Energy Conversion and Management 103 (2015) 111–118

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Municipal sewage sludge to biodiesel by simultaneous extraction

and conversion of lipids
J.A. Melero ⇑, R. Sánchez-Vázquez, I.A. Vasiliadou, F. Martínez Castillejo, L.F. Bautista, J. Iglesias,
G. Morales, R. Molina
School of Experimental Sciences and Technology (ESCET), Universidad Rey Juan Carlos, C/Tulipán s/n, Móstoles, E28933 Madrid, Spain

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

Article history: Two different approaches have been investigated for the production of biodiesel from glycerides and free
Received 16 March 2015 fatty acids (FFAs) extracted from sewage sludge. The first one is a two-step process consisting of organic
Accepted 16 June 2015 solvent extraction followed by acid-catalyzed esterification/transesterification of the isolated oil fraction.
The second one is a one-step direct transformation consisting of the simultaneous extraction and conver-
sion of the lipid fraction contained in the sewage sludge. In both alternatives, a heterogeneous acid
Keywords: Zr-SBA-15 catalyst has been used. In the two-step extraction–reaction process, conversion close to 90%
Waste water sewage sludge
of the saponifiable fraction (including FFAs and glycerides) were achieved. Remarkably, the catalyst pro-
In-situ treatment
Biodiesel
vided such high conversion in the presence of high amounts of unsaponifiable matter. Furthermore, the
Heterogeneous catalyst catalyst kept its activity in successive catalytic runs in presence of this low-quality lipid fraction. In the
Zr-SBA-15 one-step direct conversion of the dried sludge, the overall weight FAME yield, based on the initial mass of
dried sewage sludge, was around 15.5 wt% for primary and 10.0 wt% for secondary sludge. In contrast,
this FAME yield was lower than 6 wt% for two-step process when processing primary sludge (being neg-
ligible for the secondary sludge). Finally, the results of this work proof the high potential of Zr-SBA-15 as
catalyst for the production of biodiesel from a low quality oleaginous feedstock such as municipal sewage
sludge.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction membranes are lipid-rich and include phospholipids, steroids and

One of the most important challenging in the conventional bio-
diesel industry is how to overcome the high costs associated to
conventional oleaginous feedstock and eliminate the competition
between biofuels and food industries for the same raw material
[1–3]. One convenient way to simultaneously overcome both
drawbacks is using an inedible, residual and hence
inexpensive-oleaginous raw material. Therefore, important efforts
are currently being applied in biodiesel production research aiming
to find new raw materials fulfilling both requirements. One of the
potential candidates is the municipal waste water treatment sew-
age sludge, which is gaining attraction around the world as a lipid
feedstock for biodiesel production. Recent research has indicated
that the sewage sludge contains significant quantity of free lipids
(mono-, di- and triglycerides, phospholipids and fatty acids), as
well as high concentrations of microorganisms. The microorgan-
isms can suppose a significant source of oils, since their cell
⇑ Corresponding author. Tel.: +34 914887399.
E-mail address: juan.melero@urjc.es (J.A. Melero).
fatty acids, mostly in the range of C10–C18 [4,5].
Waste water treatment plants continuously produce huge
amounts of sewage sludge. More than 20 million ton of dry sludge
are produced worldwide every year. This figure is expected to
increase in the near future due to the increasing urbanization
and industrialization [6]. The management of this waste currently
presents difficult environmental challenges. Its incineration results
in emissions containing dioxins and metals [7]. In the same way,
the use of the sludge for the production of compost and fertilizers
is restricted in many countries by the presence of metals [8] and
residual pharmaceuticals [9]. Therefore, there is a need to identify
sustainable solutions for the utilization of this residue. In this con-
text, the exploitation of sewage sludge for biodiesel production is a
promising alternative that would also account as waste valorisa-
tion [10], solving at the same time energy and environmental con-
cerns. Municipal sewage sludge thus appears as an alternative
non-food feedstock offering a significant potential that could help
to overcome the competition between biofuel and food industries.
However, this feedstock usually contains, together with relevant
amounts of lipids and free fatty acids, unsaponifiable matter such

