Bioresource Technology Reports 14 (2021) 100664

Contents lists available at ScienceDirect

Bioresource Technology Reports

journal homepage: www.sciencedirect.com/journal/bioresource-technology-reports

Sugarcane bagasse based activated carbon preparation and its adsorption

efficacy on removal of BOD and COD from textile effluents: RSM based
modeling, optimization and kinetic aspects
Surafel Mustefa Beyan a, *, S. Venkatesa Prabhu b, T. Tsegaye Sissay b, A. Abraham Getahun c
a
School of Chemical Engineering, Jimma Institute of Technology, Jimma University, Jimma, Oromia, Ethiopia
b
Department of Chemical Engineering, College of Biological and Chemical Engineering, Addis Ababa Science and Technology University, Addis Ababa, Ethiopia
c
Central Laboratory, Research and Technology Transfer, Addis Ababa Science and Technology University, Addis Ababa, Ethiopia

A R T I C L E I N F O A B S T R A C T

Keywords: This study focused on the removal of biological oxygen demand (BOD) and chemical oxygen demand (COD) from
Textile wastewater textile waste effluents using activated carbon prepared from sugarcane bagasse (SBAC). Various techniques
Treatment including XRD, SEM, BET and FTIR were done to characterize the SBAC. The three independent process vari­
Sugar cane bagasse
ables, SBAC dose, pH and reaction time were selected to develop mathematical models for BOD and COD removal
Activated carbon
RSM
efficiency using response surface methodology (RSM). The predicted models were statistically optimized to
maximize BOD and COD removal. The optimal values were predicted to be a SBAC dose of 0.915 g/L, pH of
3.225, and a reaction time of 98.91 min for maximized BOD and COD removal efficiencies of 0.0225 and 0.023
mg/L min− 1, respectively. Using these optimal values, kinetics studies from the experimental data showed that
the rate constants for BOD and COD removal were 0.028 and 0.0274 mg/L min− 1, respectively.

1. Introduction biological methods. Trapping contaminants to avoid their release and

prevent pollution of uncontaminated areas can be done through various
Recently, the textile industries have gained a remarkable economic physical methods such as encapsulation, filtration, gravity settling,
significance globally. This industry includes diverse bodies that involve adsorption, and stabilization (Katheresan et al., 2018). Wastewater
various manufacturing elements with different kinds of fibers for generated in the course of processing in the fabric industry has excessive
manufacturing a wide variety of fabrics with attractive properties. In the biological oxygen demand (BOD) and chemical oxygen demand (COD)
textile industries, globally more than 8000 chemical compounds are (Oyebamiji et al., 2019; Patel and Vashi, 2015). Improper treatment of
used for preparing 400 billion m2 of fabric each year. During textile textile wastewater is one cause of environmental issues that may
manufacturing processes, various chemical treatments such as sizing, aggravate human health problems. Hence, effective treatment of the
scoring, mercerizing, coloring, printing, and finishing are carried out effluents produced by textile industries is a global issue.
(Alkaya and Demirer, 2014). Textile industries utilize various chemical Adsorption is an inexpensive, novel, efficient and easy methodology
compounds to control the properties of auxiliary textiles such as fabric for treating textile wastewater to remove pollutants. Activated carbon is
texture, serviceability and durability. During the various processes in one of the most used adsorbents. It has an appropriate pore size distri­
the textile industries, significant quantities of toxic wastewater are bution, large specific surface area and high surface reactivity. However,
generated that require complex treatments before being released back it is relatively costly and its regeneration is difficult. This has led to
into the environment (Isik et al., 2019; Lellis et al., 2019). Chemical exploration of new low cost and efficient materials to adsorb pollutants
pollutants from textile wastewaters pose serious safety risks to people in recent years. It has been found that activated carbon produced from
and the environment. Various approaches have been proposed for con­ various industrial and agricultural wastes shows high potential to
trolling textile wet processes to better protect the environment and remove pollutants. Several agricultural by-products such as rice husks,
ecosystems. Modern management for the treatment of wastewater corn cobs, banana peels, flower wastes, and sugarcane bagasse have
effluent from the textile industries can entail physical, chemical, and/or been used for preparation of activated carbon. In this regard, low-cost

* Corresponding author.
E-mail address: [email protected] (S.M. Beyan).

https://doi.org/10.1016/j.biteb.2021.100664
Received 4 November 2020; Received in revised form 13 February 2021; Accepted 14 February 2021
Available online 25 February 2021
2589-014X/© 2021 Elsevier Ltd. All rights reserved.
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

