Colloids and Surfaces A 578 (2019) 123505

Contents lists available at ScienceDirect

Colloids and Surfaces A

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

Removal of toxic dyes from aqueous solution using new activated carbon T
materials developed from oil sludge waste
Xiaoyu Lia,b, Dong Hanb, Mingyang Zhangb,c, Bin Lib,d, Zhenbo Wangb, Zhiqiang Gongb,
Peikun Liua, Yuekan Zhanga, Xinghua Yanga,
⁎

a
College of Mechanical and Electronic Engineering, Shandong University of Science and Technology, Qingdao, 266590, China
b
State Key Laboratory of Heavy Oil Processing, College of Chemical Engineering, China University of Petroleum (East China), Qingdao, 266580, China
c
School of Thermal Engineering, Shandong Jianzhu University, Jinan, 250101, China
d
School of Energy and Power Engineering, Jiangsu University, Zhenjiang, 212013, China

GRAPHICAL ABSTRACT

ARTICLE INFO ABSTRACT

Keywords: The development of waste-derived adsorbent is of great significance to realize the idea of “treating waste with
Oil sludge waste”. Herein, a novel porous activated carbon was synthesized from oil sludge waste (OSAC) in a facile way.
Activated carbon The resultant OSAC possesses favorable properties such as high specific surface area (324.98 m2 g−1) and
Dyes abundant surface oxygen-containing functional group, which contribute to outstanding adsorption capabilities
Adsorption mechanism
for Methylene Blue (MB, 316.02 mg g−1) and Direct Blue 6 (DB, 124.24 mg g−1) dyes. The adsorption capacity of
OSAC is closely related to the initial solution pH since the surface charge state of the adsorbent is affected by pH
values. With the increasing of pH values, the adsorption capacity of MB raised while that of DB decreased. The
adsorption isotherms, kinetics, thermodynamics analyses have been investigated. Although the Langmuir model
and the pseudo-second order model fitted well to the experimental data for both dyes and the thermodynamic
parameters revealed their adsorption processes were all endothermic and spontaneous, the adsorption me-
chanism of the two dyes is not exactly the same. In addition to the difference of physical adsorption caused by
dyes molecular size, the difference in chemical adsorption effects due to the different charge properties of the
two dyes was also an important aspect of the adsorption mechanisms. Moreover, the reutilization analysis shown

⁎
Corresponding author.
E-mail address: sdustyxh2018@163.com (X. Yang).

https://doi.org/10.1016/j.colsurfa.2019.05.066
Received 22 December 2018; Received in revised form 20 May 2019; Accepted 25 May 2019
Available online 27 May 2019
0927-7757/ © 2019 Elsevier B.V. All rights reserved.
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

a presentable recyclability of the OSAC. It is demonstrated that the strategy developed here would provide cost-
effective production of adsorbents for treatment of dye wastewater and offer a promising avenue of fabrication of
high-value-added products from hazardous industrial waste.