http://dx.doi.org/10.1016/j.enconman.2015.06.045
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
112 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118
as hydrocarbons, carotenes, tocopherols, sterols [7]. This matter
could disrupt the activity of catalysts in biodiesel production [1].
Additionally, water can represent more than 50 wt% of the total
weight of sewage sludge, compromising the overall yield of biodie-
sel as a consequence of esterification/hydrolysis equilibrium [11].
Biodiesel production from sewage sludge has received consider-
able attention during the last years, resulting in the development
and application of different methods for extraction and esterifica
tion/transesterification of the lipid fraction of such waste feed-
stock. Nonetheless, there are still few studies focused on the use
of heterogeneous acid catalysts, which involve several important
technical benefits as compared to their homogeneous counter-
parts, to drive the production of fatty acid methyl esters (FAME)
from sewage sludge (cheaper separation processes; reduced water
effluent load, capital and energy costs; the catalyst can be reused;
there would be no neutralization products, so a higher grade of
glycerol is produced; there would be fewer inputs and less wastes).
Nevertheless, the commercial introduction of these catalysts need
still important advances to impact in a positive way in the biodie-
sel synthesis technologies (increasing of the stability of acid sites
avoiding their leaching; increasing of thermal stability; enhance-
ment of the mass transfer avoiding diffusional limitations; milder
operation conditions and increasing of the resistance to water
and other impurities).
Pokoo-Aikins et al. [11] studied the extraction of lipids from
sewage sludge using different organic solvents, obtaining lipids
yields close to 25 wt%. Pastore et al. [12] proposed a two-step
production of FAME from municipal waste water sludge using
n-hexane in acidic ambient followed by methanolysis with sulfuric
acid allowing FAME yields between 12 and 22 wt%. In the same
way, Huynh et al. [13] reported the in-situ production of FAME,
from untreated wet activated sludge under subcritical water and
methanol conditions with sulfuric acid. Additionally, Mondala
et al. [10] investigated the feasibility of using homogeneous acid
catalysts to produce biodiesel from primary and secondary sewage
sludge by in-situ transesterification process, with sulfuric acid and
hexane to improve the solubility in the reaction mixture; obtaining
FAME yields close to 15 wt% from primary sludge and 3 wt% from
secondary sludge. The in-situ approach is gaining great interest
since it eliminates the need of extraction with organic solvents
and the subsequent separation of lipids and fatty acids from the
extraction solvent prior to the esterification/transesterification
reaction [14–16].
In this context, the aim of the present study has been to evalu-
ate the catalytic behavior of the acid catalyst Zr-SBA-15 in the pro-
duction of biodiesel from waste water treatment sewage sludge. In
previous works [17–19], Zr-SBA-15 catalyst showed high activity
as well as very good stability when treating low-grade oleaginous
feedstocks containing significant amounts of free fatty acids (FFAs),
water, alkaline metals and unsaponifiable matter. Therefore, the
present work seeks to validate the promising properties of this cat-
alyst in the production of biodiesel from sewage sludge. In order to
perform this evaluation, primary and secondary sludge from a
waste water treatment plant located at Universidad Rey Juan
Carlos were explored as renewable sources of saponifiable lipids
for biodiesel. Furthermore, the two above commented approaches
for the production of biodiesel from sewage sludge, i.e. separated
extraction–reaction and in-situ processes, have been thoroughly
assessed in the present study.
2. Materials and methods
2.1. Chemicals
Methanol (synthesis grade, Scharlab) and n-hexane (purity 96%,
Scharlab) were used as received for the extraction and reaction
assays. Tetraethylorthosilicate (TEOS, Aldrich), Pluronic P123
(Aldrich) and zirconocene dichloride (Aldrich) were used for the
preparation of the catalyst Zr-SBA-15 [17].
2.2. Sludge sampling and preparation
Sewage sludge samples used in this study were collected from
the pilot-scale waste water treatment plant (WWTP) located at
Universidad Rey Juan Carlos at Móstoles, Madrid, Spain. As shown
in Fig. 1, the WWTP has two main sludge streams, as usual in waste
water treatment facilities employing an activated sludge process.
The primary sludge, which is stored in the primary tank (sampling
point 1), is a combination of floating grease and solids collected at
the top of the flotation tank. After the primary treatment, the efflu-
ent is directed to the activated sludge system for the removal of
soluble organic contaminants, denitrification and nitrification.
The activated sludge collected in the secondary sedimentation
tank, and stored in the secondary tank (sampling point 2), is com-
posed mainly of microbial cells and suspended solids produced
during the aerobic biological treatment of waste water. Finally,
both primary and secondary sludge are collected and further trea-
ted and stabilized by anaerobic digestion. The digested sludge is
finally dewatered by centrifugation and disposed. The hydraulic
retention time (HRT) of the WWTP, calculated from the flow and
the volume of the treatment tanks, varies between 3 and 4 h in a
season-dependent manner. Some of the operational parameters
of the pilot-scale WWTP measured during the monitoring period
(2012–2013) are given in Table 1.
Sludge samples tested in the present work were collected at
sampling points 1 & 2, for primary and secondary sewage sludge,
respectively. Thereafter, they were concentrated by simple settling
and, following the work by Mondala et al. [10], the supernatant
was discarded and the wet sludge were centrifuged at 3000 rpm
for 20 min. The dewatered samples were then dried in an oven at
80 °C. Finally, dried sludge were crushed into a fine powder (with
particle size ranging from 0.5 to 1.0 mm) in order to form a suffi-
ciently homogenized suspension during the extraction/transesteri
fication process.
2.3. Extraction assays
n-hexane or methanol are the extraction solvents used in this
work. In this stage, the influence of the extraction time (2.5 and
4 h) and sewage sludge to solvent ratio (10 g of dried sewage
sludge to 100 mL and 150 mL of solvent) was assessed. The resul-
tant suspension was then filtered with nylon-membrane filters
(0.45 lm) and the solvent was removed to yield the crude oleagi-
nous residue. The collected crude oil was quantified and character-
ized by means of: (i) acid value (UNE EN ISO 660:2000), (ii) fatty
acid profile (UNE EN ISO 5508:1996 & 5509:2000), (iii) unsaponifi-
able matter content following the method described by Plank and
Lorbeer [20], and (iv) glycerides and glycerol content (UNE EN ISO
14105:2003).
2.4. Synthesis of Zr-SBA-15
Zr-SBA-15 material was prepared according to a method previ-
ously described in literature [17]. In a typical synthesis, triblock
copolymer P123, used as structure directing agent, was dissolved
in hydrochloric acid (0.67 N) at room temperature. Upon dissolu-
tion, zirconocene dichloride, used as zirconium source, was added
and the resultant suspension was stirred for 3 h and heated at
40 °C. Afterwards, TEOS was added to the synthesis medium and
vigorous stirring was kept for 20 h at 40 °C. The resultant white
suspension was hydrothermally aged (130 °C) for 24 h, the solid
was recovered by filtration and air-dried overnight. Surfactant
 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118 113

Fig. 1. Scheme of the pilot-scale WWTP of Rey Juan Carlos University.