materials that are highly efficient and can be used with simple tech­ organic/inorganic pollutants, no investigations have been made for
nology are necessary to mitigate pollutants from textile effluents. mathematical modeling and statistical optimization of COD and BOD
Recently, researchers have turned their attention to better, easily removal from textile industry effluents using SBAC. Hence, this study
available yet less expensive materials to produced activated carbon was focused on development of a model and its optimization for
(Danish et al., 2018). Sugarcane bagasse is certainly an important agro- removing COD and BOD from textile industry wastewater using SBAC as
industrial residue with a high cellulose content (Mpatani et al., 2020; an adsorbent. Furthermore, the study was aimed for statistical optimi­
Silveira et al., 2018). It mainly consists of cellulose (45%), hemi­ zation of the process parameters, SBAC dose, reaction pH and reaction
celluloses (28%), and lignin (18%) (Nikodinovic-Runic et al., 2013). time. Additionally, experimental data was subjected to kinetics analysis
These materials contain carboxylic and hydroxyl groups with the po­ for determination of removal rates for COD and BOD under optimal
tential to adsorb dye molecules by means of ion exchange and/or conditions.
complexation (Ge et al., 2017). Hence, they are attractive as they are
inexpensive and effective adsorbents. Several researches have been 2. Materials and methods
carryout using sugarcane bagasse derived activated carbon. Cueva-
Orjuela et al. (2017) investigated about the potential use of SBAC for 2.1. Materials
treatment of textile effluent. They examine to optimize the adsorption
process through a response surface design to remove the dyes, basic red Analytical grade reagents and chemicals were used in this study.
46 and BR46. The effect of factors such as point of zero charge, solution They were obtained from Sigma-Aldrich. A pH meter (Model 3520) was
pH, particle size, adsorbent dosage, initial dye concentration, contact used for pH measurements. The wastewater effluent sample was
time and ionic strength was evaluated using a statistical design of ex­ collected from Ayka Addis Textile Industry, Addis Ababa, Ethiopia. It
periments that allowed to achieve a maximum adsorption of 95.0% was taken directly from the effluent outlet of a static screen and placed
(Cueva-Orjuela et al., 2017). Ajinomoh and Salahudeen (2014) carried in clean 5 L polyethylene bottles. The samples were preserved at 4 ◦ C in a
out a study on synthesis of activated carbon from sugarcane bagasse refrigerator prior to further experimentation.
using zinc chloride as activating agent with different ratios. From these
results, the ratio of 1:1 zinc chloride to sugarcane bagasse activation 2.2. Activated carbon preparation from sugarcane bagasse
showed a good activated carbon with the highest adsorption capacity of
6.67 mg/g for phenol. Using phosphoric acid activation, a mechano­ Sugarcane bagasse (SCB) was collected from Didessa Sugar Factory,
chemical processing was carried out to prepare SCBAC by Cui et al. They Oromia, Ethiopia. It was washed thoroughly using distilled water and
found that the prepared SCBAC by mechanochemical processing acti­ dried in a laboratory oven at 105 ◦ C for 24 h. Furthermore, it was pul­
vated by phosphoric acid exhibited less processing time, greater adsor­ verized using an electric ball-mill followed by sieve analysis to obtain a
bed volume of iodine, and higher adsorption capacity for methylene uniform particle size of 250 μm. The material was subjected to a
blue (Chen et al., 2012). A study with respect to physisorption and chemical activation method in which a mixture a 1:3 (w/w) ratio of SCB
chemisorption of methylene blue and its kinetics were undertaken by particles to concentrated sulfuric acid was used. The acid acted as a
Chaiwon et al. (2017) using H2SO4 as activating agent. They observed dehydrating agent restrain the tar structure, thus producing a refined
that the kinetics of methylene blue adsorption followed the pseudo- porosity in the resulting activated carbon. Activated carbon was pre­
second-order model and suggested that the rate-limiting step the pared using the procedures of Dao and Le Luu (2020), Saad et al. (2010)
adsorption process by chemical sorption (Chaiwon et al., 2017). An and Tahir et al. (2016) with some modifications. Chemically treated SCB
Investigation on the removal of phenol from aqueous solutions was done particles were dried in an oven for 4 h at 115 ◦ C. Then, they were heated
by Akl (2014). In this study, sugarcane bagasse residues modified within a muffle furnace for 2 h at 450 ◦ C. The cinder material collected
physically by steam and chemically using NaOH were used. They found from the furnace was soaked in a 1% of sodium bicarbonate solution for
that the modification of SBAC by steam and NaOH significantly 4 h. It was cleaned once with distilled water followed by a 1% sodium
increased the oxygen-containing groups on the surface of activated bicarbonate solution for 24 h to remove residual acid. Then, chemically
carbons. Kaushik et al. (2017) found that the high molecular weight treated SCB particles were cleaned with distilled water to remove the
heterogeneous polymers, melanoidins, were effectively adsorbed by remaining bicarbonates. Chemically and thermally activated SCB par­
SCBAC followed by desorption using 25% pyridine solution. Mohamed ticles were then dried overnight in an oven at 120 ◦ C. The dried particles
et al. (2015) were carried out on the removal of formaldehyde gas using were pulverized and sieved. Particles of sizes ranging between 250 and
SCBAC. The results showed that the formaldehyde gas removal rate was 300 μm were selected for use as activated carbon (SBAC). This was
more than 67% in the plant treated chamber. Recently, Ahmad et al. selected as an adsorbent and used in further experiments.
(2020) have examined on the adsorption of 1-[2-(2-chloroethoxy)
phenyl] sulfonyl-3-(4-methoxy-6-methyl-1,3,5-triazin-2-yl) urea (tri­ 2.3. Characterization of SBAC
asulfuron) using SCBAC. They suggested that SCBAC-mediated elimi­
nation of triasulfuron is beneficial for both economically and Activated carbon prepared from sugarcane bagasse was character­
environmentally. Around 50% triasulfuron was easily detached from ized using scanning electron microscopy (SEM), Brunauer–Emmett–­
soils by this methodology. Teller (BET) analysis, Fourier transform infrared spectroscopy (FTIR)
In this study, sugarcane bagasse was used to prepare activated car­ and X-ray diffraction (XRD). The Brunauer–Emmett–Teller (BET SA-
bon to treat a textile wastewater. Prior to conducting the investigation, 9600, USA) surface area was observed from melting point analysis of
the textile wastewater was characterized to determine its primary nitrogen gas mono-layer sorption on the surface of SBAC at 77 K. Helium
physio-chemical properties. Response surface methodology (RSM) is a gas (at 2 bars of pressure) was used as a carrier for the nitrogen gas (2.5
proven and effective tool for sequential analysis to study process bar) and to purge air from the environment. To determine the specific
modeling and optimization. It gives sufficient information to test the surface area of the samples, a mixture of helium and nitrogen gases (20%
interaction effects among the process variables with a reasonable and 30%, respectively) was used. The SBAC was tempered at a tem­
number of experiments (Kidus Tekleab et al., 2020; Venkatesa et al., perature of 120 ◦ C for 1 h. Pretreated 0.1 g samples were subjected to
2019). Additionally, it uses mathematical and analytical means to BET studies. The adsorbent samples of SBAC, SEM analysis was carried
identify interrelationships between numerical input variables to obtain out (FEI, INSPCT-F50, Germany) to elucidate the surface morphology.
their optimum values (Manohari et al., 2017; Tetteh and Rathilal, 2017). SBAC samples were placed on aluminum stubs, covered with a
Even though several studies have been conducted using activated conductive carbon adhesive tape, and visualized using SEM operating at
carbon derived from sugarcane bagasse (SBAC) to remove various 15 kV with a 40 μm working distance in a vacuum. To determine the