1. Introduction through the corresponding dilution of dye stock solution with deionized
water. All the chemical reagents are purchased in Sinopharm (China)
With increasing threats of resource shortage, climate change and and used as received without further purification.
environmental pollution, achieving sustainable development have
gradually become a consensus of the world [1]. In the petroleum in-
dustry, a huge amount of oil sludge (OS) was produced during the 2.2. Preparation of OSAC
process of oil exploration, transportation, storage and refinery [2,3]. It
usually contains recalcitrant components, such as water, petroleum In a typical process, the char obtained from pyrolysis of OS sample
hydrocarbons and heavy metal salts, which seriously threaten to the at 700 °C was ground and soaked in 1 M HF solution for 24 h to remove
environment [4] and human health [5]. And even worse, more than 60 inorganic impurities. And the de-ashed char (ODSC) was generated.
million tons of the OS was increased per year. Due to the hazardous Then, the resulting OSDC was impregnated in H3PO4 (50%) and dried
composition characteristics and increasing amount, harmless disposal at 105 °C for 4 h. Subsequently, the activation processes were per-
measures have become an urgent need. More importantly, if the oil in formed at 400 °C for 1 h in a tube furnace with a heating rate of
OS cannot be effectively recovered, there will also be a great waste of 10 °C min−1 and a N2 flow of 20 mL min−1. The oil sludge-based acti-
oil resources. Therefore, appropriate utilization of OS has become a vated carbon (OSAC) was obtained after being washed with plenty of
technical problem worldwide [6]. deionized water and drying at 105 °C for 24 h. OSAC synthesized with
In recent years, varies of OS treatment technologies have been de- different H3PO4/C ratios of 2:1, 2.5:1, and 3:1 were named as OSAC-
veloped, including centrifugation, solvent extraction, incineration, 400-2, OSAC-400-2.5, OSAC-400-3, respectively. And OSAC products
pyrolysis, and so on [7]. The main purpose of these methods is to re- prepared at different activation temperatures of 300 and 500 °C with
cycle the waste oil in OS as fuel resources. However, only a few works the same H3PO4/C ratio of 2.5:1 were named as OSAC-300-2.5 and
paid attention to the utilization of OS directly as raw material to pro- OSAC-500-2.5, respectively.
duce other high-value-added products [8]. Some studies used OS to
produce construction materials such as red ceramics [9] and bricks 2.3. Characterization
[10], which exhibit a certain degree of mechanical properties. How-
ever, this method not only has the risk of toxic leaching, but also ig- The morphological features of samples were analyzed by scanning
nores the utilization of hydrocarbons contained in the OS. In response electron microscope (SEM, Hitachi S-4800, Japan) and transmission
to this problem, some researchers have suggested that the pyrolysis electron microscopy (TEM, JEOL-2010, Japan). The porosity char-
residues of OS can be used to prepare adsorbent materials [11]. S. acteristics of samples were investigated by multipoint N2 adsorption-
Mohammadi et.al [12] prepared carbonaceous adsorbents from pyr- desorption test (Micromeritics ASAP 2020, USA). The specific surface
olysis of the OS. The as-obtained carbonaceous adsorbent activated by area was calculated by Brunauer-Emmett-Teller (BET) method and pore
KOH presented good adsorption efficiency of Cd. Guo [13] synthesized size distributions (PSD) were calculated by applying Non-Local-Density-
adsorbents from OS for removing reactive brilliant red X-3B dye in Function-Theory (NLDFT) method to the N2 adsorption branch. The
aqueous solution. However, the preparation methods of them are te- surface functional groups were characterized by Fourier transform in-
dious, or the requirements of experimental conditions and equipment frared spectroscopy (FTIR, Thermo Nicolet NEXUS, USA).
are severe. In addition, the performance of these adsorbents are poor
due to the lack of purity and porosity. Therefore, more convenient and
effective methods for preparing porous activated carbon adsorbent from 2.4. Adsorption experiments
the OS need further research.
In this work, a novel porous activated carbon adsorbents was syn- In the adsorption equilibrium studies, 10.0 mg of OSAC was stirred
thesized from the OS by using an effective two-steps method for the first with 50 mL of MB or DB solution at different concentrations (from 25 to
time. Its pore size structure was developed while the purity of the 800 mg L−1 for MB, from 25 to 500 mg L−1 for DB) in a100 mL conical
material was improved. The influence of preparation parameters on the flask. The flasks were put in a thermostatic water bath oscillator at
textural evolution of OSAC was investigated. In addition, the adsorption 180 rpm for 24 h with the different temperatures of 25 °C, 35 °C and
performance of the OSAC materials for two different dyes (MB and DB) 45 °C, respectively. Then the adsorbent was immediately separated by
were studied in depth. The possible adsorption mechanism of the OSAC syringe filter. The concentrations of MB and DB dyes were determined
was also figured out. This work not only put forward an effective by UV/vis spectrophotometer (PERSEE TU-1900, China) at the wave-
method of high-value-added resource utilization of OS, but also provide length of 665 nm (MB) and 574 nm (DB), respectively [14].
a promising strategy for the preparation of efficient adsorbents for dye In the adsorption kinetic studies, 10.0 mg of OSAC adsorbents was
wastewater treatment. dosed in a series of dye solutions with a same concentration of
200 mg L−1. The residual concentrations of each flask were measured at
2. Experimental methods pre-set time intervals (0.5, 1, 2, 4, 7, 10, 15, 25, 40, 60 min), respec-
tively.
2.1. Materials In order to investigate the effect of solution pH values on the ad-
sorption capacity of OSAC, a series of dye solutions with the same
The OS sample used in this work was collected from a tank bottom concentration of 100 mg L−1 were prepared and adjusted their initial
in Shengli Oilfield Sinopec. The proximate and ultimate analyses of OS pH values from 2 to 12. The pH values were adjusted using 0.1 M HCl
are listed in Table S1. The MB and DB solutions at a concentration of and NaOH. Then 10.0 mg of OSAC adsorbent was added to the solution.
1.0 g L−1 are prepared as dye stock solutions. (The characteristics of The adsorbed amounts in each flask were measured after being equi-
two dyes are listed in Table S2). And the work solutions were derived librium.

2
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

Fig. 1. N2 adsorption-desorption isotherms of different OSAC samples.

3. Results and discussion increasing at the relative pressure near 1.0 indicates the appearance of
macropores.
3.1. Structure characterizations Fig. 2 shows the PSD curves calculated by using NLDFT method
according to these isotherms. It is clear that all the OSAC samples
As shown in Fig. 1, the N2 adsorption-desorption isotherms were present hierarchical porous structure with the majority of pore size fell
measured to evaluate the textural properties of each OSAC sample. in the range of micro- and meso- pores. As shown in Fig. 2a and b,
According to the classification standard of IUPAC, the OSC samples activation temperature has great influence on the pore size distribution.
showed type II isotherm profile, which reflects the typical physical As the temperature raising from 300 to 500 °C, more micropores were
adsorption process on non-porous or macro-pores adsorbents. At the developed first. Then the activation reaction was further deepened, and
same time, all the OSACs samples exhibited type IV isotherms with the micropores gradually expanded into larger mesopores and macro-
H4 hysteresis loop. A sharp increasing curve which attributed to the pores. In Fig. 2c and d, the PSDs of these samples almost overlap with
micropores were discovered at the low relative pressure [15]. The each other, which means that the effect of impregnating ratio is less
convex curves with inflection point afterward represented the mono- than the temperature. But in fact, for the micropores volume of these
layer adsorption. And then the multilayer adsorption gradually oc- samples synthesized at different impregnation ratios from 2 to 3, there
curred. As the pressure continues to rise, the isotherms presented an is a tendency to increase first and then decrease.
increasing slope at higher pressure, which was related to the capillary The detailed textural parameters of as-obtained OSACs are listed in
condensation of N2 in the mesopores [16]. The nitrogen adsorption Table 1. The BET surface area of OSAC-400-2.5 reached maximum

Fig. 2. (a) Pore size distributions and (b) micropore size distributions of OSACs prepared at different temperatures with the impregnation ratio of 2.5, (c) pore size
distributions and (d) micropore size distributions of OSACs prepared with different impregnation ratios at the temperature of 400 °C.