Table 1
Parameters characterization of the pilot-scale WWTP over a two-year period.

Parameter CODa (mg/L) Ammoniumb (mg/L) Nitratesc (mg/L) TSSd (mg/L) pHe
Influent pilot-plant 345.9 ± 234.9 121.8 ± 38.5 267.8 ± 84.3 0.88 ± 0.17 6.8 ± 0.3
Influent primary treatment 416.1 ± 275.8 101.3 ± 36.2 148.1 ± 94.9 0.28 ± 0.09 7.3 ± 0.2
Effluent Primary treatment/influent RBC 515.9 ± 190.0 98.9 ± 32.2 145.6 ± 79.7 0.23 ± 0.10 7.3 ± 0.5
Effluent RBC/influent disinfection 196.1 ± 187.2 24.4 ± 12.9 123.6 ± 93.5 0.02 ± 0.01 7.2 ± 0.5
Effluent pilot-plant 71.1 ± 58.2 2.0 ± 5.1 101.2 ± 60.3 0.01 ± 0.00 6.8 ± 0.6
a
Chemical oxygen demand was determined using a dichromate-reflux colorimetric method.
b
Ammonium (NH+4) was measured using an electrode equipped in Metrohm781 calibrated using standard solution of NH4Cl.
c
Nitrates (NO
3 ) were measured using a pH-meter system of Maxidirect (Lovibong).
d
Total suspended solids, the suspensions were passed through pre-weighed membrane filters (0.45 lm pore size) and then dried at 105 °C.
e
pH was measured using a pH/ion-meter (Metrohm781, Herisau/Switzerland).

removal was performed by controlled calcination in air at 450 °C was determined by ammonia TPD in a Micromeritics 2910 unit fit-
for 5 h under static conditions. ted with a TCD detector. More details about the physico-chemical
properties of this kind of catalyst can be found in literature [17].
2.5. Catalyst characterization and properties

