Renewable Energy 94 (2016) 1e9

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Impact of metals on corrosive behavior of biodieseledieseleethanol

(BDE) alternative fuel
Saravana Kannan Thangavelu a, b, Abu Saleh Ahmed c, Farid Nasir Ani a, *
a
Department of Thermo Fluids, Faculty of Mechanical Engineering, Universiti Teknologi Malaysia, UTM, 81310 Skudai, Johor D.T., Malaysia
b
Faculty of Engineering, Computing and Science, Swinburne University of Technology, 93350 Kuching, Sarawak, Malaysia
c
Department of Mechanical and Manufacturing Engineering, Faculty of Engineering, Universiti Malaysia Sarawak, Sarawak, Malaysia

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

Article history: Corrosive behavior of biodieseledieseleethanol (BDE) fuel blends upon exposure to metals, namely, mild
Received 30 July 2015 steel, copper and aluminum was studied by static immersion at room temperature and 60
C. The change
Received in revised form of fuel properties, i.e., total acid number, density, viscosity, calorific value, flash point, and color changes
26 January 2016
were investigated. Moreover, fuel compositional changes, such as water content, oxidation product level,
Accepted 3 March 2016
Available online 19 March 2016
and metal elements in the fuel blends were examined. Finally, the effect of fuel blends on corrosion rate
(CR), morphology of corrosion products, and chemical structure of metals were studied. Results revealed
that the CR of metals in BDE fuels is in the order: aluminum < mild steel < copper at both temperature
Keywords:
Biodieseledieseleethanol (BDE)
conditions. The degradation of fuel properties and corrosion rate of metals in BDE fuel blends are lower
Corrosion than neat biodiesel (B100), whereas higher than petroediesel (B0). Corrosiveness of B20D75E5 is lower
Mild steel than B20D70E10 upon exposure to metals.
Copper © 2016 Elsevier Ltd. All rights reserved.
Aluminum

1. Introduction
The increasing demands of petroleum-based fuels, due to the
rapid development of industry and automotive society, coupled
with the environmental pollution issues (greenhouse gas emis-
sions) have motivated the efforts on discovering new alternative
fuels. Biofuels, especially ethanol/bioethanol and biodiesel have
gained progressive importance as alternative fuels for internal
combustion engines [1]. The blends of diesel and ethanol could be
used in existing diesel engines without engine modification, but
the major drawback in dieseleethanol (DE) fuel blends is, that
ethanol is immiscible in diesel over a wide temperatures and water
content, because of their chemical structure and characteristics.
These can results in fuel instability due to phase separation; how-
ever, biodiesel is successfully added to DE blends to prevent the
phase separation and instability [2]. Biodiesel is highly miscible in
both diesel and ethanol; moreover, can act as an emulsifier for DE
blend to form biodieseledieseleethanol (BDE) blend, which can be
used in diesel engines. The addition of biodiesel in DE blends
dramatically improves the solubility of ethanol in diesel over a wide
range of temperature, and the BDE fuel blends are stable well below
* Corresponding author.
E-mail address: farid@mail.fkm.utm.my (F.N. Ani).
zero temperature [3]. As reported by Shahir et al. [3] in their recent
review, a maximum of 30% biodiesel (ethyl or methyl ester) and
ethanol blend can be added to diesel fuel effectively. In addition,
most of the researchers recommend that a maximum of 20% (Vol.)
biodiesel and 10% (Vol.) ethanol can be used in diesel engine for
better engine performance and emissions [4e7].
Biodieseledieseleethanol (BDE) blend represent an important
alternative fuel for diesel engines; however, changes in the fuel
composition and the introduction of new alternative fuel often
results in corrosion and degradation of the automobile fuel system
parts. The presence of water, organic acids, aldehydes, peroxides,
ketones, and esters in oxygenated fuel causes corrosion in fuel
system materials; in addition, degrades the properties of fuel [8].
The diesel engine parts made from mild steel, such as fuel tank, fuel
lines, and fuel tube outlet; copper, such as, fuel tank gasket, washer,
and bushing; and aluminum, such as fuel pump, fuel filters, and fuel
feed pump are normally affected by the fuel blends [9].
Many researchers have investigated the corrosion behavior of
metallic materials, namely, mild steel, copper, aluminum, stainless
steel, brass, magnesium and cast iron in different biodiesel, such as
palm oil, rapeseed oil and sunflower oil [10e19]. Literature show a
gap that there is no study reported on corrosion behavior of metals,
especially, mild steel and aluminum in biodieseledieseleethanol
(BDE) fuel blend, and the corrosive nature of BDE fuel on metallic