2
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

Table 1 Furthermore, the parameters were optimized to maximize the

Levels of the three independent variables studied. response using numerical optimization. ANOVA (analysis of variance)
Variables Factor Unit 1 (high -1 (low − α +α was employed to evaluate the statistical constraints. The optimal values
coding level) level) of the chosen variables were attained by solving a regression equation.
SBAC A g/L 1.1 0.5 0.295 1.304 Interaction effects among the factors were analyzed by developing
dosage response surfaces and contour plots. CCD-RSM analyses were performed
pH B – 7 3 1.64 8.4 using Design-Expert 12.0 software. To validate the developed model,
Contact C min 100 60 46.36 113.6 prediction using the point prediction feature of Design-Expert 12 soft­
time
ware, experimental runs was done in triplicate.

presence of functional groups in the SBAC, the sample was subjected to 2.6. Kinetics approach for BOD and COD removal
FTIR (iS50 ABX, Germany) within the spectral region range of 4000
cm− 1 to 400 cm− 1. These wavelengths were generated using an IR light Since these processes involve reaction rate controlling mechanisms,
source and KBr pellet method, adapted to do FTIR analysis. Additionally, a pseudo-first order kinetics model (Eq. (1)) was employed to assess the
XRD analysis was carried out on SBAC to investigate the amorphous and rate of removal of BOD/COD by evaluating rate constant, kp (Kumar
crystalline nature of the acquired product. XRD investigations were done et al., 2020; Ma et al., 2011). Adsorption studies were carried out using
using an Olympus BTX-528 XRD by Cu–Co with a radiation source the optimal conditions determined from RSM analysis. The BOD and
working at a voltage of 40 kV and a 25 mA current using diffraction COD removals were calculated at various times until reaching an
angles (2θ) of 5o to 55o. The split threshold value was kept as 4. optimal time (Sundramurthy et al., 2020; Venkatesa Prabhu and Baskar,
2015) as:
2.4. Physiochemical characterization of textile effluents
dCp ( )
= kp Cpi − Cpt (2)
dt
The pH of a collected raw textile effluent was determined on-site.
Furthermore, turbidity, total solids, electrical conductance, total sus­ (
Cpi
)
pended solids (TSS), total dissolved solids (TDS), BOD, and COD were ln = kp t (3)
Cpi − Cpt
analyzed before experimentation according to the methods prescribed
by APHA (Franson, 1998). Color measurements were made according to
where, ( dt p ) is the incremental rate of BOD/COD removal, kp is the rate
dC
the Platinum-Cobalt unit.
constant of BOD/COD removal, CPi is the total BOD/COD content in the
Area of Crystalline peak raw wastewater and Cpt is BOD/COD content in a treated solution at
Crystallinity (%) = × 100 (1)
Area of Peak (Crystalline + Amorophours) time t. By integrating Eq. (2) between the initial time (t = 0 min, Cpt = 0)
and the conditions corresponding to a time t (t = t m, Cpt = Cpt), a linear
2.5. Batch adsorption experiments for BOD and COD removal using RSM form can be obtained as Eq. (3). Using this equation, the rate constant
( )
value (kp) was determined from the slope of a plot of ln Cpi − piCpt vs time
C

Prior to the design of experiments, prelimenary adsorption experi­

ments were carried out using SCB and SBAC to determine the feasibility and was reported in units of mg/L min− 1. The linear fit of the data was
of COD and BOD removal at pH 4, with a contact time of 60 min and a evaluated based the plot’s correlation coefficient (R) value.
dose of 1 g/L. From this study, it was observed that the removal of BOD
and COD by SBAC adsorbent was 80% greater than for SCB. Based on 3. Result and discussion
this result, SBAC was chosen as the adsorbent for optimization studies.
The study focused on optimization and investigation of the interac­ 3.1. Characterization of SBAC and SCB
tion effects of SBAC dose, solution pH and contact time on the COD and
BOD removal efficiency from the textile effluent. Agitation speed was The specific surface area of SBAC was found to be 591 m2/g using
450 rpm at a constant room temperature. For RSM analysis, the com­ BET analysis. Chemical activation using sulfuric acid created porous
bination of the selected parameters was determined using central com­ area on the SBAC. However, the observed BET surface area of SBAC is
posite design (CCD). This procedure ensures that all factors and their comparatively smaller than that of commercially available activated
interactions are systematically explored. The chosen parameters and carbon (Mohamed et al., 2015). The surface morphology observed from
their respective levels are given in Table 1. Based on the number of SEM imagery for SBAC. From the SEM image, it can be seen that SBAC
parameters, CCD indicated 20 combination experiments that account for subjected to sulfuric acid activation had significant effects. The size and
factorial, axial and center points. Hence, all the experimental runs were shape of the pores present in the SBAC are artificially created with
carried out in triplicate and the mean value of the response (percent different physio-chemical properties than found in the raw material.
removal of BOD and COD) was reported. Additionally, this material could be affected by an oxidizing agent,
The experimental observations obtained from the CCD were statis­ chemical treatment and the impregnation ratio (Kim, 2004).
tically analyzed and fitted to a second order polynomial model to FTIR spectrum analysis clearly shows the functional groups SBAC.
develop an empirical mathematical model to indicate the relationship From the analysis, it can be understood that the sorption mechanism and
between the response and independent factors. The second order poly­ adsorption/desorption behaviors were dominated by the functional
nomial equation takes the form: groups present in the adsorbent. The FTIR spectrum showed that the
SBAC had a large number of peaks. This indicates the complex nature of
∑ ∑ ∑∑ SBAC. There were two strong bands at around 1000 cm− 1 and 550 cm− 1.
k k k
R = β0 + βj X j + βjj X2j + βji Xi Xj + ei
j=1 j=1 i <j=2 These typical bands indicate the presence of inorganic phosphates.
Strong peaks in the range of 1085–1024 cm− 1 indicate the presence of
where, R is the response [BOD/COD removal (%)]. Xi and Xj refer var­ C–O stretching vibrations in primary alcohol, secondary alcohol and
iables (i and j ranging from 1 to k). β0 refers the intercept coefficient. βj, aliphatic ether (Mistar et al., 2020). The peaks in the range of
βjj and βij are known interaction coefficients for the respective terms, k is 1070–1030 cm− 1 and 1050–1040 cm− 1 are observed. These peaks
the number of chosen independent variables (k = 3 in this study), and ei clearly represent strong S– – O stretching corresponding to sulfoxide
is an error term. compounds and strong CO–O–CO stretching related to anhydride

