Journal of Environmental Management 92 (2011) 2454e2460

Contents lists available at ScienceDirect

Journal of Environmental Management

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

Remediation of biodiesel wastewater by chemical- and electro-coagulation:

A comparative study
Krit Ngamlerdpokin a, Sasipan Kumjadpai a, Preeya Chatanon a, Ungsika Tungmanee a,
Sulalit Chuenchuanchom a, Pattaraluk Jaruwat b, Prarinya Lertsathitphongs c, Mali Hunsom a, d, *
a
Department of Chemical Technology, Faculty of Science, Chulalongkorn University, Phaya Thai Rd., Bangkok 10330, Thailand
b
Department of Environmental Science, Faculty of Science, Chulalongkorn University, Phaya Thai Rd., Bangkok 10330, Thailand
c
The Bangchak Petroleum Public Company Limited, 210 Moo 1, Sukhumvit 64, Sukhumvit Rd., Bangchak, Phrakanong, Bangkok 1026, Thailand
d
Center for Petroleum, Petrochemicals and Advanced Materials, Chulalongkorn University, Bangkok 10330, Thailand

a r t i c l e i n f o a b s t r a c t

Article history: The remediation of biodiesel wastewater was carried out using chemical and electrochemical techniques.
Received 12 May 2010 Initially the fatty acid methyl esters (FAME or biodiesel) and free fatty acids (FFA) were chemically
Received in revised form removed from the wastewater using three types of mineral acids, H2SO4, HNO3 and HCl, at different pH
29 April 2011
values within the range of 1.0e8.0. Optimally, approximately 24.3 ml/l of FAME/FFA were removed from
Accepted 8 May 2011
the wastewater when using H2SO4 to set a final pH of 2.5 for 7 min. All pollutant levels were markedly
Available online 8 June 2011
reduced during this step. That is, approximately 38.94%, 76.32% and 99.36% of COD, BOD5 and oil &
grease were respectively removed. The acidic aqueous phase left after the removal of the FAME/FFA
Keywords:
Coagulation
phase was then treated by chemical- and electro-coagulation processes. The results demonstrated that
Electro-coagulation both investigated treatment processes were effective for treating wastewater from a biodiesel production
Biodiesel wastewater plant. The chemical coagulation provided a lower operating cost (1.11 USD/m3) compared with the
Oil-rich phase electro-coagulation process (1.78 USD/m3). However, the latter process provided a better quality of
wastewater compared with the former process, with the exception of the BOD levels.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction in the presence of a basic or acidic catalyst. The untreated biodiesel

Since fossil fuels such as petroleum, natural gas and coal are
non-renewable and limited together with the increasing energy
demands of the present day, there is an increasing urgency to
search for new sustainable and renewable power sources that can
produce a sufficient quantity of power with acceptable safety
(environmental and human), efficiency and reliability criteria.
Biodiesel or fatty acid methyl esters (FAME) is one of the most
promising candidates as an alternative fuel for the transportation
sector because it is biodegradable, non-toxic, burns with a low
sulfur, carbon monoxide and aromatic-free emission profile, and is
environmentally beneficial in terms of, for example, recycling of
spent oils and fats (Karmee and Chadha, 2005). Biodiesel can be
produced from the transesterification of triglycerides (from vege-
table oils or animal fats) and alcohol, such as methanol or ethanol,
* Corresponding author. Department of Chemical Technology, Faculty of Science,
Chulalongkorn University, Phaya Thai Rd., Bangkok 10330, Thailand. Tel.: þ662
2187523; fax: þ662 2555831.
E-mail address: mali.h@chula.ac.th (M. Hunsom).
contains several impurities, such as free glycerol, soap, metals,
methanol, free fatty acids (FFA), catalyst, water and glycerides,
which will impact on the performance and durability of the diesel
engine. For example, the presence of glycerides and soap in bio-
diesel leads to the formation of high carbon residues, which can
plug or block the injector, whilst the presence of glycerol can
increase aldehyde and acrolein emissions (Berrios and Skelton,
2008). Therefore, the purification stage is essential. The more
traditional purification method is wet washing, which involves
using water or a weak acid to remove some of the excess
contaminants and leftover production chemicals from the biodiesel
production. However, the inclusion of additional water to the
process offers many disadvantages, including an increased cost and
production time, the generation of a highly polluting effluent
(wastewater) that needs to be treated prior to environmental
discharge (Berrios and Skelton, 2008) and the significant loss of
biodiesel into the wastewater phase (Canakci and Gerpen, 2001).