3
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

Table 1
The porosity characteristics of OSACs.
Sample SBET Smirco Smeso Vpore Vmirco Vmeso Daver
(m2 g−1) (m2 g−1) (m2 g−1) (cm3 g−1) (cm3 g−1) (cm3 g−1) (nm)

OSAC-300-2.5 144.48 59.05 85.4 0.202 0.032 0.17 2.79

OSAC-400-2.5 324.98 274.02 50.96 0.249 0.106 0.143 1.53
OSAC-500-2.5 234.86 80.36 154.5 0.259 0.045 0.214 2.21
OSAC-400-2 266.03 228.48 37.55 0.22 0.089 0.131 1.65
OSAC-400-3 240.88 211.54 29.34 0.179 0.085 0.094 1.49

because of the highest micropores ratio. It means that the moderate

activation conditions bring the best activation effect, while excessive
heat treatment or activation agent dosage may arouse the widening or
collapse of the existing micro- or meso-pores structure.
The surface morphology of the OSACs was characterized by SEM
images and TEM images. The surface of OSC samples was rough
(Fig. 3a), but almost no obvious pores can be found in it. The SEM
images of OSDC (Fig. 3b) shows a sponge-like morphology which
caused by acid etching. For the OSAC-400-2.5 sample with the largest
specific surface area, its structure was severely corroded by some
broken fragments and bubbles (Fig. 3c). This may be attributed to the
melting effect of activation agents. The internal pores structure of the
OSAC sample can be further investigated in TEM images. As shown in
Fig. 3d, OSAC-400-2.5 possessed amorphous and 3D-interconnected
porous structure which may enhance its adsorption properties of dyes. Fig. 4. The adsorption capacity of different adsorbents for MB and DB dyes.
[adsorbent dosage: 10.0 mg, solution volume: 50 mL, solution concentration:
100 mg L−1, temperature: 25 °C, initial pH: 7].
3.2. Adsorption capacity of different samples

To investigate the effects of temperatures and activation agent ratio consistent with the trends of specific surface area changes. It suggests
on the adsorption capability of OSACs, the adsorption experiments were that the adsorption performance of OSACs has an important relation-
carried out by adding 10 mg of different samples into the same con- ship with their pore structures. It is worth noting that the pore structure
centration (100 mg L−1) of MB and DB solutions. The adsorption has a lower effect on DB dye adsorption than MB. That may be due to
properties of each sample are different as shown in Fig. 4. With the the large molecular size of DB hindering them from reaching the ad-
increase of temperature from 300 to 500 °C, the two dyes removal rates sorption sites on the surface of OSACs. Among these adsorbents, OSAC-
were all increased first and then decreased. The same tendency was 400-2.5 exhibits the highest adsorption capability for both dyes
observed with the increasing of H3PO4 ratios from 2 to 3. These are (266 mg g−1 for MB and 73.5 mg g−1 for DB). Therefore, unless

Fig. 3. SEM images of (a) OSC, (b) OSDC and (c) OSAC-400-2.5, TEM image of OSAC-400-2.5.

4
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

Fig. 5. (a) FTIR spectra of OSDC and OSAC-400-2.5, (b) determination of the pHpzc for OSAC-400-2.5 by pH drift method, (c) the effect of initial solution pH on the
adsorption capacity of MB and DB [adsorbent dosage: 10.0 mg, solution volume: 50 mL, solution concentration: 200 mg L−1, temperature: 35 °C].

otherwise stated, the following analysis about adsorption performance play important roles in the adsorption process of dye molecules. Fig. 5a
of OSAC is focused on the OSAC-400-2.5. also shows the FTIR spectrum of OSDC, which is the precursor before
activation process. The main difference between the two samples is the
3.3. Effect of initial pH values on adsorption peaks in the range of 1000 cm−1 to 1500 cm−1, which attributed to the
oxygen functional groups primarily as mentioned above. That is be-
The effect of the initial pH values of solutions on the adsorption cause the H3PO4 activation process can introduce more oxygen-con-
performance of OSAC was investigated in this work. At first, the FTIR taining functional group to the surface of OSDC carbon materials. It has
test was performed on OSAC-400-2.5, which is illustrated in Fig. 5a. been manifested by many literatures that oxygen-containing functional
The band around 3420 cm−1 represents the OeH stretching vibration of groups are beneficial to the adsorption of ions dyes in aqueous solution.
the hydroxyl group [17]. The peaks at 2925 cm−1 and 2850 cm−1 were As shown in Fig. 5b, the pHpzc value for OSAC was then determined
assigned to −CH2 group [18]. The band at 1595 cm−1 represents C]C to be about 2.94, which may be caused by the oxygen-contained
stretching vibration of pyrolysis char [19]. The peaks at 1398 cm−1 and functional group as mentioned above. It is well known that when the
1045 cm−1 attributed to C]O stretch carboxylates groups [20]. And pH is higher than pHpzc, the adsorption of positive-charged molecules
the peaks at 1310 cm−1, 1220 cm−1 and 1118 cm−1 can be assigned to was promoted, because the adsorbent surface exhibits a negative net
C–O and C–OeC groups [18]. The wide stretch of −OH group com- charge. On the other hand, the solution pH which is lower than pHpzc
bined with stretch of C]O is diagnostic for carboxylic acid, which may helps to increase the adsorption of negative-charged molecules due to

Fig. 6. Adsorption isotherms of (a) MB and (b) DB onto OSAC-400-2.5, [adsorbent dosage: 10.0 mg, solution volume: 50 mL, initial pH: 7, initial MB concentration:
25–800 mg L−1, initial DB concentration: 25–500 mg L−1].