The most relevant physicochemical properties for the 2.6. Catalytic tests: two-steps extraction–reaction and in-situ method
Zr-SBA-15 catalyst are summarized in Table 2. XRD experiments
were recorded on a Philips X́Pert diffractometer using the Cu Ka FAME production tests were accomplished in a 25 mL
line in the 2h angle range of 0.6–5.0° with a step size of 0.02° for stainless-steel autoclave (Autoclave Engineers) fitted with temper-
low angle analysis, and a step size of 0.04° in the 2h angle range ature controller, mechanical stirrer and pressure transducer for
of 5.0–50.0° for high angle analysis. Nitrogen adsorption isotherms monitoring the reaction conditions. The tests were performed tak-
were carried out at 77 K using a Micromeritics TRISTAR 3000 unit ing into account the reaction conditions previously optimized for
(outgassing conditions with nitrogen: a heating rate of 5 °C min1 the use of Zr-SBA-15 catalyst in biodiesel production from veg-
up to 90 °C kept for 30 min and thereafter increase up to 120 °C etable oils under autogenous pressure (temperature: 209 °C; stir-
and kept for 720 min). Both analyses evidenced high mesoscopic ring speed: 2000 rpm; 50:1 methanol to saponifiable matter
ordering and high surface area (531 m2/g) with a narrow pore size molar ratio; 12.5 wt% catalyst based on lipids mass) [17].
distribution around 13 nm (size enough to avoid steric hindrances For the two-steps extraction–reaction method, in a typical assay
to the mass transfer of relatively bulky substrates such as five grams of the extracted crude oil were mixed together with
tri-glycerides with and average molecular size of ca. 4 nm). methanol (50:1 methanol to saponifiable matter molar ratio) and
Zr-SBA-15 catalyst contains 9.7 wt% of zirconium determined by the catalyst (12.5 wt% catalyst based on lipids mass) in the reaction
Inductively Coupled Plasma-Optical Emission Spectroscopy using vessel. The system was hermetically closed and the temperature
a Varian Vista AX CCD Simultaneous unit. The amount of acid sites (209 °C) and stirring conditions (2000 rpm) were set up. The reac-
tion was allowed to proceed for 3 or 6 h, before cooling down in an
ice bath. The reaction suspension was then filtered using a nylon
Table 2 filter to recover the catalyst for reuse.
Physicochemical properties of the Zr-SBA-15 material.
For the in-situ method, the same temperature reaction condi-
Sample Zra H+b SBETc Vpd Dpe (Å) a0f (Å) tions were used (209 °C), but directly using dry sludge (2.5 g)
(% wt) (meq/g) (m2/g) (cm3/g) which was added to the reactor and mixed with methanol and cat-
Zr-SBA-15 9.7 0.28 531 1.41 128 136 alyst. In order to make a direct comparison, the ratio of dried
a sludge to methanol is similar to that used for the extraction step
Metal loading calculated by means of ICP–OES.
b
Acid loading calculated by NH3 temperature programmed desorption analysis. in the two-step extraction–reaction method, that mean 0.10 g
c
Specific surface area calculated by the B.E.T. method. dried sludge/mL methanol. A lower methanol quantity (0.25 g
d
Total pore volume recorded at p/p0 = 0.985. dried sludge/mL methanol) was also studied. The catalyst amount
e
Mean pore size calculated as the maximum of the B.J.H. pore sizes distribution was 12.5 wt% based on the amount of lipids potentially extracted
applying the K.J.S. correction.
f p
Unit cell size calculated as 2/( 3d100), where d100 is the Bragg’s lattice
in the dried sewage sludge determined in the extraction analysis.
parameter. Likewise, the effect of the reaction time (3 h and 6 h) was
investigated.
114 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118
The yield of the transformation of fatty acid alkyl chains into
fatty acid methyl esters was calculated from 1H NMR measure-
ments following the method described by Wahlen et al. [21].
Catalyst reusability was assessed for the two-step and in-situ pro-
cedures. For the two-step process, the catalyst after reaction was
recovered by filtration and calcined at 450 °C in air during 5 h to
remove organic adsorbed compounds on the catalyst surface. In
the in-situ procedure, the residual solid was recovered by filtration
and calcined at 450 °C in air during 5 h to remove all the organic
matter and to obtain the spent catalyst. Afterward, the recovered
catalyst was tested again under identical reaction conditions.
3. Results and discussion
3.1. Lipids extraction and characterization
The first aim of the present work was to select the best solvent
for the lipids extraction as well as to assess the composition of
extracted lipid fraction, identifying the compounds which could
exert a more relevant effect in the catalytic performance of
Zr-SBA-15 catalyst.
3.1.1. Lipids extracted from primary sewage sludge
Fig. 2 shows the amount of extracted crude oil from primary
sludge, obtained under different extraction conditions by refluxing
with n-hexane or methanol. The values represent the average of
the results obtained from three different replicas of the extraction
process. As shown, methanol achieved a significantly greater lipids
extraction yield (expressed as wt% based on the starting dry
sludge) as compared to n-hexane under each assayed extraction
conditions.
Table 3 summarizes the amount of total extracted lipids,
together with their composition in terms of unsaponifiable matter,
glycerides and FFA. The results exhibited a strong difference in the
FFA and glycerides fractions, which might be due to the different
polarity of the used solvents. Thus, methanol is able to extract a
higher fraction of the more polar FFAs, whereas n-hexane appears
as a better extraction agent for the non-polar glycerides. It must be
noted that the lipids content of the primary sewage sludge are
expected to contain a higher FFA content because of their aqueous
origin, which promotes the hydrolysis of glycerides into free fatty
acids. Hence the use of methanol allows an enhancement of the
16
Primary sewage Secondary sewage
sludge
14
Extraction Yield (wt%)
12
10
8 The fatty acid profiles of the extracted lipids from primary and
6 extracted lipid mainly contain palmitic acid (C16:0), stearic acid
4 in the relative ratios. Thus, there is a larger amount of unsaturated
2
Methanol
n-Hexane
0 acid (C20:1) (7 mol%) and eicosapentaenoic acid (C20:5) (7 mol%).
A B C D
Fig. 2. Extraction yield (wt% based on dry sludge) from primary and secondary
sludge (10 g), obtained under different extraction conditions with n-hexane or
methanol. (A) Primary sludge 2.5 h 100 mL of solvent, (B) primary sludge 4 h
100 mL of solvent, (C) primary sludge 4 h 150 mL of solvent and (D) secondary
sludge 2.5 h 100 mL of solvent.
Table 3
Total extracted lipids (expressed as wt% based on the starting dry sludge) and their
composition in terms of unsaponifiable matter, glycerides and free fatty acids
obtained from primary and secondary sludge (10 g) by two-step extraction–reaction
with methanol or n-hexane (100 mL per 10 g of dried sludge). Extraction time: 2.5 h.
Sludge type Solvent Total lipids Lipid fraction composition (wt%)
(wt%)
Glycerides FFA Unsaponifiable
matter
Primary sludge n-Hexane 7.4 36.5 17.0 46.5
Primary sludge Methanol 13.6 5.2 43.0 51.8
Secondary sludge Methanol 2.1 62.3 20.7 16.9
extraction yield. It is noteworthy the high content of the
unsaponifiable matter in the extracted lipid fraction (50 wt%).
The presence of such impurities in the sewage sludge makes the
composition of the extracted oil more complex for producing bio-
diesel using heterogeneous acid catalysts.
Additionally, in the range of study the amount of extracted
lipids seems to be independent on the time and volume of solvent
used in experiments A to C. Therefore, the extraction conditions
used in the rest of the experiments were fixed at 100 mL of solvent,
10 g of dried sludge and refluxing for 2.5 h.
In view of the results, methanol was selected for subsequent
experiments due to the high extraction yields as well as it acts as
dual esterification/transesterification agent in the acid-catalyzed
production of biodiesel.
3.1.2. Lipids extracted from secondary sewage sludge
Table 3 and Fig. 2 illustrate the comparison between primary
and secondary sludge. Although the amount of extracted lipids
from primary sludge was much higher than that from secondary
sludge, the latter gave in relative terms a greater proportion of
saponifiable matter (glycerides plus FFA) in the extracted lipids
(83.0% vs. 48.2%) when using methanol as extraction solvent. On
the other hand, n-hexane was not effective for the extraction of
secondary sludge, leading to negligible yields. Differences can be
attributed to the nature in the original feedstocks, both in terms
of composition and form in which the lipids are present.
Secondary sludge is mostly composed of microbial cells and the
extracted lipids come from phospholipids in the cell membranes
of such microorganisms [4]. Due to their nature, such lipids exhi-
bits low extractability through a conventional method like the
one herein used because they first need to be released from their
bacteria membrane [22]. On the other hand, primary sludge con-
sists of a combination of floating grease and solids, which are much
sludge
more amenable to be recovered by solvent washing [23]. For the
same reason, the fraction of FFAs is lower in these lipids than in
those from primary sludge, as such lipids have been protected from
the aqueous hydrolytic environment.
3.1.3. Fatty acid profile of extracted lipids
secondary sludge are shown in Fig. 3. It was found that both of the
(C18:0) and oleic acid (C18:1), but there are significant differences
fatty acids in the secondary sludge than in the primary one, linoleic
acid (C18:2) (10 mol%), linolenic acid (C18:3) (7 mol%), eicosaenoic
In consequence, the biodiesel produced from secondary sewage
sludge will have better cold flow properties [12]. The melting point
of biodiesel product strongly depends on chain length and degree
of unsaturation, therefore the biodiesel produced from primary
sludge will have unfavourable cold temperature behavior [24].
These results are in agreement with those reported by Mondala
 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118 115