3
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

Table 2 absorption found at 3300 cm− 1. These distinct structures and functional
The physicochemical characteristics of collected raw wastewater from groups would have a promising effect on the adsorption of organic
Ayka Addis textile industry. molecules through hydrogen bonding, chelation and electrostatic
Parameters Textile effluent values adsorption.
pH 8.9
The SBAC sample was examined using X-ray diffraction (XRD). The
Turbidity 135 NTU crystal structure of SBAC at 2θ (o) ~ 26.689 and d-spacing ~3.3375 was
Color (Pt–Co) 1140 an orthorhombic Haklaite with lattice parameters of a = 9.3246 Å, b =
Suspended solid 198 mg/L 5.4992 Å and c = 9.7892 Å. These values exactly matched the current
Total dissolved solid 11,540 mg/L
AMCSD (American Mineralogist Crystal Structure Database) card (set 2
Chemical oxygen demand 960 mg/L
Biological oxygen demand 490 mg/L ~ file ~7194 and subfile ~ mineral). It had a plane of (213). A phase of
Sitinakite with a tetragonal arrangement was found at 2θ (o) ~ 11.279
and d-spacing ~7.8389 with lattice parameters a = b = 7.819 Å and c =
compounds. A peak at 1249.67 cm− 1 indicate the presence of medium 12.099 Å. It had a space group of P2/mcm and a (312) plane. Further­
C–N stretching in amines. Alternatively, there were no peaks found in more, at 2θ (o) ~ 32.356 and d-spacing ~2.764, a BGaO4Sr phase with
SCB. The small peak in the spectra at 1520 cm− 1 was due to aromatic an orthorhombic crystal system within the space group of P2_12_12
skeletal vibration (Kan et al., 2017; Kumar and Jena, 2016). Small bands existed. Peaks for Pyrargyrite and Gillespite were located at 2θ (o) ~
at 1619 cm− 1 and 1729 cm− 1 are observed that correspond to strong 32.123 and 26.495, respectively. A broad peak was observed showing
stretching of C–– O bonds. This can be attributed to α and β-unsaturated the presence of an amphoras structure. Additionally, the percent crys­
ketones and strong carboxylic acid. The broad peak at 3337 cm− 1 was tallinity was calculated using Eq. (1) for SBAC and found to be 52.89%.
found to be due to the O–H stretching vibrations of hydroxyl groups.
The band at 2900 cm− 1 can be attributed to a medium C–H vibrational
3.2. Physiochemical characteristics of textile effluents
stretching in alkenes. Also, the presence of mono-substituted acetylenes
was identified by the strong –C=C–H (carbon hydrogen) stretching
Physicochemical characterization results on the textile wastewater

Table 3
CCD matrix of adsorption experiments with factor setup, actual and predicted response values.
Run Factor 1 Factor 2 Factor 3 Response (% removal)

A: BAC dosage B: pH C: Contact time (min) COD BOD

(g/L)
Actual Predicted Actual Predicted

1 0.3 5 80 50.98 52.40 53.14 53.37

2 0.8 5 46 45.46 47.27 43.76 44.21
3 1.1 7 60 52.36 54.53 55.17 56.93
4 0.8 5 80 84.65 82.74 87.52 83.81
5 0.8 8.36 80 49.15 48.95 51.46 51.38
6 0.8 5 114 85.72 80.12 84.35 81.21
7 1.1 3 60 80.61 80.49 83.62 83.58
8 0.5 7 100 59.75 62.34 63.79 65.73
9 0.8 5 80 80.25 82.74 83.45 83.81
10 0.8 5 80 84.46 82.74 85.56 83.81
11 0.8 5 80 83.58 82.74 86.75 83.81
12 0.8 5 80 79.93 82.74 78.66 83.81
13 0.5 3 60 42.06 42.97 40.15 41.65
14 1.3 5 80 88.69 83.77 90.58 87.66
15 0.8 1.63 80 84.56 81.26 86.72 84.11
16 1.1 3 100 92.16 98.73 95.23 98.56
17 0.5 3 100 74.6 74.80 77.86 78.01
18 0.8 5 80 82.95 82.74 80.45 83.81
19 1.1 7 100 60.56 62.12 64.16 64.57
20 0.5 7 60 45.25 41.16 40.14 37.71

Fig. 1. Scatter plot of predicted vs actual for the percent removal of COD and BOD using SBAC.

4
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

Table 4
Analysis of variance (ANOVA) and regression coefficients of quadratic model for percent of BOD and COD removal.
Source COD removal (%) BOD removal (%)

Sum of square df Mean square F-value P-value Sum of square df Mean square F-value P-value

Model 5840.11 9 648.90 53.48 <0.0001 Significant 6203.85 9 689.07 56.07 <0.0001 Significant
A 1282.63 1 1282.36 105.70 <0.0001 1418.95 1 1418.95 115.43 <0.0001
B 1357.57 1 1357.57 111.88 <0.0001 1293.30 1 1293.30 105.21 <0.0001
C 1426.97 1 1426.97 117.60 <0.0001 1652.45 1 1652.45 134.42 <0.0001
AB 279.54 1 279.54 23.04 0.0007 258.10 1 258.10 21.00 0.0010
AC 99.33 1 99.33 8.19 0.0169 207.67 1 207.67 16.89 0.0021
BC 51.46 1 51.46 4.24 0.0665 34.78 1 34.78 2.83 0.1235
A 390.70 1 390.70 32.19 0.0002 318.53 1 318.53 25.91 0.0005
B 564.70 1 564.70 45.54 <0.0001 465.06 1 465.06 37.83 0.0001
C 648.27 1 648.27 53.43 <0.0001 802.19 1 802.19 65.26 <0.0001
Residual 121.34 10 12.13 122.93 10 12.29
Lack of fit 99.95 5 19.99 4.67 0.058 Not-Significant 55.49 5 11.90 0.9379 0.5272 Not-Significant
Pure error 21.39 5 4.28 63.44 5 12.69
Cor total 5961.45 19 6326.78 19
Std. Dev 3.48 3.51
Mean 70.33 71.63
C.V % 4.95 4.90
R 0.979 0.98
Adj. R 0.961 0.96
Pre. R 0.864 0.91
Adeq. Precision 26.264 24.94
PRESS 807.29 568.08