http://dx.doi.org/10.1016/j.renene.2016.03.015
0960-1481/© 2016 Elsevier Ltd. All rights reserved.
2 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9
materials. In a previous study, we have done the preliminary
investigation on corrosion behavior of copper in B45D35E20 (45%
biodiesel, 35% diesel and 20% ethanol) at room temperature and
50
C, and the results showed that the corrosion of copper in
B45D35E20 at 50
C is comparatively lower than neat biodiesel
(B100), whereas it was higher than that of diesel (B0) [20]. In
another previous study, we have investigated the impact of
B20D75E5 (20% biodiesel, 75% diesel and 5% ethanol) on degrada-
tion behavior of elastomers, such as nitrile rubber (NBR) and pol-
ytetrafluroethylene (PTFE), and it was found that degradation of
NBR is high compared to PTFE in B20D75E5 [21].
The objective of this study is to investigate the impact of metals
(mild steel, copper and aluminum) on degradation of BDE fuel
blends, such as B20D75E5 (20% biodiesel, 75% diesel and 5%
ethanol) and B20D70E10 (20% biodiesel, 70% diesel and 10%
ethanol). In addition, the effect of these blends (B20D75E5 and
B20D70E10) on corrosion behavior of metals was also investigated.
2. Materials and methods
The palm biodiesel (98.8% ester content) was procured from
Senari Biofuels Sdn. Bhd, Sarawak. As described by the supplier, the
impurities in the palm biodiesel were methanol (~0.1%), free glyc-
erol (~0.015%), monoglyceride (~0.32), diglyceride (~0.1%), glycerol
(~0.18%), triglyceride (~0.08%) and water (~310 ppm). All the
specifications were below the standard limit mentioned by Euro-
pean biodiesel standard (EN 14214). The acid value (Total acid
number) of received the palm biodiesel was ~0.25 mg KOH/g. The
anhydrous ethanol (99.5% pure) was acquired from SigmaeAldrich
and the petroediesel was procured from local petrol station at
Kuching, Sarawak (Shell Malaysia). Fuels were kept in inert atmo-
spheres prior to and after use, to prevent oxidation and water ab-
sorption. Three different fuel blends, namely, B0 (100% diesel),
B20D75E5 (20% biodiesel, 75% diesel and 5% ethanol) and
B20D70E10 (20% biodiesel, 70% diesel and 10% ethanol), were
prepared for immersion test. To prevent reaction of blends with
oxygen atmosphere, they were prepared 30 min before the exper-
iment, and kept in sealed glass bottles. The properties of fuel blends
(B0, B20D75E5 and B20D70E10) were determined according to
American Society for Testing and Materials (ASTM) standards and
are reported in Table 1.
The mild steel specimen of size 40 mm length, 25 mm width and
3 mm thickness; copper specimen of size 40 mm length, 22 mm
width and 3 mm thickness; and aluminum specimen of size 40 mm
length, 24 mm width and 3 mm thickness were made from long flat
plate by machining and grinding. The density of mild steel (0.2% C,
0.4% Mn and 99.4% Fe), copper (99.99%) and aluminum (99%
commercially pure) used in this study is 7.85, 8.96 and 2.7 g/cm3,
respectively.
To suspend the specimen into test fuel blends in the glass screw
bottle, a hole of 3 mm diameter was drilled on the edge of the
specimens. Using a silk string through the hole, the specimen was
suspended into the fuel, and the test glass bottle is completely
sealed during immersion. Before immersion, the specimens were
Table 1
The properties of BDE fuel blends.
Properties B0 B100
Density 853.4 864.42
(kg/m3 at 15
C)
Kinematic viscosity 2.15 4.71
(mm2/s at 40
)
Flash point (
C) 69 122
Cetane number 51.5 62.4
Heat of combustion (MJ/kg) 45.8 39.84
polished using abrasive (SiC) grading papers (grades 180 to 800)
until removing the scratches produced during cutting and grinding,
and thereafter washed and degreased with acetone. Finally, the
specimens were rinsed with deionized water, and then oven dried
at 60
C for 15 min.
Static immersion test was conducted for the metal specimens in
B0, B20D75E5 and B20D70E10 at room temperature (25e30
C) for
800 h and 60
C for 400 h. The weight of the specimens before and
after the immersion test was measured using balance with accuracy
of four decimal points. Immersion tests for triplicate specimens
(three replicates) in each of the fuel blends were also conducted.
The weight loss was calculated by deducting weight of the spec-
imen after immersion from its weight before immersion. At the end
of the immersion test, corrosion behaviors of metals were studied
by investigating corrosion rate (CR), morphology of corrosion
products and chemical structure of metals. The mean of weight loss
measured from triplicate test specimens was converted into
corrosion rate using Eq. (1) [10e21].
Corrosion rate (CR) ¼ W  534/D  T  A (1)
Where corrosion rate ‘mpy’ stands for mils (0.001 inch) per year, W
is the mean weight loss (mg), D is the density (g/cm3), T is the
exposure time (h), and A is the exposed surface area (square inch).
The morphology of corrosion products was characterized by optical
microscope (OM). Fourier transform infrared spectroscopy (FTIR)
was used to investigate the chemical structure of metals, before and
after immersion. FTIR spectrum was recorded between 4000 and
400 cm1 using Shimadzu spectrometer (Shimadzu, Japan) at
1 cm1 resolution and 10 scans per sample.
The degradation of fuel blends were studied by investigating the
change of fuel properties, such as total acid number, density, vis-
cosity, calorific value, flash point, color changes, water content,
oxidation products, and the metal elements in the fuel. Total acid
number (TAN) was determined using standard 0.1 M KOH solution
(potassium hydroxide) by Potentiometric Titration method ac-
cording to ASTM D664. A digital stabinger viscometer (SVM 3000,
Anton Paar) was used to find the density at 15
C (according to
ASTM D7042) and kinematic viscosity (according to ASTM D445) at
40
C. FTIR spectroscopy was used to investigate the oxidation
products (carbonyl groups) in the fuel blends according to ASTM
D7414. Water content in the fuel was measured by standard test
method according to ASTM E203, using volumetric Karl Fischer