60 100 % FFA
100
Primary sludge % Gly
Secondary sludge
50 80 80

Molar composition (%)

(% molar composition)

FAMEYield (%)
40 60 60

30 40 40

20
20 20

10
0 0
A B C

0 Fig. 4. FAME yield and molar composition of unreacted glycerides (Gly) and free
:0

:0

:0

:1

:2

:3

:0

:1

:2

:4

:5

:0

:1

:0
fatty acids (FFA) obtained from saponifiable extracted fraction from primary sewage
14

16

18

18

18

18

20

20

20

20

20

22

22

24
sludge achieved by Zr-SBA-15 catalyst. (A) Reaction product after 3 h with fresh
C

C
Fig. 3. Fatty acid profiles of extracted lipids from primary and secondary sludge.
et al. [10], Olkiewicz et al. [23], Huynh et al. [13] and Siddiquee and
Rohani [7].
3.2. Transformation of sewage sludge into biodiesel
Both conventional two-steps extraction–reaction and in-situ
methods were applied to the acid-catalyzed production of
biodiesel over Zr-SBA-15 catalyst. Conventional two-steps
extraction–reaction method was only used for primary sludge,
since the extraction yield from secondary sludge was too low to
be practically exploited as biodiesel source (as above mentioned).
On the other hand, the in-situ approach was applied to both types
of sewage sludge.
3.2.1. Biodiesel production by the two-steps extraction–reaction
method
The starting oleaginous material used as feedstock in the reac-
tion step of this method came from the extraction step of primary
sludge using the previously selected extraction conditions. Fig. 4
shows the FAME yield and the composition of unreacted glycerides
and FFAs obtained from saponifiable extracted fraction, through
methanolysis of primary sewage sludge extracted lipid. Two exper-
iments were conducted: using a fresh sample of catalyst (Fig. 4,
catalyst; (B) reaction product after 6 h with fresh catalyst; and (C) reaction product
after 3 h with used and regenerated catalyst. Reaction conditions: 209 °C, 2000 rpm,
50:1 methanol to saponifiable matter molar ratio, 12.5 wt% catalyst based on lipids
mass.
method with the in-situ method; (ii) decrease the amount of
required methanol and (iii) study the effect of reaction time.
Fig. 5 shows the FAME yield and the composition of unreacted
glycerides and FFAs obtained from saponifiable extracted fraction
achieved by Zr-SBA-15 catalyst in the in-situ method. Zr-SBA-15
provided again very good catalytic activity in terms of transesterifi
cation/esterification to give FAME from the FFA and glycerides frac-
tions, regardless of the presence of a large portion of
non-extractable solid residue in the reaction medium. Thus, 92%
of the extractable saponifiable fraction in the primary sewage
sludge was converted into FAME after 3 h (Fig. 5, experiment A).
Note that this reaction was conducted using a ratio of dried sludge
to methanol similar to that used for the methanol extraction step
in the conventional two-step extraction–reaction method (0.1 g
dried sludge/mL methanol) in order to make a direct comparison.
On the other hand, a higher ratio (i.e. less alcohol) of sewage sludge
100
100
% FFA
% Gly
80 80

Molar composition (%)

experiment A) and reusing the catalyst sample in a second consec-
FAMEYield (%)

utive run with intermediate regeneration by thermal treatment at

60 60
450 °C in air (Fig. 4, experiment C). The catalytic activity displayed
by Zr-SBA-15 was outstanding in both a first and a second use
(Fig. 4, experiments A and C), converting almost 90% of the 40 40
saponifiable extracted fraction (free fatty acids and glycerides) into
FAME after 3 h of reaction. Although a small amount of FFA and
20 20
glycerides remained unconverted (Fig. 4, experiment A), the final
FAME yield was almost the same achieved after 6 h of reaction
(Fig. 4, experiment B). 0 0
Blank A B C D E