Where, A is SBAC dosage (g/L), B is pH and C is Contact time (min).

effluent collected from Ayka Addis Textiles are presented in Table 2. The where, A is SBAC dose (g/L), B is pH and C is contact time (min). Co­
wastewater sample was bluish in color and alkaline with a pH of 8.9. The efficients with negative values present negative leverage in the
light absorbance capacity was examined on untreated wastewater and quadratic models for COD and BOD removal. Although adsorption effi­
found to be 95%. Its discharge into the environment without proper ciency of the SBAC adsorbent powder was lower when the dose was in
treatment could cause serious harm to the fauna, flora and microbial the range of 0.5 to 0.7 g/L, the high amount COD and BOD removal
biota. Such excessive alkalinity is a measure of textile wastewater occurred at this level. When increasing the SAC dose above 0.8 g/L,
strength resulting from dyeing processes (Yaseen and Scholz, 2019). particularly in the range between 0.9 and 1.3 g/L, the highest removal of
Textile effluents are a complex combination of chemicals, varying in both COD and BOD was observed. Hence, the SBAC dose can be one of
composition with time and from factory to factory as well as on a tem­ the most significant factors that needs to be considered to obtain optimal
poral basis at individual factories based on the materials used in wet conditions for adsorption. However, insufficient dose may result in poor
processing (Drumond Chequer et al., 2013). adsorption. Hence, it is required to determine the optimal dose to
minimize sludge formation, costs, also to attain improved performance
3.3. RSM analysis of COD and BOD removal during treatment.
It was observed that the variation in pH had a significant effect on the
Through the various combinations of parameters, 20 experiments removal efficiency of BOD and COD. As shown in the Table 3, the highest
were designed and executed. The experimental and predicted responses COD and BOD removal were detected in the pH range of 1 to 5.
obtained from CCD experiments are given in Table 3. Improved adsorption could be achieved at lower pH levels, which can be
attributed to the presence of excessive levels of H+ ions. This might be a
3.4. Model fitting and regression analysis for COD and BOD removal reason for neutralizing the negatively charged adsorbent surface. By
comparison, lower adsorption efficiencies were observed at higher pH
The weight of each factor in the model was evaluated using ANOVA. values due to the abundance of OH− ions and the resultant ionic
Several different types of mathematical models were evaluated. A repulsion between the negatively charged active sites of the adsorbent
quadratic model proposed by Design-Expert 12.0 software was chosen surfaces. Similar outcomes were found for the reduction of COD and
after analyzing R2 values, lack of fit, and p-values. Experimental and color from cotton textile using bamboo-based activated carbon as an
predicted values for percent removal of COD and BOD were obtained adsorbent (Ahmad and Hameed, 2009). Although lower COD and BOD
and compared to validate this quadratic model. The linear correlation removal was observed with shorter contact times, the highest percent­
between the experimental and predicted values is given in Fig. 1 The ages of removal were found in the range of 80–100 min of reaction time.
models for COD and BOD removal showed adjusted R2 values of 0.96 When reaction time exceeded 100 min, there was no substantial increase
and 0.96, and predicted R2 values of 0.87 and 0.91, respectively, indi­ in removal. Possibly, at the beginning, solutes molecules were adsorbed
cating good model fit. The difference between the adjusted R2 and at the exterior surfaces of SBAC particles. Therefore, the adsorption rate
predicted R2 was less than 2. Thus, predictions of this quadratic model was higher. When the adsorption at these pores reached saturation,
are acceptable. Eqs. (4) and (5) can be respectively applied to predict the additional molecules could not be adsorbed. Then, the molecules could
response for COD and BOD removal percent using SBAC. diffuse through the pores of the adsorbent to the interior surfaces of the
particles. Since this process would require time, there would be an
COD Removal (%) = 82.74 + 9.76A − 10.04B + 10.30C − 6.04AB optimal adsorption time.
− 3.04AC + 2.08BC − 5.18A2 − 6.23B2 − 6.68C2 (4)

BOD Removal (%) = 83.81 + 10.19A − 9.73B + 11.00C − 5.68AB 3.5. Analysis by ANOVA
2
− 5.10AC − 2.08BC − 4.7A − 5.68B − 7.46C 2 2
(5)
CCD coupled with RSM gave better results for the adsorption treat­
ments for textile effluents using SBAC. The significance of the coefficient

5
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

Fig. 2. 3D interactive surface plot for COD and BOD removal percent from textile effluent by SBAC.

terms was examined using the p-values and F-values (Table 4). Higher F- which indicated adequate signal for COD and BOD removal, respec­
values and lower p-value indicate a more significant model. For both tively. A ratio of more than four signifies the importance of this
COD and BOD removal, the model had p-values less than 0.0001 and F- quadratic model. Thus, these models for BOD and COD removal can be
values of 53.48 and 56.07, respectively (Table 4). The lack of fit values used to navigate the design space. The smaller value of coefficient of
for COD and BOD removal were 4.67 and 0.94, which were not signif­ variation (for COD = 4.95% and for BOD = 4.90%) favors precision with
icant relative to the pure error. The signal to noise ratio was measured the experiments results (Venkatesa et al., 2019). The positive value of A
using Adeq precision. Their values were observed to be 24.26 and 24.95, and C coefficients show these parameters have a positive effect on the

6
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

Table 5 has negative effect on the BOD and COD removal. In addition, it was
RSM found an optimum condition for COD and BOD (%) removal at a desir­ observed that interaction of pH and contact time (BC), pH and adsorbent
ability level of 1. dosage (BA), and interaction of contact time and adsorbent dosage (CA)
Factors and Response Goal Importance Value also showed negative quadratic effect on the BOD and COD removal
SBAC dose (g/L) In range 3 0.5–1.1
efficiency.
pH In range 3 3–7
Contact time (min) In range 3 60–100 3.6. Interactive effects of SBAC dose, pH and contact time on adsorption
COD removal (%) Maximum 5 94.16 processes
BOD removal (%) Maximum 5 95.23