titration (Mettler Toledo-Karl Fischer Compact Titrator). Heat of
combustion (calorific value) of the blends was determined by EKA
2000 Bomb Calorimeter according to ASTM D240. Flash point was
measured by Pensky-Martens Closed cup apparatus according to
ASTM D93. Metal elements in the fuel blends were measured using
inductively coupled plasma mass (ICP-MS) spectrometer (Perki-
nElmer Elan 9000) according to the method described by Sanchez
et al. [23]. A strong concentrated nitric acid was used for acid
digestion to prepare the sample in ionic form prior to entering the
mass analyzer (ICP-MS) in order to be detected the metals. Color
changes in the fuel blends were studied by visual inspection.
B20D75E5 B20D70E10 Method
851.9 846.8 ASTM D7042
2.756 2.4796 ASTM D445
18 16 ASTM D93
53.7 50.4 ASTM D613
41.124 40.524 ASTM D240
 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9
3. Results and discussion
3.1. Corrosion rate
The corrosion rate (CR) of metals exposed to different fuel
blends is shown in Fig. 1, which indicates the corrosion of metals in
BDE fuels is higher than that of B0 (diesel) at both temperature
conditions (room temperature and 60
C). A slight variation in the
CR of metals among the BDE blends at 60
C, whereas at room
temperature, the CR in B20D70E10 is slightly higher than that of
B20D75E5. The CR of metals in the BDE fuel blends is in the order:
aluminum < mild steel < copper at both temperature conditions,
while the observed CR of metals at 60
C is comparatively higher
than at room temperature. This is in good agreement with the
literature where it has been reported that metals could be sus-
ceptible to enhanced corrosion as a result of higher temperature
[11]. A comparison on CR of metals obtained in the present study
with previous studies is summarized in Table 2.
At room temperature, the CR of mild steel in the present study
was 0.1572 and 0.1817 mpy in B20D75E5 and B20D70E10, respec-
tively, which is higher than B0 (0.0523 mpy). As compared to
previous studies [14,18] reported for B100 (neat biodiesel), these CR
Fig. 1. Corrosion rate of metals in different fuel blends (a) at room temperature (b) at
60
C.
3
in BDE fuel are slightly higher. According to Baena et al. [22], the
fuel containing ethanol is more susceptible corrosion when
exposed to ferrous material, which is due to occurrence of water
and oxygen in the ethanol; moreover, no passivation exists in mild
steel, and corrosion products are not protective at low temperature
atmosphere. The CR of mild steel obtained in the present study at
60
C (0.243 mpy in B20D75E5) is less than Cursaru et al. [16] and
Hu et al. [15] as reported in Table 2,whereas, Fazal et al. [18] ob-
tained only 0.059 mpy at 80
C for 1200 h in palm biodiesel. This is
due to the investigated mild steel in that study contained 0.2% C,
0.5% Mn, 0.5% Si and 98.8% Fe in composition, whereas the mild
steel used in the present study contains only 0.2% C, 0.4% Mn and
99.4% Fe. Moreover, condensation of water molecules is high at
80
C compared to 60
C in ethanol atmosphere, which slightly
reduced the reaction of water molecules to enhance corrosion in
carbon steel.
At room temperature, the CR of copper in B20D75E5 and
B20D70E10 was 0.1624 and 0.1911 mpy, respectively, which is
slightly higher than in B0 (0.1107 mpy). These BDE fuel values are
less than the previous studies [13,14,16,20] as reported for neat
biodiesel at room temperature (as in Table 2). However, Haseeb
et al. [10] obtained only 0.042 mpy at 25e30
C for 2640 h in palm
biodiesel. This is due to longer immersion time in their study
(2640 h), and the formation of oxides acted as barriers for corro-
sion. Moreover, the corrosion rate of copper was maximum limit at
approximately 600e1200 h, after which the corrosion rate drops
[14]. At 60
C, the CR of copper observed in the present study for
BDE fuel is lower than the previous studies [15e18,20], as reported
in Table 2.
At room temperature, the CR of aluminum was 0.0614 and
0.0681 mpy in B20D75E5 and B20D70E10, respectively, which is
slightly higher than in B0 (0.0515 mpy). As compared to previous
studies [13,16,19], as reported in Table 2, these values are lower. At
60
C, the CR of aluminum is 0.212 mpy and 0.218 mpy in B20D75E5
and B20D70E10, respectively. These values are comparable to Hu
et al. [15] and Fazal et al. [11]; moreover, these values are lower
than Norouzi et al. [17] and Cursaru et al. [16], as reported in Table 2.
According to Hu et al. [15], aluminum is favorable in the formation
of metal oxide films, which prevents the oxidation, and allows for
lower corrosion. The metal oxide film not only protects the
aluminum from oxygen, but also prevents the fuel sample contact
with metal surface, and thus the corrosion rates are lower.
3.2. Change of fuel properties and compositions
The change of fuel properties, such as total acid number, density,
viscosity, calorific value, flash point, and color changes, exposed to
metals are discussed in this section. In addition, the compositional
changes in the fuel blends, namely, water content, oxidation
products, and metal elements in the fuels are also described in this
section.
3.2.1. Total acid number
TAN is the measure of acid concentration in a non-aqueous so-
lution. Fig. 2 shows the total acid number (TAN) value of fuel blends,
before and after exposure to metals at room temperature and 60
C.
As shown in Fig. 2, there is a slight variation in TAN value between
the BDE blends at room temperature and 60
C, whereas theses
values are higher than B0. At room temperature, the TAN value of
B20D70E10 exposed to metals is in the order: aluminum (0.94 mg
KOH/g) < mild steel (1.15 mg KOH/g) < copper (1.33 mg KOH/g). At
60
C, the TAN value of B20D70E10 exposed to metals is in the
order: aluminum (1.02 mg KOH/g) < mild steel (1.59 mg KOH/
g) < copper (1.75 mg KOH/g). The permissible TAN limit for bio-
diesel as per ASTM standard D6751 (0.8 mg KOH/g) has been
4 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9