5
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

the positive charge on the surface of the adsorbent [21]. Since MB higher than KL of DB. It suggests that OSAC has stronger adsorption
usually forms positive-charged ions, and DB is presented as negative- capability for MB than DB. Moreover, Freundlich model can also fit well
charged ions in the solution. Therefore, the adsorption processes of both with the experimental data. The values of heterogeneity factor 1/n of
dyes were all pH-dependent. As is illustrated in Fig. 5c, with the change Freundlich model were all less than 1, which also indicates that the
of the initial pH, the adsorption capacity of the two dyes expressed an adsorption was favorable [24]. But as is well known that the physical
opposite trend. As the initial pH increased from 2 to 12, the adsorption adsorption is exothermic, which preferring lower temperature [25].
capacity of MB increased from 255.7 mg g−1 to 334.8 mg g−1, while the This is obviously contrary to the above-mentioned results. Therefore,
adsorption capacity of DB decreased from 145.2 mg g−1 to we can infer that these adsorption processes also contain chemical
82.4 mg g−1. These results respond to the surface charges of the OSAC, adsorption effects which may result in higher adsorption capacities. The
which were getting negative with the increasing of pH values. It is Temkin adsorption model considers the chemical adsorption interac-
beneficial to the adsorption of MB cations rather than DB anions. This tions of adsorbate and adsorbent [18]. The well-fitting of this model
means that the chemical adsorption caused by electrostatic attraction suggests that the electrostatic attractions or repulsion contributes to the
plays a significant role in the process of dyes adsorption onto OSAC. mechanism for MB and DB dyes adsorbed by OSAC. The Temkin model
predicts a uniform distribution of binding energies over the population
of surface binding adsorption sites [26]. The range and distribution of
3.4. Adsorption isotherms analysis
the binding energies should depend on the density and distribution of
functional groups both on the dyes and adsorbent surfaces [27,28].
The adsorption isotherms of OSAC towards MB and DB are shown in
Temkin model was usually used to determine the adsorption potentials
Fig. 6. The adsorption capacities of these two dyes increased with the
of the adsorbent for adsorbates. As shown in Table 2, the Temkin ad-
increase of the initial concentration of dyes. That is due to higher
sorption potential KT of MB is much higher than DB, which means that
concentration may produce stronger driving force to promote diffusion
there was electrostatic attraction between MB and OSAC, but electro-
of dye molecules from aqueous phase to the solid phase. Additionally,
static repulsion between DB and OSAC.
the maximum adsorption amount of dyes increased with the increasing
In addition, the maximum adsorption capacities of OSAC predicted
temperature, which are 284.2, 302.9, 316.1 mg g−1 for MB and 104.3,
by Langmuir model were 284.90, 303.95, 317.46 mg g−1 for MB and
112.5, 124.2 mg g−1 for DB at 25, 35 and 45 °C, respectively.
108.11, 123.61, 134.95 mg g−1 for DB at 25, 35, 45 °C, respectively. For
The isotherms experimental data were fitted by Langmuir,
one thing, the MB and DB adsorption capacities increased against the
Freundlich, Dubinin-Radushkevich (D-R) and Temkin isotherm models,
increment of the operating temperature, which further confirms that
respectively. The fitting curves are illustrated in Figs. S1 and S2 and the
increasing temperature favored the adsorption process. For the other
isotherms fitting parameters are presented in Table 2. The Langmuir
thing, these values are higher than many adsorbents reported in other
model fit the experimental data better for the two dyes, since the cor-
literatures (as shown in Table S3). Therefore, this novel OSAC ad-
relation coefficients (R2) were largest and higher than 0.99. As we all
sorbent has strong practical value for dyes removal.
know, Langmuir model considers the adsorption process is monolayers
adsorption phenomenon on homogeneous surfaces. Hence, the mono-
layer physical adsorption played a significant role in the adsorption of 3.5. Adsorption kinetic analysis
MB or DB on OSAC [22]. In addition, the separation factors (RL) which
provide important information on the adsorption nature are also pre- Adsorption of dyes onto porous solid material is often complex,
sented in Table 2. The values of them fell into the range of 0–1, which which may be affected by a variety of factors [22]. Therefore, besides
indicates that the adsorption of dyes on OSAC is favorable [23]. the isotherms analysis, the kinetic investigation is an important step to
However, the Langmuir constant KL of the MB adsorption is much gain insight into the rate-controlling factors and possible mechanisms of

Table 2
Adsorption isotherm parameters for MB and DB adsorption onto OSAC at different temperature.
Adsorbate MB DB

Temperature (°C) 25 35 45 25 35 45

−1
qe.exp (mg g ) 284.17 302.88 316.02 104.32 112.53 124.24
Langmuir ce
=
1
+
ce
RL =
1
qe qm KL qm 1 + KL ce
qm (mg g-1) 284.90 303.95 317.46 108.11 123.61 134.95
KL (L mg−1) 0.3642 0.3730 0.4249 0.0202 0.0190 0.0268
RL 0.0032 0.0036 0.0037 0.0722 0.0993 0.0944
˜0.782 ˜0.8375 ˜0.8717 ˜0.6913 ˜0.7594 ˜0.7564
R2 0.9999 0.9999 0.9999 0.9989 0.9947 0.9963

Freundlich ln qe = ln KF +
1
ln ce
n
KF [(mg g−1)(L mg−1)1/n] 155.069 160.7548 166.1099 20.6689 18.3373 20.7198
1/n 0.11077 0.11544 0.11718 0.2702 0.30387 0.30002
R2 0.8002 0.8255 0.8399 0.9555 0.9798 0.9821