3D surface plots (Fig. 2) presenting the interaction effect among AB,

COD decrease. This means that the COD of wastewater could be AC and BC against the response aid in understanding the meaning of the
decreased by working at SBAC dosage and contact time closed to their effects of parameter combinations. This study reveals that the effects of
higher limits of the studied interval. Moreover, the negative sign of pH interactions for SBAC dose, pH and contact time are very important to
coefficient indicated that working at lower pH could improve the explain the effectiveness of SBAC in removing COD and BOD from textile
removal of COD at the end of the treatment process. The p values for A, effluents. The data given in Fig. 2 shows that an increase in SBAC dose
B, C, AB, AC, A2, B2, and C2 were highly significant (p value lower than and contact time results in incrementally higher removal of both COD
0.005). This implies that these factors contribute significantly towards and BOD. Nevertheless, increased pH values tend to decrease the
removal of COD during the adsorption. Based on the model developed removal of COD and BOD due to a deficiency in H+ ion concentration.
for removal efficiency of BOD and COD, the most dominant parameter is An increased contact time beyond 80 min and SBAC dose greater than
the contact time (C) followed by the adsorbent dosage (A) with a posi­ 0.8 g/L do not show a substantial increase in removal by the adsorbent
tive effect for COD removal, then the solution pH (B) but with a negative since the pores of the adsorbent reached saturation and sludge formed.
effect on the BOD and COD removal. Pure quadratic (A2, B2 and C2) also Additionally, a pH less than 4 had a similar effect on the removal of BOD
and COD. It was shown that these three parameters determined the

Fig. 3. COD and BOD removal with respect to time (a) and data fitting to psedo-first order kinetics approch (b).

7
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

performance of the adsorption process. Devi et al. (2008) did experi­ previously.
mentation to remove BOD and COD from a coffee processing plant
effluent using a batch mode adsorption method. They found about in­ Role of the funding source
fluence of adsorbent dose, effluent pH, contact time and stirring rate on
adsorption process. Not applicable.

3.7. RSM optimization and validation of the model CRediT authorship contribution statement

The outcome of this study revealed that at low pH with contact times Author Contribution
Surafel Mustefa Beyan Experimentation and draft preparation
longer than 80 min, and an SBAC dose greater than 0.9 g/L, the removal
Sundramurthy Venkatesa Prabhu Experimental plan and final formatting
efficiency of the adsorbent was greater. RSM numerically optimized drafting
values for maximum removal of COD and BOD are given in Table 5. Tsegaye Sissay T and Abraham Kinetics and characterization
Under the optimal conditions given in this table, a numerical regression Getahun A
was done using RSM and aided by Design-Expert 12.0 software to
optimize the adsorption process. The resulting solution, with a desir­
ability of 1, presented optimal values for SBAC dose (0.915 g/L), pH
(3.225) and contact time (98.91 min) yielding 96.13% COD and 96.74% Declaration of competing interest
BOD removal. Additionally, triplicate experiments were carried out to
validate the model. The experiment conducted at optimum conditions The authors declare that they have no known competing financial
showed (94.75% COD and 95.49% BOD) good agreement with the interests or personal relationships that could have appeared to influence
predicted values. the work reported in this paper.