Currently, greater than 70,000 l/day of such biodiesel waste-
water are produced in Thailand (Jaruwat et al., 2010), making this
a significant problem. The common procedure for managing such
wastewater is carried out by the installation of a pre-treatment

0301-4797/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvman.2011.05.006
 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
Table 1
Chemical and physical properties of raw wastewater from a biodiesel production plant in Thailand, and the same after treated by the chemical- and electro-coagulation
processes.
Properties Thai standard Raw BDW
pH
COD (mg/l) 400 312,000e588,800
BOD5 (mg/l) 60 168,000e300,000
Oil & grease (mg/l) 5 18,000e22,000
TDS (mg/l) 3000 26,580e30,527
TKN (mg/l) 100 439e464
a
Not analyzed.
system prior to deliver to a treatment facility of wastewater agency
(most likely IEAT). This plant accepts wastewater for which the
BOD does not exceed 500 mg/l. Factories discharging wastewater
exceeding the BOD limit have to pay penalties and additional
charges required to treat the wastewater. However, due to biodiesel
wastewater contained extremely high concentration of BOD, the
pre-treated wastewater from the pre-treatment system is usually
higher than 500 mg/l. If the biodiesel wastewater generators send
their wastewater to the wastewater agency, they have to pay
around 128.45e160 USD/m3 (Kumjadpai et al., 2011). It is very
expensive, alternative option to manage this wastewater is the
incineration in cement industry. Because of this wastewater
contains small amount of biodiesel which lose during the washing
step, it still has small heating value. By incinerating in cement
industry, the biodiesel wastewater producers have to play around
60.35 USD/m3, 2-time lower than sending to the treatment effi-
ciency. Nevertheless, this option is still expensive compared with
the treatment cost of other industrial wastewaters.
To find out the suitable treatment procedure for managing
biodiesel wastewater for both environmental and economic
reasons, several processes have been developed to treat biodiesel
wastewater. For example, by biological process, Suehara et al.
(2005) reported that the oil degradable yeast, Rhodotorula
mucilaginous, was effective to treat biodiesel wastewater at the
dilution ratio of 1:1 (v/v). At optimum concentration of yeast (1 g/l,
C/N ratio ¼ 17e68), greater than 97% of oil was degraded. Siles et al.
(2010) pointed out that the anaerobic co-digestion of glycerol and
wastewater derived from biodiesel manufacturing is a good alter-
native for treating and revalorizing both wastes. Following the pre-
treatment of glycerol by acidification and the electro-coagulation of
the wastewater, the mixture showed a high level of anaerobic
biodegradability (around 100%), permitting a substantial quantity
of methane to be obtained (310 mL CH4/g CODremoved, at 1 atm
and 25  C). Besides, by using the electro-coagulation process,
Chavalparit and Ongwandee (2009) found that the level of COD, oil
& grease and SS was reduced by 55.43%, 98.42%, and 96.59%,
respectively, at the optimum conditions of pH 6.06, applied voltage
18.2 V, and reaction time 23.5 min. Consequently, Jaruwat et al.
(2010) used a chemical recovery approach and electrochemical
treatment to treat biodiesel wastewater. They pointed out that
approximately 6e7% (w/w) of likely biodiesel was recovered by the
protonation reaction with H2SO4 at a pH of 2e6, and the levels of
BOD, COD and oil and grease were reduced by more than 13e24%,
40e74% and 87e98%, respectively, depending upon the pH. The
residual denser aqueous phase was consequently treated by elec-
trooxidation with or without the addition of NaCl, the latter being
found to be optimal at 0.061 M, where greater than 95% of BOD, and
100% of COD and oil and grease levels were eliminated within 6, 7
and 2 h, respectively, using a current density of 4.28 mA/cm2. By
using the chemical process, Kumjadpai et al. (2011) found that
the chemical recovery by the addition of H2SO4 at pH 1e2.5 and
2455
After FAME/FFA After FAME/FFA removal and treated by
Chemical coagulation Electro-coagulation
2.5 3.4 12.5
271,000e341,712 6480 1600
6739e67,389 3000 11,823
210e421 105 80
7310e17,495 16,612 NA
NAa NA NA
coagulation treatment by Al2(SO4)3 was effective to treat biodiesel
wastewater. Greater than 98.3%, 97.7%, and 99.2% of COD, BOD5, and
oil and grease were respectively removed using Al2(SO4)3 at 2 g/l,
whilst that achieved by PAC coagulation (at 1 g/l) was slightly lower
at 98.2%, 96.5%, and 98.6%, respectively. According to these works,
each of them has been reported in several studies to be the most
efficient and economic way to reduce the environmental impacts of
biodiesel wastewater. However, each study started with a different
source of biodiesel wastewater with different concentrations of
pollutants. Therefore, it is hard to compare and justify what is the
most effective process for treating biodiesel wastewater.