3.2.2. Biodiesel production by the in-situ method
Primary sludge was also used to study the efficiency of the
in-situ method (simultaneous extraction and FAME production in
the presence of methanol). In this case, dried primary sludge was
used directly as reaction substrate with no pre-treatment. As in
the conventional two-step extraction–reaction method, the perfor-
mance of the Zr-SBA-15 catalyst and the capability of its reuse
were tested.
Three sets of catalytic experiments were conducted in order to:
(i) compare the conventional two-steps extraction–reaction
Fig. 5. FAME yield and molar composition of unreacted glycerides (Gly) and free
fatty acids (FFA) obtained from saponifiable extracted fraction by Zr-SBA-15
catalyst in the in-situ transformation of primary sewage sludge. (Blank) Reaction
product after 3 h without catalyst and dry sludge to methanol ratio of 0.25 g/mL. (A)
Reaction product after 3 h with fresh catalyst and a dry sludge to methanol ratio of
0.1 g/mL; (B) reaction product after 3 h with fresh catalyst and a dry sludge to
methanol ratio of 0.25 g/mL; (C) reaction product after 6 h with fresh catalyst an a
dry sludge to methanol ratio of 0.25 g/mL; (D) reaction product after 3 h with
recovered and regenerated catalyst an a dry sludge to methanol ratio of 0.25 g/mL;
and (E) reaction product after 3 h with fresh catalyst and wet sludge (0.25 g sludge
in dry-basis/mL methanol); reaction conditions: 209 °C, 2000 rpm, 12.5 wt%
catalyst based on lipids mass potentially extracted in the dry sewage sludge.
116 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118
to solvent (0.25 g/mL) was studied in order to evaluate the effect
on the in-situ production of FAME. Far from negatively affecting
the FAME yield, this experiment resulted in a slight increase in
the production of FAME, totalizing 95.0 wt% of the extractable
saponifiable fraction (Fig. 5, experiment B), which can be ascribed
to a better contact between catalyst and saponifiable substrates in
a less diluted system. After reaction, the catalyst was recovered as
described in Experimental Section and used again under the same
reaction conditions. The reusability of the catalyst was thus con-
firmed using the ratio of 0.25 g dried sewage sludge/mL methanol,
leading to a similar FAME yield (95.0%) as compared with the first
use (Fig. 5, experiment D vs. Fig. 5, experiment B). Finally, in order
to further study the activity of the catalyst, reaction time was
increased up to 6 h. The results showed a slight increase in the
amount of FAME produced, though too small to compensate the
longer time (Fig. 5, experiment C). A blank run (no catalyst) has
also been performed in the same conditions in order to determine
the extension of thermally-driven reactions (Fig. 5). This blank
gave 46.9% yield to FAME of the extractable saponifiable fraction,
which can be mainly attributed to the thermal esterification of free
fatty acids present in the sludge [25].
Since the tolerance of the catalyst to the presence of water is a
key point for the potential exploitation of sewage sludge as biodie-
sel feedstock, and bearing in mind that this kind of waste material
usually contain high water loadings, our study has been completed
by treating a sample of wet primary sludge (72 wt% water content)
by the in-situ procedure. In a remarked result, the FAME produc-
tion remained very similar to that obtained with dried sludge even
though the ratio of dry sludge to methanol is higher (Fig. 5, exper-
iment E). The relative low acid strength of the active sites of
Zr-SBA-15 catalyst do not promote the hydrolysis of formed
methyl esters in this highly aqueous medium and hence keeping
high the yield to FAME [17]. These results demonstrate that, due
to the high water tolerance of the Zr-SBA-15 catalyst, the drying
pre-treatment of the primary sludge could be avoided. In such case,
there would be fewer inputs and less waste outputs associated
with the treatment of the raw material, reducing the overall pro-
duction costs.
Regarding the in-situ transformation of secondary sewage
sludge, the method was performed using the optimal operating
conditions resulting from the in-situ treatment of primary sludge
(209 °C, 0.25 g dried sludge/mL methanol, 12.5 wt% catalyst based
on lipids mass potentially extracted in the dried sewage sludge,
2000 rpm). Fig. 6 depicts the results based on dried sludge mass,
showing that a mass FAME yield of ca. 10% is obtained (almost
85% of the saponifiable fraction) after 3 h, keeping the activity in
two consecutive reactions runs with intermediate catalytic ther-
mal regeneration (not shown). Though this value is a bit lower than
that achieved with primary sludge, probably as a result of lower
extractability of lipids in these samples, it is indeed much higher
than that reported in literature using a similar secondary sewage
sludge [10]. This data confirms that Zr-SBA-15 is an excellent cat-
alyst to convert such a complex substrate into biodiesel. This result
is quite remarkable since it implies that, under the used reaction
conditions, the lipids from the microorganisms can be simultane-
ously extracted and esterified/transesterified into FAME by the
Zr-SBA-15 catalyst.
3.2.3. Product distribution: comparison of the transformation methods
Fig. 7 illustrates a comparison of the products distribution,
based on dried sludge mass, obtained applying both treatments,
the two-steps extraction–reaction (Fig. 7A) and the in-situ methods
(Fig. 7B), both of them starting from primary sludge. Solid residue
(non-extractable and non-transformable under the reaction condi-