D-R 2 1
lnq e = lnqm KD-R RT ln 1+ ( ) 1
ce
E=
2KD R

qm (mg g−1) 269.02 283.71 294.25 84.13 88.27 97.52

KD-R (mol2 kJ−2) 0.1515 0.1121 0.0809 37.429 37.347 32.795
E (kJ mol−1) 1.8169 2.1118 2.4853 115.578 115.705 123.475
R2 0.8834 0.8718 0.8615 0.6402 0.5811 0.6103

Temkin qe =
RT
ln KT +
RT
ln ce
b b
KT (L mg−1) 1331.741 964.7855 936.6856 0.5346 0.3268 0.3557
b 111.2341 105.4763 104.2430 131.763 114.125 108.783
R2 0.8605 0.8923 0.9052 0.9775 0.9833 0.9903

6
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

Fig. 7. Adsorption kinetics of (a) MB and (b) DB dye molecules on OSAC-400-2.5 at different solution temperatures, [adsorbent dosage: 10.0 mg, solution volume:
50 mL, initial concentration: 200 mg L−1, initial pH: 7].

Table 3
Adsorption kinetic parameters of OSAC towards MB and DB sollution at different temperature.
Adsorbate MB DB

Temperature (°C) 25 35 45 25 35 45

qe.exp (mg g−1) 281.35 302.69 313.28 88.46 94.54 105.5

Pseudo-first-order ln(q e.cal qt )=lnq e.cal k1t
qe.cal (mg g−1) 14.806 15.044 14.939 30.117 31.846 34.708
k1 (min−1) 0.0811 0.1995 0.0946 0.0917 0.1130 0.1249
R2 0.8490 0.8986 0.8851 0.9461 0.9545 0.9585

Pseudo-second-order t
=
1
+
1
t
qt k2q 2e.cal q e.cal

qe.cal (mg g−1) 281.69 303.03 313.48 89.686 95.877 106.952

k2 (g mg−1 min−1) 0.0278 0.0478 0.0308 0.0110 0.0119 0.0118
R2 0.9999 0.9999 0.9999 0.9997 0.9998 0.9998

Elovich 1
qt = ln(1+ t)
α 5.9805 4.807 2.4749 3695.872 4727.738 10051.9
×1021 2 × 1028 ×1023
β 0.1847 0.2242 0.1773 0.1086 0.1026 0.0986
R2 0.9285 0.9019 0.8997 0.9643 0.9504 0.9438

Intra-particle diffusion qt = kip t1/2+C

kip1 17.6636 13.7039 19.2861 18.3858 19.8663 16.7862
C1 243.15 272.63 272.69 34.70 37.99 50.37
R12 0.9806 0.9991 0.9833 0.7386 0.8088 0.8201

kip2 5.8219 5.0962 5.3384 11.1880 12.9509 12.0319

C2 259.37 285.07 293.32 44.96 47.27 59.63
R22 0.9910 0.9877 0.9378 0.9548 0.9834 0.9363

kip3 0.7626 0.3226 0.7854 1.7376 1.3287 1.6773

C3 275.59 300.51 307.60 75.93 85.06 93.94
R32 0.9740 0.5851 0.8463 0.8655 0.8392 0.5190

Table 4
Thermodynamic parameters for adsorption of MB and DB onto OSAC.
Adsorbate Temperature △G0 △H0 △S0
(K) (kJ mol−1) (kJ mol−1) (J mol−1 K−1)

MB 298 −11.787 6.025 59.626

308 −12.244
318 −12.986

DB 298 −7.274 10.961 60.635

308 −7.361
318 −8.509

the adsorption process.

The kinetic analyses of MB and DB dyes adsorbed onto OSAC were
performed at the temperatures of 25, 35 and 45 °C, respectively. As
shown in Fig. 7, the adsorption amount of MB and DB increased with
the contact time. The adsorption rate of MB was fast in the first 10 min.
Fig. 8. Thermodynamics analysis for the adsorption of MB and DB on OSAC-
400-2.5. This may be due to the presence of available adsorption active sites on

7
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

molecules migrated into the interior regions of the adsorbents. And

then, they were adsorbed on the interior surfaces of the pores of the
adsorbents [18]. One or any combination of these three steps may be
rate-limiting [30]. As shown in Table 3, the intra-particle diffusion
model was also well fitted to the experimental data. It is worth noting
that all the C values of these fitting parameters are not equal to zero,
indicating the intra-particle diffusion was not the unique rate-limiting
step during these adsorption processes [25]. As shown in Figs. S3 and
S4, the adsorption process of dyes could be divided into three steps,
which are rapid initial stage controlled by the external boundary layer
diffusion, the period of the slower uptake represented by the intra-
particle diffusion, and the equilibrium stage caused by the saturation of
adsorption active sites on the interior surface, respectively.