In this work, due to the presence of a high concentration of
pollutants, the FAME and FFA were first removed by acid proton-
ation and then the leftover wastewater was further treated by two
further processes; chemical coagulation with Al2(SO4)3 and electro-
coagulation. The effectiveness of each treatment of the biodiesel
wastewater was evaluated and judged in terms of the treatment
efficiency and estimated likely operating cost.
2. Materials and methods
The treatment of wastewater from a conventional biodiesel
production plant that uses waste used vegetable oil as feedstock
was carried out on a laboratory scale at ambient temperature with
two steps; the removal of FAME and FFA by acid protonation fol-
lowed by the chemical and electrochemical treatments of the
residual aqueous waste liquid. The wastewater utilized in this work
was collected from the washing unit of a biodiesel production
process and its properties are summarized in Table 1. For the first
treatment step, the wastewater was acidified in a separation funnel
by the addition of one of three types of mineral acid; H2SO4 (98.08%
Fig. 1. Experimental set-up used for electro-coagulation: (1) Reactor; (2) Pump; (3)
Valve; (4) Inlet stream and (5) Outlet stream.
2456 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
Fig. 2. Evaluation of the oil-rich phase removal from biodiesel wastewater by acid
protonation for 15 min with H2SO4 (,), HCl ( ) or HNO3 ( ) at the indicated pH
values.
Mallinckrodt chemicals), HNO3 (65%, Lab scan) or HCl (35.4%, BHD
laboratory), at different final concentrations, measured as the final
pH value, within the range of pH 1.0e8.0. The mixture was then
shaken for 1.5 h before then being left for 2 h to allow the complete
phase separation between the upper oil-rich phase and the lower
acidic aqueous phase. The structure of the oil-rich phase was then
characterized using 1H-NMR (Mercury 400, Varian), as previously
described (Miyake et al., 1998).
The remaining discharged wastewater (aqueous phase) from the
first step after extraction of the salvaged FAME/FFA phase, was then
treated by the chemical coagulation and the electro-coagulation
process as follows. For the chemical coagulation process, the pH of
the aqueous wastewater phase was adjusted to be within the
preferred range (pH 4.5e10) by the addition of CaO. Subsequently,
approximately 2 l of wastewater was conventionally treated by the
addition of Al2(SO4)3 (range; 0e6 g/l) as the coagulant in a Jar
reactor with a fast stirring rate (range; 100e300 rpm) for 1 min
followed by a slow stirring rate of 40 rpm for 20 min. For the
electro-coagulation treatment, the cell was constructed with
Plexiglas with a total volume of wastewater in each trial of
approximately 3 l. A low-cost iron plate with a total surface area of
Fig. 3. Representative FTIR spectra of (a) the oil-phase substances recovered by H2SO4 acidification at a pH of 2.5 and (b) the biodiesel obtained by the transesterification reaction.
161 cm2 was used as the sacrificial electrode. In each batch, six iron
plates were arranged in a monopolar configuration and the
distance between the plates was fixed at approximately 5 cm
(Fig. 1). To achieve a good mass transfer in the electrochemical
reactor, a magnetic pump (Model NH-5PX type) was used to
circulate the electrolyte in the reactor. A regulated DC power supply
(ZS 3205-2X type) was employed to supply the external electricity.
During each of the above treatments a 5 ml sample was removed at
the indicated time periods and assayed for the chemical oxygen
demand (COD), biological oxygen demand (BOD5) and the oil &
grease levels of the wastewater following standard methods (APHA,
AWWA, WEF, 1998).