3.8. Kinetics of BOD and COD removal Appendix A. Supplementary data

The design of the experiments coupled with RSM were applied in this FTIR and XRD data were provided in supplementary material sec­
study for removal of BOD and COD from textile wastewater. A first order tion. Supplementary data to this article can be found online at https://
kinetics model was employed to determine the rate constants (kp) for doi.org/10.1016/j.biteb.2021.100664.
BOD and COD removal. Statistically optimized parameters from the RSM
analysis (dose of SBAC, 0.915 g/L and pH value, 3.225) were adopted References
from the experiments to accomplish the kinetics analysis. In this context,
various experiments were done with different contact times (0–100 Ahmad, A.A., Hameed, B.H., 2009. Reduction of COD and color of dyeing effluent from a
min). The percent removal of BOD and COD with respect to the different cotton textile mill by adsorption onto bamboo-based activated carbon. J. Hazard.
Mater. 172, 1538–1543. https://doi.org/10.1016/j.jhazmat.2009.08.025.
contact times is presented in Fig. 3(a). Using the experimental data, plots Ahmad, K.S., Amjad, I., Ali, D., 2020. Adsorption and sugarcane-bagasse-derived
of ln[Cpi/(Cpi − Cpt)] vs time were generated and are presented in Fig. 3 activated carbon-based mitigation of 1-[2-(2-chloroethoxy)phenyl]sulfonyl-3-(4-
(b). The linear fitness of the data was examined based on statistical methoxy-6-methyl-1,3,5-triazin-2-yl) urea-contaminated soils. Open Chem. 18,
1433–1443. https://doi.org/10.1515/chem-2020-0185.
measure (using the correlation coefficient [R value]) of how close the Ajinomoh, C.S., Salahudeen, N., 2014. Production of activated carbon from sugar cane
data are to the fitted linear regression line. The R value for BOD and COD bagasse. Aust. J. Ind. Res. 1, 16–22.
removal were observed to be 0.9677 and 0.9819, respectively, which Akl, M.A., 2014. Efficient removal of phenol from water samples using sugarcane bagasse
based activated carbon. J. Anal. Bioanal. Tech. 5 https://doi.org/10.4172/2155-
sound good. So, the determined kp value has been expressed as the rate 9872.1000189.
of removal of BOD/COD with respect to time (within the optimal time). Alkaya, E., Demirer, G.N., 2014. Sustainable textile production: a case study from a
These values indicate the fitness and suitability of the first-order woven fabric manufacturing mill in Turkey. J. Clean. Prod. 65, 595–603. https://doi.
org/10.1016/j.jclepro.2013.07.008.
kinetics model for BOD/COD removal processes within the optimal
Arbib, Z., Ruiz, J., Álvarez-Díaz, P., Garrido-Pérez, C., Perales, J.A., 2014. Capability of
contact time. However, based on R value, the experimental data used in different microalgae species for phytoremediation processes: wastewater tertiary
the first order kinetics model fitted the data for COD removal slightly treatment, CO2 bio-fixation and low cost biofuels production. Water Res. 49,
465–474. https://doi.org/10.1016/j.watres.2013.10.036.
better. According to the first order kinetics model, the rate constant
Chaiwon, T., Jannoey, P., Channei, D., 2017. Preparation of activated carbon from
values for BOD and COD removal were 0.0225 and 0.023 mg/L min− 1, sugarcane bagasse waste for the adsorption equilibrium and kinetics of basic dye.
respectively. These results are supported by Arbib et al. (2014), who Key Eng. Mater. 751, 671–676. https://doi.org/10.4028/www.scientific.net/
studied the kinetics of organic load removal from wastewaters. Trapp KEM.751.671.
Chen, C.-X., Huang, B., Li, T., Wu, G.-F., 2012. Preparation of phosphoric acid activated
et al. (2007) also used the applicability of kinetics models to investigate carbon from sugarcane bagasse by mechanochemical processing. BioResources 7,
bioremediation of pollutants from contaminated soils and waters. 5109–5116. https://doi.org/10.15376/biores.7.4.5109-5116.
Cueva-Orjuela, J.C., Hormaza-Anaguano, A.D.S., Merino-Restrepo, A., 2017. Sugarcane
bagasse and its potential use for the textile effluent treatment. DYNA 84, 291–297.
4. Conclusions https://doi.org/10.15446/dyna.v84n203.61723.
Danish, M., Ahmad, T., Majeed, S., Ahmad, M., Ziyang, L., Pin, Z., Shakeel Iqubal, S.M.,
This study focused on development of a process model followed by 2018. Use of banana trunk waste as activated carbon in scavenging methylene blue
dye: kinetic, thermodynamic, and isotherm studies. Bioresour. Technol. Reports 3,
optimization of the removal of BOD and COD from a textile wastewater 127–137. https://doi.org/10.1016/j.biteb.2018.07.007.
effluent using sugarcane bagasse activated carbon (SBAC). The inde­ Dao, T.M., Le Luu, T., 2020. Synthesis of activated carbon from macadamia nutshells
pendent process variables, adsorbent dose, reaction pH, and reaction activated by H2SO4 and K2CO3 for methylene blue removal in water. Bioresour.
Technol. Reports 12, 100583. https://doi.org/10.1016/j.biteb.2020.100583.
time were considered and statistically optimized for the maximum
Devi, R., Singh, V., Kumar, A., 2008. COD and BOD reduction from coffee processing
removal of BOD and COD using RSM technique. The optimal values were wastewater using Avacado peel carbon. Bioresour. Technol. 99, 1853–1860. https://
SBAC dose of 0.915 g/L, pH value of 3.225 and contact time of 98.91 doi.org/10.1016/j.biortech.2007.03.039.
Drumond Chequer, F.M., de Oliveira, G.A.R., Anastacio Ferraz, E.R., Carvalho, J., Boldrin
min. Pseudo-first order kinetics showed that the rate constants for BOD
Zanoni, M.V., de Oliveir, D.P., 2013. Textile Dyes: Dyeing Process and
and COD removal were 0.0225 and 0.023 mg/L min− 1, respectively. Environmental Impact, in: Eco-Friendly Textile Dyeing and Finishing. InTech.
https://doi.org/10.5772/53659.
Submission declaration Franson, M.A.H., 1998. Standard Methods for the Examination of Water and Wastewater,
20th ed. Amer Public Health Assn, Washington, D.C.
Ge, M., Du, M., Zheng, L., Wang, B., Zhou, X., Jia, Z., Hu, G., Jahangir Alam, S.M., 2017.
The authors assure that this work has not been published/submitted A maleic anhydride grafted sugarcane bagasse adsorbent and its performance on the