3.6. Thermodynamic analysis

Fig. 9. Reutilization performance of OSAC during adsorption-desorption cycles, In order to further understand the adsorption behavior, the ad-
[adsorbent dosage: 10.0 mg, solution volume: 50 mL, initial concentration: sorption thermodynamic analyses were conducted. The △H0 and △S0
100 mg L−1, initial pH: 7, temperature: 25 °C]. can be calculated [31] from the slope and the intercept of the plot of ln
K0 vs. 1/T (as shown in Fig. 8). And the adsorption thermodynamic
parameters are listed in Table 4. The negative values of △G0 for both of
two dyes at different temperatures suggest their adsorption process
were spontaneous. Moreover, the higher negative value reflects more
energetically favorable adsorption. It means that the higher tempera-
ture helps to promote dye adsorption, being consistent with the pre-
vious experimental results. The positive values of △S0 imply that the
randomness of the adsorption systems were increasing, which made the
system more stable. In addition, the positive values of △H0 were 6.025
and 10.961 for MB and DB, respectively. It confirms that the adsorption
processes are endothermic, which may be due to the higher tempera-
ture reduced the amounts of water molecule adsorbed on the adsorbent
surface by hydrogen bonding, and more adsorption sites for dye mo-
lecules are exposed [32].

3.7. Reutilization performance

Reutilization capacity is commonly an important property to eval-

uate the practicability of an adsorbents [33]. Firstly, the fresh OSAC
adsorbents were dosed into two types of dyes solutions to reach equi-
librium, respectively. Then the fully-adsorbed OSAC powders were
rinsed by 0.1 M NaOH and deionized water repeatedly after being se-
parated. Subsequently, the recovered OSAC adsorbents were subjected
to the same adsorption process and repeated five times in succession. It
is evident from Fig. 9 that the adsorption capacity qe decreased gra-
Fig. 10. Proposed mechanism for MB and DB adsorption onto OSAC.
dually with the increase of cycles. On the one hand, it is due to the
porous structure of the adsorbent makes the internal adsorption sites
the surface of OSAC in the beginning. And as the adsorption process could not be penetrated quickly by the eluent, on the other hand, it is
continues, the active sites were gradually occupied by adsorbates. Then, inherently difficult to desorb the molecules that bind to the OSAC by
the adsorption rate was gradually slowed down, thus reach equilibrium forming chemical bonds. It is worth noting that qe for MB relatively
in 1 h. The DB adsorption probably shows the same trend, but takes a decreased by 13.3%, while qe for DB decreased 9.2%. It indicates that
little longer to reach equilibrium. OSAC has a stronger reutilization ability when adsorbing DB than MB.
The adsorption kinetic experimental data were analyzed by using This is mainly caused by the difference in the ways of the adsorption of
Pseudo-first order model, Pseudo-second order model, Elovich model, two dyes on OSAC. Chemical adsorption effect plays a more important
and Intra-particle diffusion model, which illustrated in Figs. S3 and S4. role on the MB adsorption process, while DB adsorption is dominated
And as presented in Table 3, the Pseudo-second order model fitted the by physical adsorption. Moreover, SEM images of OSAC adsorbents
experimental data better with the higher R2 values more than 0.999. with MB and DB adsorbed which has been used by six times are shown
Furthermore, the value of the equilibrium adsorption capacity (qe.cal) in Fig. S5. The pore structure of the OSAC was not as obvious as the
calculated from Pseudo-second order model was more agreeable to original state. In general, our OSAC exhibits excellent reutilization
those determined experimentally (qe.exp). The Pseudo-second order capability, especially as a waste-based adsorbent material, which gives
model implies chemical adsorption as an important adsorption me- it a higher economic and practical value. However, realizing industrial
chanism [29] and involves the chemical interaction between MB/DB production to improve the production rate and reduce the production
molecules and the polar functional group at the surface of OSAC, which cost is worth to further study.
characterized by FTIR above.
Generally, several adsorption steps involved in adsorption dy- 3.8. Adsorption mechanism
namics. Firstly, the adsorbates molecules in the solution moved through
the boundary layer to the surface of the adsorbents. Secondly, the It is extremely difficult to determine the adsorption mechanisms,