3. Results and discussion
3.1. Removal of FAME/FFA by acid protonation
Three types of strong mineral acid, H2SO4, HNO3 and HCl, were
used to extract the residual FAME/FFA from the biodiesel waste-
water by protonation at different final concentrations, measured as
the more relevant pH values. Generally, when each acid was added
directly into the biodiesel wastewater, the mixture automatically
separated into two phases. The upper yellow colored phase was the
oil-rich phase containing the FAME/FFA, whereas the lower
aqueous phase had a low turbidity and was transparent. For all
three acids and concentrations (pH values) evaluated, the oil-rich
phase was rapidly formed within the first 7 min and thereafter
the phase separation (formation) rate reached a plateau. Higher
yields, shown as the quantity of the oil-rich phase removed, were
obtained at lower pH conditions (pH 1 and 2.5) compared with that
attained under weakly acidic conditions (pH 4.5 and, especially, pH
6), whilst essentially no recovery was noted with weakly alkaline
(pH 8) conditions (Fig. 2). Nevertheless, the three different acids
yielded different levels of oil-rich phase formation (i.e. FAME/FFA
removal efficiencies), where H2SO4 provides the most efficient
removal of the FAME/FFA at the same pH value. Thus, approxi-
mately 24.5, 15.1 and 21.2 ml/l of FAME/FFA were removed from the
acidified wastewater using H2SO4, HNO3 and HCl, respectively, at
pH 2.5.
From the FTIR spectra (Fig. 3), the functional groups of the
removed oil-rich phase were similar to those of the original FAME
 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
obtained from the transesterification of wastewater used oil, con-
sisting of CH3 alkane stretching (2923 and 2853 cm1) and bending
(1436 and 1463 cm1). However, the oil-rich phase also had two
peaks at 1742 cm1 and 1711 cm1 that are ascribed to the C]O
peak of the methyl ester (COOMe) and the carboxylic group of the
FFA, respectively. This indicates that the principle components in
the removed oil-rich phase are FAME and FFA, which was further
supported by the 1H-NMR analysis (Fig. 4). Here, a strong charac-
teristic peak at 3.6 ppm, arising from the protons of the methyl
ester moiety, was clearly observed as well as (i) peaks near 0.8 ppm,
related to the terminal methyl proton of saturated, mono-
unsaturated and n-6 polyunsaturated fatty acids, (ii) a strong signal
at 1.2 ppm from the methylene protons of saturated carbon atoms,
(iii) a signal at 1.6 ppm from the b-carbonyl methylene protons, (iv)
a multiplet at 2.2 ppm related to the a-carbonyl methylene protons,
and (v) signals associated with unsaturation at 2.0, 2.8 and 5.3 ppm
which are assigned to the allyl methylene protons, bis-allylic or
divinyl methylene protons and olefinic protons and methine proton
in the glyceryl group, respectively. The quantity of FAME/FFA
obtained from the integration of 1H-NMR peaks (Knothe, 2000;
Yeung et al., 2008) were around 44.8% and 55.1%, respectively.
This emphasized that FAME, as well as fatty acids, were removed
from the biodiesel wastewater by acid protonation.
The removal of FAME and FFA from biodiesel wastewater in the
presence of a strong proton donor may be speculated that the
Hþ from the H2SO4 would quickly neutralize any residual alkali
catalyst in the biodiesel wastewater and substitute the Na atom in
the soap molecule, arising from the reaction between triglyceride
and base catalyst (ex. NaOH) in the presence of water in the initial
feedstock or by esterification of FFA in the feedstock with alcohol,
leading to the formation of uncharged FFA. In addition, it can also
substitute the H2O molecule combining biodiesel leading to the
formation of the free FAME (Jaruwat et al., 2010). Therefore, the
product from this acid protonation step is the mixture of FAME
Fig. 4. Representative 1H-NMR spectra of the oil-rich phase recovered from biodiesel wastewater by acidification with H2SO4 to pH 2.5.
2457
and FFA, which can be used as a raw material for the biodiesel
production process.