tions) has also been included in order to have a better idea of the
efficiency of each method. The achieved yield to FAME was clearly
100
Solid Residue
FAME
Glycerides
80
Mass composition (wt%)
FFA
Unsaponifiable Matter
60
40
15.5 wt. %
20
10.0 wt. %
0
Primary Secondary
Fig. 6. Mass composition of the reaction product achieved by Zr-SBA-15 catalyst in
the in-situ transformation of primary and secondary sewage sludge based on dried
sludge mass. Reaction conditions: 209 °C, 3 h, 2000 rpm, 0.25 g dried sludge/mL
methanol, 12.5 wt% catalyst based on lipids mass potentially extracted in the dry
sewage sludge.
higher in the one-step in-situ treatment (15.5% vs. 5.8%).
Furthermore, the overall extracted fraction, comprising FAME,
glycerides, FFA and unsaponifiable matter, is likewise larger for
the in-situ method (34.9% vs. 13.5%). This can be explained by
the different temperature used in each treatment, refluxing
methanol, approx. 65 °C, in the two-step extraction–reaction
method versus 209 °C in the in-situ transformation. Moreover,
the low temperature used in the drying stage of the two-step
extraction–reaction method hinders the extraction of saponifiable
matter from the sludge [22]. Therefore, these results demonstrate
that the in-situ treatment is a more efficient approach in terms of
FAME yield for the catalytic production of biodiesel from primary
sewage sludge, potentially leading to a more cost-efficient and
environmentally sustainable process, in as much as the lipids
extraction and transformation into FAME are performed at the
same process. It must be pointed out that although a final purifica-
tion step at the end of the process is necessary to remove the solid
residue and unsaponifiable matter [12], also in the case of the
two-step extraction–reaction method, wherein there are no solid
residues in the reaction product, a purification process is needed
to remove the remaining unsaponifiable matter.
3.3. Biodiesel refining and characterization
Aside of the excellent catalytic results shown above, a crucial
issue for the future use of sewage sludge as a substrate for the pro-
duction of FAME is the quality of the biodiesel resulting therefrom.
Such biodiesel should meet the main requirements established by,
e.g., the European quality Standard UNE-EN 14214. As shown in
the previous sections, the reaction product, either from primary
or secondary sludge, is not only composed by FAME but also unre-
acted glycerides and FFA, and a remarkable unsaponifiable fraction
which could comprise wax esters, steroids, terpenoids and phar-
maceutical chemicals. Therefore, a purification treatment is
mandatory to achieve the standardized FAME purity (>96 wt%).
In a preliminary attempt aiming to explore the feasibility of such
a purification treatment, we performed a simple aqueous washing
followed by a drying step on the biodiesel obtained from the in-situ
treatment of the primary sludge. Thus, the reaction product was
washed with distilled water in a volume ratio of water to reaction
product of 2 to 1, and subsequently dried in an oven at 40 °C under
low vacuum (0.5 bar) for 12 h. After this simple washing–drying
treatment, the unsaponifiable matter decreases from 44% to 8%,
 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118
Unsaponifiable FFA 0.62%
A Maer 7.0 %
Glycerides 0.1%
FAME 5.8%
Solid residue
86.4%
Two-step process
Unsaponifiable
B Maer 14.0%
FFA 1.9%
Glycerides 3.5%
FAME 15.5%
Solid residue
65.1%
In-situ process
Fig. 7. Mass products distribution based on dried sludge mass achieved by: (A)
two-step extraction–reaction method for primary sewage sludge; (B) in-situ
method for primary sewage sludge (0.25 g sludge/mL methanol). Reaction condi-
tions: 209 °C, 3 h, 2000 rpm, 12.5 wt% catalyst based on lipids mass potentially
extracted in the dry sewage sludge.
the FFA content slightly decreases from 6% to 5% and the glycerides
content remains the same (6%), resulting in an increase of the
FAME content from 46 wt% to 80 wt%. Probably, a more rigorous
purification treatment would improve the FAME content, getting
closer to the objective of >96 wt%. Another alternative to achieve
the commercial standards would be by blending with high-purity
biodiesel from a different source. Anyway, these results show the
potential of producing biodiesel from sewage sludge following a
process heterogeneously catalyzed by Zr-SBA-15.
4. Conclusions
The results show that the catalytic activity displayed by
Zr-SBA-15 was outstanding in both the two steps extraction–reac-
tion method and the in-situ method, converting more than 90% of
the saponifiable extracted oil fraction (free fatty acids and glyc-
erides) into FAME regardless of the presence of large amounts of
solid residue and unsaponifiable matter in the raw-material.
Likewise, the in-situ treatment is more efficient than the
117
two-step extraction–reaction method in terms of FAME yield and
potentially leading to a more cost-efficient and environmentally
sustainable process. This in-situ approach avoids the use of organic
solvents for lipids extraction and allows the reduction of treatment
steps. On the other hand, Zr-SBA-15 showed a high catalytic per-
formance in the in-situ processing of wet sludge and hence the
energetic requirements for the pre-treatment of the primary
sludge might be minimized. Additionally, an increase of the
FAME content from 46 wt% to 80 wt% resulted after a simple wash-
ing–drying treatment of the reaction product coming from the
in-situ treatment of primary sludge. The catalytic results of this
work proof the high potential of Zr-SBA-15 as catalyst for the pro-