8
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

because adsorption itself is a complex interface reaction process con- [5] Z. Gong, Z. Wang, Z. Wang, Study on migration characteristics of heavy metals
trolled by various factors simultaneously. According to the analyses during oil sludge incineration, Pet. Sci. Technol. (2018) 1–6.
[6] R.A. Tahhan, R.Y. Abu-Ateih, Biodegradation of petroleum industry oily-sludge
above, the adsorption process of both MB and DB onto OSAC were all using Jordanian oil refinery contaminated soil, Int. Biodeterior. Biodegrad. 63
controlled by the effects combined physical adsorption and chemical (2009) 1054–1060.
adsorption, but somewhat different from each other. First and foremost, [7] G. Hu, J. Li, G. Zeng, Recent development in the treatment of oily sludge from
petroleum industry: a review, J. Hazard. Mater. 261 (2013) 470–490.
the OSAC’s porous structure would be important for its enhanced ad- [8] X. Li, Z. Wang, L. Guo, D. Han, B. Li, Z. Gong, Manganese oxide/hierarchical porous
sorption properties, as it provides efficient molecules motion pathway carbon nanocomposite from oily sludge for high-performance asymmetric super-
and sufficient active adsorption sites. However, the adsorption equili- capacitors, Electrochim. Acta 265 (2018) 71–77.
[9] V. Mymrine, M.J.J.S. Ponte, H.A. Ponte, N.M.S. Kaminari, U. Pawlowsky,
brium capacity of MB is much higher than DB. Except for the smaller G.J.P. Solyon, Oily diatomite and galvanic wastes as raw materials for red ceramics
molecular made MB is much easier to access the internal of OSAC, their fabrication, Constr. Build. Mater. 41 (2013) 360–364.
own unique adsorption mechanism was also important reasons. As [10] D. Eliche-Quesada, R. Azevedo-Da Cunha, F.A. Corpas-Iglesias, Effect of sludge from
oil refining industry or sludge from pomace oil extraction industry addition to clay
shown in Fig. 10, OSAC is typically negatively charged in solution due
ceramics, Appl. Clay Sci. 114 (2015) 202–211.
to the presence of oxygen-containing functional group. Thus, it prefers [11] T.J. Bandosz, K. Block, Municipal sludge-industrial sludge composite desulfuriza-
MB which is positively charged through electrostatic interactions than tion adsorbents: synergy enhancing the catalytic properties, Environ. Sci. Technol.
DB which is negatively charged. It suggests that the effect of chemical 40 (2006) 3378–3383.
[12] S. Mohammadi, N. Mirghaffari, Optimization and comparison of Cd removal from
adsorption in MB adsorption process is greater than DB adsorption aqueous solutions using activated and non-activated carbonaceous adsorbents
process. In addition, the interaction between the adsorbent and MB or prepared by pyrolysis of oily sludge, Water Air Soil Pollut. 226 (2015).
DB molecules could also through π-π stacking and H-bonding interac- [13] H. Guo, H. Lin, Y. Li, X.L. Li, S.P. Feng, Y. Zhu, Adsorbents synthesized from de-
watered oily sludge and their adsorption of reactive brilliant red X-3B in aqueous
tions [34–38]. Among them, H-bonding may also root in the surface solution, Desalin. Water Treat. 57 (2016) 25267–25277.
functional group of OSAC, and π-π stacking mechanism occurs between [14] L.L. Ding, B. Zou, W. Gao, Q. Liu, Z.C. Wang, Y.P. Guo, X.F. Wang, Y.H. Liu,
the aromatic rings of the dye molecules and π-electron system of OSAC Adsorption of rhodamine-B from aqueous solution using treated rice husk-based
activated carbon, Colloids Surf. A Physicochem. Eng. Asp. 446 (2014) 1–7.
structure. [15] K.F. Fu, Q.Y. Yue, B.Y. Gao, Y. Wang, Q. Li, Activated carbon from tomato stem by
chemical activation with FeCl2, Colloids Surf. A Physicochem. Eng. Asp. 529 (2017)
4. Conclusion 842–849.
[16] X. Li, K. Liu, Z. Liu, Z. Wang, B. Li, D. Zhang, Hierarchical porous carbon from
hazardous waste oily sludge for all-solid-state flexible supercapacitor, Electrochim.
In this work, we proposed a facile way to fabricate porous activated Acta 240 (2017) 43–52.
carbon from oil sludge waste. The obtained OSAC possess porous [17] P. Tamilarasan, S. Ramaprabhu, Graphene based all-solid-state supercapacitors
with ionic liquid incorporated polyacrylonitrile electrolyte, Energy 51 (2013)
carbon structure with high specific surface area (318.3 m2 g−1) and rich
374–381.
oxygen functional group. Due to these properties, it exhibits out- [18] L.J. Leng, X.Z. Yuan, G.M. Zeng, J.G. Shao, X.H. Chen, Z.B. Wu, H. Wang, X. Peng,
standing adsorption capabilities of 316.02 and 124.24 mg g−1 for MB Surface characterization of rice husk bio-char produced by liquefaction and appli-
and DB, respectively. The adsorption isotherms of both two dyes were cation for cationic dye (Malachite green) adsorption, Fuel 155 (2015) 77–85.
[19] R.S. Hastak, P. Sivaraman, D.D. Potphode, K. Shashidhara, A.B. Samui, All solid
fitted well by Langmuir model, and adsorption kinetic was good de- supercapacitor based on activated carbon and poly [2,5-benzimidazole] for high
scribed by pseudo-second order equation. Moreover, the thermo- temperature application, Electrochim. Acta 59 (2012) 296–303.
dynamic analyses implied that their adsorption processes are all en- [20] Y. Song, J. Yang, K. Wang, S. Haller, Y. Wang, C. Wang, Y. Xia, In-situ synthesis of
graphene/nitrogen-doped ordered mesoporous carbon nanosheet for supercapacitor
dothermic and spontaneous. In addition, solution pH appreciably application, Carbon 96 (2016) 955–964.
affected the adsorption capacities of OSAC. With the increasing of pH [21] S.P.D. Monte Blanco, F.B. Scheufele, A.N. Módenes, F.R. Espinoza-Quiñones,
values, the adsorption capacity of MB raised while that of DB decreased. P. Marin, A.D. Kroumov, C.E. Borba, Kinetic, equilibrium and thermodynamic
phenomenological modeling of reactive dye adsorption onto polymeric adsorbent,
Based on the experimental results, it can be concluded that the ad- Chem. Eng. J. 307 (2017) 466–475.
sorption mechanism of OSAC towards dye molecules includes physical [22] A.L. Cazetta, O. Pezoti, K.C. Bedin, T.L. Silva, A. Paesano Junior, T. Asefa,
and chemical adsorption. The electrostatic interactions, π-π stacking V.C. Almeida, Magnetic activated carbon derived from biomass waste by concurrent
synthesis: efficient adsorbent for toxic dyes, ACS Sustain. Chem. Eng. 4 (2016)
and H-bonding interactions could play certain roles on the adsorption
1058–1068.
process of both dyes. Moreover, the OSAC showed an excellent re- [23] X. Hu, Z.H. Ding, A.R. Zimmerman, S.S. Wang, B. Gao, Batch and column sorption
utilization performance. This study explores new prospects for high- of arsenic onto iron-impregnated biochar synthesized through hydrolysis, Water
Res. 68 (2015) 206–216.
value-added recycling of oil sludge and other industrial wastes.
[24] U.R. Lakshmi, V.C. Srivastava, I.D. Mall, D.H. Lataye, Rice husk ash as an effective
adsorbent: evaluation of adsorptive characteristics for Indigo Carmine dye, J.
Acknowledgements Environ. Manage. 90 (2009) 710–720.
[25] A.T. Xie, J.D. Dai, X. Chen, P. Ma, J.S. He, C.X. Li, Z.P. Zhou, Y.S. Yan, Ultrahigh
adsorption of typical antibiotics onto novel hierarchical porous carbons derived
This work was supported by the Fundamental Research Funds for from renewable lignin via halloysite nanotubes-template and in-situ activation,
the Central Universities (No. 17CX06031) and the Natural Science Chem. Eng. J. 304 (2016) 609–620.
Foundation of Shandong Province (No. ZR2017BEE042). [26] D. Angin, Utilization of activated carbon produced from fruit juice industry solid
waste for the adsorption of Yellow 18 from aqueous solutions, Bioresour. Technol.
168 (2014) 259–266.
Appendix A. Supplementary data [27] K. Ada, A. Ergene, S. Tan, E. Yalcin, Adsorption of Remazol Brilliant Blue R using
ZnO fine powder: equilibrium, kinetic and thermodynamic modeling studies, J.
Hazard. Mater. 165 (2009) 637–644.
Supplementary material related to this article can be found, in the [28] L. Wang, Application of activated carbon derived from ‘waste’ bamboo culms for the
online version, at doi:https://doi.org/10.1016/j.colsurfa.2019.05.066. adsorption of azo disperse dye: kinetic, equilibrium and thermodynamic studies, J.
Environ. Manage. 102 (2012) 79–87.
[29] J.T. Petrovic, M.D. Stojanovic, J.V. Milojkovic, M.S. Petrovic, T.D. Sostaric,
References
M.D. Lausevic, M.L. Mihajlovic, Alkali modified hydrochar of grape pomace as a
perspective adsorbent of Pb2+ from aqueous solution, J. Environ. Manage. 182
[1] P. Hadi, M. Xu, C. Ning, C.S.K. Lin, G. McKay, A critical review on preparation, (2016) 292–300.
characterization and utilization of sludge-derived activated carbons for wastewater [30] A. Salima, B. Benaouda, B. Noureddine, L. Duclaux, Application of Ulva lactuca and
treatment, Chem. Eng. J. 260 (2015) 895–906. Systoceira stricta algae-based activated carbons to hazardous cationic dyes removal
[2] J. Wang, J. Yin, L. Ge, J. Shao, J. Zheng, Characterization of oil sludges from two oil from industrial effluents, Water Res. 47 (2013) 3375–3388.
fields in China, Energy Fuels 24 (2010) 973–978. [31] Y. Liu, Is the free energy change of adsorption correctly calculated? J. Chem. Eng.
[3] J. Liu, X. Jiang, X. Han, Devolatilization of oil sludge in a lab-scale bubbling flui- Data 54 (2009) 1981–1985.
dized bed, J. Hazard. Mater. 185 (2011) 1205–1213. [32] H. Laksaci, A. Khelifi, M. Trari, A. Addoun, Synthesis and characterization of mi-
[4] X. Wang, Q. Wang, S. Wang, F. Li, G. Guo, Effect of biostimulation on community croporous activated carbon from coffee grounds using potassium hydroxides, J.
level physiological profiles of microorganisms in field-scale biopiles composed of Clean. Prod. 147 (2017) 254–262.
aged oil sludge, Bioresour. Technol. 111 (2012) 308–315. [33] X.-y. Li, D. Han, J.-f. Xie, Z.-b. Wang, Z.-q. Gong, B. Li, Hierarchical porous