The properties of the aqueous phase attained after the proton-
ation and extraction of FAME/FFA by H2SO4 at different pH values
was then characterized in terms of COD, BOD5 and oil & grease
levels. All investigated pollutants were significantly reduced within
the first 15e30 min reaction time, and thereafter remained stable
or increased slightly (data not shown). A significantly higher
removal efficiency of all pollutants was observed under strong
acidic conditions (pH 2.5 and especially 1) compared with that
obtained at weakly acidic (pH 4 and 6) or weakly alkaline (pH 8)
conditions (Fig. 5). The removal of oil & grease was close to 100% at
pH 1 for all three acids. With respect to the removal of COD and
BOD5, the presence of H2SO4 at pH 1 and 2.5 promoted the highest
removal level compared with that obtained by the other two
mineral acids and pH conditions (Fig. 5 and Table 1). This is
attributed to the fact that a large amount of FAME/FFA were sepa-
rated from the wastewater under these acidic conditions leading to
a lower amount of organic substances, particularly oil & grease, in
the residual wastewater. According to these results, the optimum
tested condition for FAME/FFA removal from biodiesel wastewater
was achieved by using H2SO4 to set a low pH range (pH 1 and 2.5).
From the point of view of the efficiency and operating costs of each
treatment, further treatment would be required to deal with the
aqueous phase obtained after biodiesel recovery at pH w2.5.
3.2. Removal of residue pollutants by chemical and electrochemical
methods
3.2.1. Chemical coagulation
The optimum pH and coagulant dose were explored with a fast
stirring rate of 300 rpm for 1 min, followed by a slow stirring rate of
40 rpm for 20 min. Before being subjected to chemical treatment
with Al2(SO4)3, the acidified wastewater (pH w2.5) attained from
2458 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
Fig. 5. The reduction in BOD5 (a), COD (b) and oil & grease (c) levels in the water-rich
phase, after removal of the oil-rich phase, obtained by acidification of the wastewater
with H2SO4 (,); HNO3 ( ) or HCl ( ) to the indicated final pH for15 min.
the first protonation treatment step was adjusted in pH to the
indicated value (range from 4.5 to 10) by the addition of CaO.
Consequently, different doses of Al2(SO4)3, in the range of 0e6 g/l,
were added. The addition of CaO only promoted a significantly
increased level of removal of all COD, and especially BOD5 and oil
and grease, and this was pH dependent being higher as the final pH
of the wastewater was increased (Fig. 6). This is because CaO can
work as coagulant coupling with the pollutant molecules and
leading to the formation of the organic precipitated sludge. More-
over, the subsequent addition of Al2(SO4), at different wastewater
pH values within the range of 4.5e6.0, significantly increased the
removal level obtained for all three categories of pollutants. Thus,
increasing the initial wastewater pH from 4.5 to 6.0 in the presence
of 2 g/l Al2(SO4)3 lead to an increase in the levels of COD, BOD5
and oil & grease removal from 83.2 to 97.5%, 79.3 to 98.6% and 89.0
to 98.9%, respectively. However, the presence of Al2(SO4)3 in
Fig. 6. Reduction in COD (a), BOD5 (b) and oil & grease (c) levels in biodiesel waste-
water after coagulation with the indicated different doses of Al2(SO4)3 at a pH of 4.5
(), 6.0 (A), 7.0 (:), 8.0(-), 9.0 () and 10.0 (þ), set by the addition of CaO.
wastewater with an initial pH within the range of 6.0e10.0 had only
a slight effect on the BOD5 removal efficiency, particularly at a high
coagulant dose (>1.0 g/l), whereas it had no effect on COD and oil &
grease removal. This is because Al2(SO4)3 is effective at reducing
pollutants in wastewater over a relatively wide pH range of 6e8
(Davis and Cornwell, 1998). The effect of Al2(SO4)3 was clearly dose-
dependent, but varied between the three pollutant categories, with
the COD, BOD5 and oil & grease levels being almost completely
removed in the presence of Al2(SO4)3 at 2, 1 and 0.5 g/l, respectively.
Regardless, under these conditions greater than 97.5, 97.2 and
98.2% of COD, BOD5 and oil & grease, respectively, were removed at
an initial wastewater pH of 6.0, and so for the efficient simulta-
neous removal of pollutants, 2 g/l of Al2(SO4)3 at an initial waste-
water pH of 6.0 was selected as the optimal set of conditions for
further study.