duction of biodiesel from sewage sludge.
Acknowledgements
Financial support from the Spanish Science and Innovation
Ministry through the project CTQ2008-01396 and from the
Regional Government of Madrid through the project
S2009-ENE1743 is gratefully acknowledged. The Spanish govern-
ment is kindly acknowledged by the award of a FPI grant to
RS-V. I.A. Vasiliadou thanks the Marie Curie program (action
FP7-PEOPLE-2010-IEF-273654) for a post-doctoral fellowship.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.enconman.2015.
06.045.
References
[1] Morales G, Bautista LF, Melero JA, Iglesias J, Sánchez-Vázquez R. Low-grade oils
and fats: effect of several impurities on biodiesel production over sulfonic acid
heterogeneous catalysts. Bioresour Technol 2011;102:9571–8.
[2] Chen Y, Xiao B, Chang j, Fu Y, Lv P, Wang X. Synthesis of biodiesel from waste
cooking oil using immobilized lipase in fixed bed reactor. Energy Convers
Manage 2009;50:668–73.
[3] Mahesh Sneha E, Ramanathan Anand, Meera KM, Begum S, Narayanan
Anantharaman. Biodiesel production from waste cooking oil using KBr
impregnated CaO as catalyst. Energy Convers Manage 2015;91:442–50.
[4] Karbo DM. Biodiesel production from municipal sewage sludge. Energy Fuels
2010;24:2791–4.
[5] Olkiewicz M, Fortuny A, Stüber F, Fabregat A, Font J, Bengoa C. Effects of pre-
treatment on the lipid extraction and biodiesel production from WWTP sludge.
Fuel 2015;141:250–7.
[6] Iglesias J, Morales G. Biodiesel from waste oils and fats: methods and catalysis.
In: Luque R, Melero JA, editors. Advances in biodiesel preparation. Second
generation processes and technologies. Cambridge: Woodhead Publisher Ld.;
2012. p. 154–78.
[7] Siddiquee MN, Rohani S. Lipid extraction and biodiesel production from
municipal sewage sluges: a review. Renew Sustain Energy Rev
2011;15:1067–73.
[8] Smith SR. A critical review of the bioavailability and impacts of heavy metals in
municipal solid waste composts compared to sewage sludge. Environ Int
2009;35:142–56.
[9] Motoyama M, Nakawaga S, Tanoues R, Sato Y, Nomiyama K, Shinohara R.
Residues of pharmaceutical products in recycled organic manure produced
from sewage sludge and solid waste from livestock and relationship to their
fermentation level. Chemosphere 2011;84:432–8.
[10] Mondala A, Liang K, Toghiani H, Hernandez R, French T. Biodiesel production
by in situ transesterification of municipal primary and secondary sludges.
Bioresour Technol 2009;100:1203–10.
[11] Pokoo-Aikins G, Heath A, Mentzer RA, Mannan MS, Rogers WJ, El-Halwagi MM.
A multi-criteria approach to screening alternatives for converting sewage
sludge to biodiesel. J Loss Prev Proc Ind 2010;23:412–20.
[12] Pastore C, Lopez A, Lotito V, Mascolo G. Biodiesel from dewatered wastewater
sludge: a two-step process for a more advantageous production. Chemosphere
2013;92:667–73.
[13] Huynh LH, Nguyen PLT, Ho QP, Ju Y. Catalyst free fatty acid methyl ester
production from wet activated sludge under subcritical water and methanol
conditions. Bioresour Technol 2012;123:112–6.
[14] Choi OK, Song JS, Cha DK, Lee JW. Biodiesel production from wet municipal
sludge: evaluation of in situ transesterification using xylene as a cosolvent.
Bioresour Technol 2014;166:51–6.
118 J.A. Melero et al. / Energy Conversion and Management 103 (2015) 111–118
[15] Gunuawan F, Kurniawan A, Gunuawan I, Ju Y-H, Ayucitra A, Soetaredjo FE,
et al. Synthesis of biodiesel from vegetable oils wastewater sludge by in-situ
subcritical methanol transesterification: process evaluation and optimization.
Biomass Bioenergy 2015;69:28–38.
[16] Sangaletti-Gerhard N, Cea M, Risco V, Navia R. In situ biodiesel production
from greasy sewage sludge using acid and enzymatic catalysts. Bioresour
Technol 2015;179:63–70.
[17] Melero JA, Bautista LF, Iglesias J, Morales G, Sánchez-Vázquez R. Zr-SBA-15
acid catalyst: optimization of the synthesis and reaction conditions for
biodiesel production from low-grade oils and fats. Catal Today
2012;195:45–53.
[18] Iglesias J, Melero JA, Bautista LF, Morales G, Sanchez-Vazquez, Andreola MT,
et al. Zr-SBA-15 as an efficient acid catalyst for FAME production from crude
palm oil. Catal Today 2011;167:46–55.
[19] Iglesias J, Melero JA, Bautista LF, Morales G, Sanchez-Vazquez. Continuous
production of biodiesel from low grade feedstock in presence of Zr-SBA-15:
catalyst performance and resistance against deactivation. Catal Today
2014;234:174–81.
[20] Plank C, Lorbeer E. Minor components in vegetable oil methyl esters: sterols in
rape seed oil methyl ester. Fat Sci Technol 1994;96(10):379–86.
[21] Wahlen BD, Barney BM, Seefeldt LC. Synthesis of biodiesel from mixed
feedstocks and longer chain alcohols using an acid-catalyzed method. Energy
Fuel 2008;22:4223–8.
[22] Olkiewicz M, Caporgno MP, Fortuny A, Stüber F, Fabregat A, Font J, et al. Direct
liquid–liquid extraction of lipid from municipal sewage sludge for biodiesel
production. Fuel Process Technol 2014;128:331–8.
[23] Olkiewicz M, Fortuny A, Stüber F, Fabregat A, Font J, Bengoa Ch. Evaluation of
different sludges from WWTP as a potential source for biodiesel production.
Procedia Eng 2012;42:634–43.
[24] Carrero A, Pérez A. Advances in biodiesel quality control, characterization and
standards developments. In: Luque R, Melero JA, editors. Advances in biodiesel
preparation. Second generation processes and technologies, vol.
5. Cambridge: Woodhead Publisher Ld.; 2012. p. 91–130.
[25] Pasias SA, Barakos NK, Papayannakos NG. Catalytic effect of free fatty acids on
cotton seed oil thermal transesterification. Ind Eng Chem Res
2009;48:4266–73.