9
X. Li, et al. Colloids and Surfaces A 578 (2019) 123505

activated biochar derived from marine macroalgae wastes (Enteromorpha pro- mesh size of porous activated carbon prepared from Pinus eldarica and its effects on
lifera): facile synthesis and its application on Methylene Blue removal, RSC Adv. 8 simultaneous removal of dyes: multivariate optimization, Chem. Eng. Res. Des. 125
(2018) 29237–29247. (2017) 408–421.
[34] S.A. Hosseinpour, G. Karimipour, M. Ghaedi, K. Dashtian, Use of metal composite [37] R. Karimi, F. Yousefi, M. Ghaedi, K. Dashtian, Back propagation artificial neural
MOF-5-Ag2O-NPs as an adsorbent for the removal of Auramine O dye under ul- network and central composite design modeling of operational parameter impact
trasound energy conditions, Appl. Organomet. Chem. 32 (2018) e4007. for sunset yellow and azur (II) adsorption onto MWCNT and MWCNT-Pd-NPs:
[35] E. Sharifpour, M. Ghaedi, F. Nasiri Azad, K. Dashtian, H. Hadadi, M.K. Purkait, Zinc isotherm and kinetic study, Chemom. Intell. Lab. Syst. 159 (2016) 127–137.
oxide nanorod-loaded activated carbon for ultrasound-assisted adsorption of sa- [38] S. Kheirandish, M. Ghaedi, K. Dashtian, F. Heidari, F. Pourebrahim, S. Wang,
franin O: central composite design and genetic algorithm optimization, Appl. Chitosan extraction from lobster shells and its grafted with functionalized MWCNT
Organomet. Chem. 32 (2018) e4099. for simultaneous removal of Pb2+ ions and eriochrome cyanine R dye after their
[36] M. Jafari, M.R. Rahimi, M. Ghaedi, K. Dashtian, ZnO nanoparticles loaded different complexation, Int. J. Biol. Macromol. 102 (2017) 181–191.

10