Previous studies have indicated that the rate of mixing affects
the coagulation efficiency (Zhou et al., 2008). Therefore, the
 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
influence of rapid mixing on the coagulation performance of bio-
diesel wastewater, at an initial pH of 6, with 0.2% (w/v) Al2(SO4)3,
was investigated within the range of 100e300 rpm for the initial
mixing rate for 1 min, and then followed by a fixed slow stirring
rate of 40 rpm for 20 min (Fig. 7). The obtained removal levels of
both COD and oil & grease were found to be independent of the
mixing rate, with greater than 95.0% and 98.8% of COD and oil &
grease being removed with all the tested initial mixing rates
(100e300 rpm). However, the removal efficiency of BOD5 was seen
to be dependent upon the initial mixing rate and increased from
73.5% at 100 rpm to 96.1% at 250 rpm. The improvement in coag-
ulation performance with the mixing rate is due to the fact that the
collision frequency increases with increasing velocity gradient and
consequently the efficiency will increase (Zhou et al., 2008).
However, increasing the mixing rate above 250 rpm promoted only
a slight further increase (to 97.7%) in the BOD5 removal level
obtained, suggesting then that the reduction of BOD5 was limited
by the kinetics of the coagulation reaction. Therefore, the optimum-
mixing rate should be 250 rpm.
3.2.2. Electro-coagulation technique
The effect of the initial pH of the wastewater on the removal of
BOD5, COD and oil & grease was first explored within the pH range
of 2.6e9.0 at a current density of 7.45 mA/cm2 and a circulating
flow rate of 7.26 l/min. The removal efficiency of COD and BOD5
increased dramatically after the current was applied and reached
their plateau after 4 h of electrolysis, whereas that of oil & grease
reached their maximum removal levels within 0.5 h of electrolysis
time (data not shown).
Using electro-coagulation at a current density of 12.4 mA/cm2,
no detectable difference was seen in the oil & grease removal levels
(within the tested pH range) at pH levels greater than 4.0, since an
almost maximal level (99.5% removal) was achieved with the
lowest tested current density (Fig. 8). For the removal of COD and
BOD5 removals, increasing the initial pH of the wastewater from
2.6 to 7.4 led to the increase in the removal efficiencies of both
pollutants, but they then remained constant as the initial pH of the
wastewater was increased above 7.4. This is most likely explained
by the fact that, in acidic conditions, a large amount of iron ions are
in the form of Fe2þ ion, which is soluble and stable in water
resulting in a low formation of coagulant and pollutant removal.
In contrast, in basic solutions, the redox equilibrium is shifted
to Fe3þ ions which, in monomeric forms such as Fe(OH)3, or in
polyhydroxyl iron (III) complexes such as Fe(OH)2þ, Fe(OH)2þ,
Fe(H2O)5OH2þ and Fe(H2O)4(OH)2þ, are reactive and can couple
Fig. 7. The removal of BOD5 (A), COD (,) and oil & grease (:) levels in biodiesel
wastewater, as a function of the initial mixing rate, by chemical coagulation with 2 g/l
Al2(SO4)3 at an initial pH of 6.
2459
Fig. 8. Electro-coagulation based removal of BOD5 (A), COD (,) and oil & grease (:),
as a function of the initial wastewater pH, at current density of 12.42 mA/cm2,
a circulating flow rate of 7.26 l/min and an electrolysis time of 4 h.
with the pollutants in wastewater (Mollah et al., 2004; Chen, 2004)
leading to a lower amount of pollutants in the wastewater.
The effect of varying the current density was then explored for
this treatment process because the supply of current influences the
amount of ferrous ions produced from the sacrificial electrodes. A
large current allows the use of a small electro-coagulation unit but
wastes electrical energy in heating up the water (Chen, 2004). Thus,
different current densities within the range of 6.21e18.63 mA/cm2
were investigated, with a fixed circulating flow rate of wastewater
of 7.26 l/min at an initial pH of 7.4. The removal efficiency of COD
and BOD5 increased dramatically after the current was applied and
reached a plateau after 4 h of electrolysis, whereas that of oil &
grease reached its maximum level within 1.0 h of electrolysis (data
not shown). As demonstrated in Fig. 9, the oil & grease removal
levels (efficiencies) showed no detectable differences in the
explored current density range, with removal of more than 98.7%
in all cases, probably as they were already almost maximal at
the lowest tested current density. For COD and BOD5 removal,
increasing the current density from 7.45 to 12.42 mA/cm2 lead to an
initial increase in the removal levels obtained for both pollutants up
to 99.6% and 91.5% of COD and BOD5 being removed, respectively, at
a current density of 12.42 mA/cm2. However, further increases in
the current density above 12.42 mA/cm2 led to a significant and
slight decrease in the level of BOD5 and COD removal, respectively.
