development, in: Fundamental Issues in Control of Carbon Gasication Reactivity,
Kluwer Academic Publishers, Dordrecht, Boston, 1991.
[29] P.J.M. Carrott, Suhas, M.M.L. Ribeiro Carrott, C.I. Guerrero, L.A. Delgado, Journal of
Analytical and Applied Pyrolysis 82 (2008) 264.
[30] M.T. Gonzalez, F. Rodrguez-Reinoso, A.N. Garcia, A. Marclla, Carbon 35 (1997) 159.
[31] T.C. Drage, A. Arenillas, K.M. Smith, C. Pevida, S. Piippo, C.E. Snape, Fuel 86 (2007) 22.
[32] M.C.B. Lopez, A.M. Alonso, J.M.D. Tascon, Microporous and Mesoporous Materials
34 (2000) 171.
[33] P.J.M. Carrott, J.M.V. Nabais, M.M.L.R. Carrott, J.A. Pajares, Carbon 39 (2001) 1543.
[34] C.L. Mangun, K.R. Benak, J. Economy, K.L. Foster, Carbon 39 (2001) 1809.
[35] J. Starck, P. Burg, S. Muller, J. Bimer, G. Furdin, P. Fioux, C.V. Guterl, D. Begin, P.
Faure, B. Azambre, Carbon 44 (2006) 2549.
[36] B.S. Girgis, El-Hendawy, A. Abdel-Nasser, Microporous and Mesoporous Materials
52 (2002) 105.
[37] T. Yang, A. Chong Lua, Materials Chemistry and Physics 100 (2006) 438.
[38] M. Benzina, A. Bellagi, Annali di Chimica 15 (1990) 315.
[39] J. Starck, P. Burg, D. Cagniant, J.M.D. Tascon, A. Martinez-Alonso, Fuel 83 (2004) 845.
[40] H.P. Boehm, Carbon 32 (1994) 759.
[41] A. Bacaoui, A. Yaacoubi, A. Dahbi, C. Bennouna, R. Phan Tan Luu, F.J. MaldonadoHodar, J. Rivera-Utrilla, C. Moreno-Castilla, Carbon 39 (2001) 425.
[42] A.C. Lua, T. Yang, Journal of Colloid and Interface Science 274 (2004) 594.
[43] C. Bouchelta, M.S. Medjram, O. Bertrand, J.-P. Bellat, Journal of Analytical and
Applied Pyrolysis 82 (2008) 70.
[44] Q. Cao, K.C. Xie, Y.K. Lv, W.R. Bao, Bioresource Technology 97 (2006) 110.
[45] T. Yang, A.C. Lua, Journal of Colloid and Interface Science 267 (2003) 408.
[46] A.M. Puziy, O.I. Poddubnaya, A. Martinez-Alonso, F. Suarez-Garcia, J.M.D. Tascon,
Carbon 40 (2002) 1493.
[47] L.J. Kennedy, J.J. Vijaya, K. Kayalvizhi, G. Sekaran, Chemical Engineering Journal
132 (2007) 279.
[48] C. Pechyen, D. Atong, D. Ahtong, V. Sricharoenchaikul, Journal of Solid Mechanics
and Materials Engineering 1 (2006) 498.
[49] T. Yang, A.C. Lua, Materials Chemistry and Physics 100 (2006) 438.
[50] R. Baccar, J. Bouzid, M. Feki, A. Montiel, Journal of Hazardous Materials 162 (2009) 1522.
[51] M.A. Montes-Mor n, D. Su rez, J.A. Menendez, E. Fuente, Carbon 42 (2004) 1219.
[52] L.J. Bellamy, The Infrared Spectra of Complex Molecules, Advances in Infrared
Group Frequencies, vol. 2, Chapman and Hall, London and New York, 1980pp.
294298.
[53] Z. Zhang, M. Xu, H. Wang, Z. Li, Chemical Engineering Journal 160 (2010) 571.
[54] K.B. Bota, G.M.K. Abotsi, Fuel 73 (1994) 1354.
[55] A. Sari, M. Tuzen, D. Citak, M. Soylak, Journal of Hazardous Materials 148 (2007) 387.
[56] A. Boualia, A. Mellah, T.T. Aissaoui, K. Menacer, A. Silem, Applied Clay Science 7
(1993) 431.
[57] B.H. Hameed, M.I. El-Khaiary, Journal of Hazardous Materials 154 (2008) 237.
[58] A.D. Souza, P.S. Pina, V.A. Leao, C.A. Silva, P.F. Siqueira, Hydrometallurgy 89 (2007)
72.