8
S.M. Beyan et al. Bioresource Technology Reports 14 (2021) 100664

removal of methylene blue from related wastewater. Mater. Chem. Phys. 192, potted plants exposure chamber. J. Air Waste Manage. Assoc. 65, 1413–1420.
147–155. https://doi.org/10.1016/j.matchemphys.2017.01.063. https://doi.org/10.1080/10962247.2015.1100141.
Isik, Z., Arikan, E.B., Bouras, H.D., Dizge, N., 2019. Bioactive ultrafiltration membrane Mpatani, F.M., Aryee, A.A., Kani, A.N., Wen, K., Dovi, E., Qu, L., Li, Z., Han, R., 2020.
manufactured from Aspergillus carbonarius M333 filamentous fungi for treatment of Removal of methylene blue from aqueous medium by citrate modified bagasse:
real textile wastewater. Bioresour. Technol. Reports 5, 212–219. https://doi.org/ kinetic, Equilibrium and Thermodynamic study. Bioresour. Technol. Reports 11,
10.1016/j.biteb.2019.01.020. 100463. https://doi.org/10.1016/j.biteb.2020.100463.
Kan, Y., Yue, Q., Li, D., Wu, Y., Gao, B., 2017. Preparation and characterization of Nikodinovic-Runic, J., Guzik, M., Kenny, S.T., Babu, R., Werker, A., O Connor, K.E.,
activated carbons from waste tea by H 3 PO 4 activation in different atmospheres for 2013. Carbon-Rich Wastes as Feedstocks for Biodegradable Polymer
oxytetracycline removal. J. Taiwan Inst. Chem. Eng. 71, 494–500. https://doi.org/ (Polyhydroxyalkanoate) Production Using Bacteria, pp. 139–200. https://doi.org/
10.1016/j.jtice.2016.12.012. 10.1016/B978-0-12-407673-0.00004-7.
Katheresan, V., Kansedo, J., Lau, S.Y., 2018. Efficiency of various recent wastewater dye Oyebamiji, O.O., Boeing, W.J., Holguin, F.O., Ilori, O., Amund, O., 2019. Green
removal methods: a review. J. Environ. Chem. Eng. 6, 4676–4697. https://doi.org/ microalgae cultured in textile wastewater for biomass generation and
10.1016/j.jece.2018.06.060. biodetoxification of heavy metals and chromogenic substances. Bioresour. Technol.
Kaushik, A., Basu, S., Singh, K., Batra, V.S., Balakrishnan, M., 2017. Activated carbon Reports 7, 100247. https://doi.org/10.1016/j.biteb.2019.100247.
from sugarcane bagasse ash for melanoidins recovery. J. Environ. Manag. 200, Patel, H., Vashi, R.T., 2015. Characterization and Treatment of Textile Wastewater.
29–34. https://doi.org/10.1016/j.jenvman.2017.05.060. Characterization and Treatment of Textile Wastewater. Elsevier. https://doi.org/
Kidus Tekleab, W., Beyan, S.M., Balakrishnan, S., Admassu, H., 2020. Chicken feathers 10.1016/C2014-0-02395-7.
based Keratin extraction process data analysis using response surface-box-Behnken Saad, S.A., Isa, K.M., Bahari, R., 2010. Chemically modified sugarcane bagasse as a
design method and characterization of keratin product. Curr. Appl. Sci. Technol. 20, potentially low-cost biosorbent for dye removal. Desalination 264, 123–128. https://
163–177. https://doi.org/10.14456/cast.2020.6. doi.org/10.1016/j.desal.2010.07.015.
Kim, D.-S., 2004. Activated carbon from peach stones using phosphoric acid activation at Silveira, M.H.L., Chandel, A.K., Vanelli, B.A., Sacilotto, K.S., Cardoso, E.B., 2018.
medium temperatures. J. Environ. Sci. Heal. Part A 39, 1301–1318. https://doi.org/ Production of hemicellulosic sugars from sugarcane bagasse via steam explosion
10.1081/ESE-120030333. employing industrially feasible conditions: pilot scale study. Bioresour. Technol.
Kumar, A., Jena, H.M., 2016. Preparation and characterization of high surface area Reports 3, 138–146. https://doi.org/10.1016/j.biteb.2018.07.011.
activated carbon from Fox nut (Euryale ferox) shell by chemical activation with H 3 Sundramurthy, V.P., Rajoo, B., Srinivasan, N.R., Kavitha, R., 2020. Bioleaching of Zn
PO 4. Results Phys. 6, 651–658. https://doi.org/10.1016/j.rinp.2016.09.012. from sphalerite using Leptospirillum ferriphilum isolate: effect of temperature and
Kumar, V., Singh, J., Nadeem, M., Kumar, P., Pathak, V.V., 2020. Experimental and kinetic aspects. Appl. Biol. Chem. 63 https://doi.org/10.1186/s13765-020-00528-8.
kinetics studies for biogas production using water hyacinth (Eichhornia crassipes Tahir, H., Sultan, M., Akhtar, N., Hameed, U., Abid, T., 2016. Application of natural and
[Mart.] Solms) and sugar mill effluent. Waste and Biomass Valorization 11, 109–119. modified sugar cane bagasse for the removal of dye from aqueous solution. J. Saudi
https://doi.org/10.1007/s12649-018-0412-9. Chem. Soc. 20, S115–S121. https://doi.org/10.1016/j.jscs.2012.09.007.
Lellis, B., Fávaro-Polonio, C.Z., Pamphile, J.A., Polonio, J.C., 2019. Effects of textile dyes Tetteh, E.K., Rathilal, S., 2017. Application of Response Surface Methodology (RSM) -
on health and the environment and bioremediation potential of living organisms. Reduction of Industrial Wastewater Chemical Oxygen Demand. CBU Int. Conf. Proc.
Biotechnol. Res. Innov. 3, 275–290. https://doi.org/10.1016/j.biori.2019.09.001. 5, pp. 1226–1232. https://doi.org/10.12955/cbup.v5.1101.
Ma, Y., Huang, M., Wan, J., Hu, K., Wang, Y., Zhang, H., 2011. Hybrid artificial neural Trapp, S., Ücisik, A.S., Romano, P.D., Larsen, M., 2007. The Role of Plants and Bacteria in
network genetic algorithm technique for modeling chemical oxygen demand Phytoremediation - Kinetic Aspects, in: Bioremediation of Soils Contaminated with
removal in anoxic/oxic process. J. Environ. Sci. Heal. Part A 46, 574–580. https:// Aromatic Compounds. Springer Netherlands, Dordrecht, pp. 41–49. https://doi.org/
doi.org/10.1080/10934529.2011.562821. 10.1007/978-1-4020-5693-2_4.
Manohari, Singh, J., Kadapakkam Nandabalan, Y., 2017. Copper(II) bioremoval by a Venkatesa Prabhu, S., Baskar, R., 2015. Kinetics of heavy metal biosolubilization from
rhizosphere bacterium, stenotrophomonas acidaminiphila MYS1-process electroplating sludge: effects of sulfur concentration. J. Korean Soc. Appl. Biol.
optimization by RSM using Box–Behnken design. Int. J. Environ. Res. 11, 63–70. Chem. 58, 185–194. https://doi.org/10.1007/s13765-015-0027-9.
https://doi.org/10.1007/s41742-017-0007-5. Venkatesa, P.S., Girma, G., Gizachew, A.K., Surafel .G,.R., B.M., 2019. Biosolubilization
Mistar, E.M., Alfatah, T., Supardan, M.D., 2020. Synthesis and characterization of of Cr (VI) from tannery sludge: process modeling, optimization, rate kinetics and
activated carbon from Bambusa vulgaris striata using two-step KOH activation. thermodynamics aspects. Int. J. Recent Technol. Eng. 8, 4808–4816. https://doi.
J. Mater. Res. Technol. 9, 6278–6286. https://doi.org/10.1016/j.jmrt.2020.03.041. org/10.35940/ijrte.d7771.118419.
Mohamed, E.F., El-Hashemy, M.A., Abdel-Latif, N.M., Shetaya, W.H., 2015. Production Yaseen, D.A., Scholz, M., 2019. Textile dye wastewater characteristics and constituents
of sugarcane bagasse-based activated carbon for formaldehyde gas removal from of synthetic effluents: a critical review. Int. J. Environ. Sci. Technol. 16, 1193–1226.
https://doi.org/10.1007/s13762-018-2130-z.