This might be attributed to the fact that excess Fe2þ ions produced
Fig. 9. Electrocoagulation mediated removal of BOD5 (A), COD (,) and oil & grease
(:) levels as a function of the current density at an initial pH of 7.4, a circulating flow
rate of 7.26 l/min and an electrolysis time of 4 h.
2460 K. Ngamlerdpokin et al. / Journal of Environmental Management 92 (2011) 2454e2460
Table 2
Operating cost analysis of FAME-contaminated wastewater treatment by the two-
stage chemical- and electro-coagulation treatment.
Estimated cost Raw data per m3 Saving cost
wastewater (USD/m3)
1. Removal of FAME and FFAb
- H2SO4 (liters) 2.67
- Recovered FAME (liters) 24.3 12.42
0.35
2. Removal of residue pollutants by coagulation process
- CaO (kg) 5 0.28
- Al2(SO4)3 (kg) 2 0.33
- Electricity (kW-h) 0.385
- Sludge (kg) 5.64
3. Removal of residue pollutants by electro-coagulation process
- Electrode (kg) 0.278
- Electricity (kW-h) 10.667
- Sludge (kg) 0.495
Treatment cost after managementc
Conventional management procedured
a
1 USD ¼ 31.14 Baht.
b Under Graduate Students (IRPUS) (Grant No. I251A13015), the
FAME recovery step using H2SO4 to pH 2.5.
c
pH adjustment and biological treatment.
d
Deliver to incinerate in high temperature furnace in the cement industry
(Kumjadpai et al., 2011).
The bold values are the summation of estimated cost of each step.
by the anode oxidation at high current densities can react with the
dissolved oxygen in the wastewater leading to an increase in the
amount of oxygen required by the microorganisms to degrade
organic materials present in the wastewater (Kortangsakul and
Hunsom, 2009).
The analysis of the likely operating costs of biodiesel wastewater
treatment by the chemical and electrochemical techniques were
estimated, and are summarized in Table 2. In this analysis, it was
assumed that the removed FAME/FFA phase would be re-used as
a raw material for feeding to the biodiesel production, and so its
value was credited accordingly. For the first treatment stage, the
saving cost obtained from the value of the recovered FAME/FFA was
around 12.42 USD/m3, greater than the expense (cost of H2SO4) by
around 0.35 USD/m3. With respect to the second treatment stage,
the coagulation process consumed a lower level of electricity usage
but had a higher sludge production cost compared with that of the
electro-coagulation process. Nevertheless, the coagulation process
required an operating cost of 1.07 USD/m3, some 1.62-fold lower
than that of the electro-coagulation technique. Overall, for the
whole treatment system, the estimated total expenses of both
investigated treatment process were lower than the cost of
conventional procedures, where the waste is incinerated in a high
temperature furnace of cement industry (60.35 USD/m3), by at least
54.37- and 33.9-fold for the coagulation and electro-coagulation
methods, respectively. The characteristics of the wastewater after
treatment by both investigated processes are summarized in
Table 1. All pollutants were markedly reduced but they were still
higher than the levels permitted by the Thai government for
discharge to environment. This is likely to be because the original
wastewater contained a very high concentration of pollutants.
Should this, however, remain the case then post-treatment of the
wastewater after the treatments proposed above should be carried
out, for example by biological treatment in sedimentation ponds
and pH adjustment.
4. Conclusions
The remediation of biodiesel wastewater was carried out at
a laboratory scale at ambient temperature. Initially, the FAME/FFA
were removed from the wastewater by acid protonation using
H2SO4 to a final pH of 2.5, yielding greater than 24.5 ml/l of FAME/
FFA. The attained wastewater discharged from the first step was
Expense then treated by chemical- and electro-coagulation. Both processes
(USDa/m3) were effective forcing oil & grease removal, reducing its concen-
tration in the treated wastewater to around 105 and 80 mg/l,
12.07
respectively. The chemical coagulation process was more effective
at reducing the BOD levels than that of the electro-coagulation
process, and it also provided a lower operating cost compared
with that of electro-coagulation. However, the principle drawbacks
of the coagulation process, the requirement of a large treatment
0.03
0.43
area and the contamination of chemical coagulants in the treated
1.07 wastewater, remain. Therefore, the operation should be balanced
between its treatment efficiency and operational requirements.
0.843
0.856
0.038 Acknowledgments
1.737
0.39
The authors would like to acknowledge the financial support
60.35
from the Thailand Research Fund, under the Industrial Projects for
Center for Petroleum, Petrochemicals and Advanced Materials for
facility support, and the Bangchak Petroleum Co., Ltd., for samples
and materials. Also, we thank the Publication Counseling Unit
(PCU) of the Faculty of Science, Chulalongkorn University, and Dr.
Robert D. J. Butcher for comments, suggestions and checking the
grammar.
References
APHA, AWWA, WEF, 1998. Standard Methods for the Examination of Water and
Wastewater, twentieth ed.
Berrios, M., Skelton, R.L., 2008. Comparison of purification methods of biodiesel.
Chem. Eng. J. 144, 459e465.
Canakci, M., Gerpen, J.V., 2001. Biodiesel production from oils and fats with high
free fatty acids. Trans. ASAE 44 (6), 1429e1436.
Chavalparit, O., Ongwandee, M., 2009. Optimizing electrocoagulation process for
the treatment of biodiesel wastewater using response surface methodology.
J. Environ. Sci. 21, 1491e1496.
Chen, G., 2004. Electrochemical technologies in waste water treatment. Sep. Purif.
Technol. 38, 11e41.
Davis, M.L., Cornwell, D.A., 1998. Introduction to Environmental Engineering, third
ed. McGraw-Hill, Singapore.
Jaruwat, P., Kongjao, S., Hunsom, M., 2010. Management of biodiesel wastewater by
the combined processes of chemical recovery and electrochemical treatment.
Energy Convers. Manag. 51, 531e537.
Karmee, S.K., Chadha, A., 2005. Preparation of biodiesel from crude oil of Pongamia
pinnata. Bioresour. Technol. 96, 1425e1429.
Knothe, G., 2000. Monitoring a progressing transesterification reaction by fiber-
optic near infrared spectroscopy with correlation to 1H nuclear magnetic
resonance spectroscopy. J. Am. Oil Chem. Soc. 77 (5), 489e493.
Kortangsakul, S., Hunsom, M., 2009. The optimization of the photo-oxidation
parameters to remediate wastewater from the textile dyeing industry in
a continuous stirred tank reactor. Korean J. Chem. Eng. 26 (6), 1637e1644.
Kumjadpai, S., Ngamlerdpokin, K., Chatanon, P., Lertsathitphongs, P., Hunsom, M.,
2011. Management of fatty acid methyl ester (FAME) wastewater by a combined
two stage chemical recovery and coagulation process. Can. J. Chem. Eng. 89,
369e376.
Miyake, Y., Yokomizo, K., Matsuzaki, N., 1998. Determination of unsaturated fatty
acid composition by high-resolution nuclear magnetic resonance spectroscopy.
J. Am. Oil Chem. Soc. 75, 1091e1094.
Mollah, M.Y.A., Morkovsky, P.J.A., Gomes, G., Kesmez, M., Parga, J., Cocke, D.L., 2004.
Fundamentals, present and future perspective of electrocoagulation. J. Hazard.
Mater. 114 (1e3), 199e210.
Siles, J.A., Martín, M.A., Chica, A.F., Martín, A., 2010. Anaerobic co-digestion of
glycerol and wastewater derived from biodiesel manufacturing. Bioresour.
Technol. 101 (16), 6315e6321.
Suehara, K., Kawamoto, Y., Fujii, E., Coda, J., Nakano, Y., Yano, T., 2005. Biological
treatment of wastewater discharged from biodiesel fuel production plant with
alkali-catalyzed transesterification. J. Biosci. Bioeng. 100, 437e442.
Yeung, D.K.W., Lam, S.L., Griffith, J.F., Chan, A.B.W., Chen, Z., Tsang, P.H., Leung, P.C.,
2008. Analysis of bone marrow fatty acid composition using high-resolution
proton NMR spectroscopy. Chem. Phys. Lipids 151, 103e109.
Zhou, Y., Liang, Z., Wang, Y., 2008. Decolorization and COD removal of secondary
yeast wastewater effluents by coagulation using aluminum sulfate. Desalination
225, 301e311.