Table 2
Corrosion rate (mpy) of metals in biodiesel.

Reference Biodiesel (B100) MS Cu Al Temp

Haseeb et al. [10] palm oil e 0.042 e 25e30
C

Fazal et al. [13] palm oil e 0.3928 0.1731
Fazal et al. [14] palm oil 0.0688 0.94255 e
Cursaru et al. [16] sunflower oil 0.18 0.34 0.17
Fazal et al. [18] palm oil 0.052 e e
Chew et al. [19] palm oil e e 0.123
Thangavelu et al. [20] B45D35E20 e 0.277 e
This study B20D75E5 0.1572 0.1624 0.0614
Hu et al. [15] rapeseed oil 0.7167 0.9196 0.1277 43e80
C
Norouzi et al. [17] rapeseed oil e 0.92 0.35
Haseeb et al. [10] palm oil e 0.053 e
Fazal et al. [11] palm oil e 0.586 0.202
Thangavelu et al. [20] B45D35E20 e 0.327 e
Cursaru et al. [16] sunflower oil 0.32 0.64 0.33
Fazal et al. [18] palm oil 0.059 e e
This study B20D75E5 0.243 0.2892 0.212

exceeded in both the BDE blends upon exposure to metals, at both
temperature conditions. TAN value of BDE blends exposed to cop-
per is higher at both room temperature and 60
C, which indicates
that the copper particles are participants in the oxidation process,
eventually increasing the acid concentration in the fuel blends. TAN
value does not directly measure the rate of oxidation, but it merely
measures the by-products of oxidation in the fuel blends, hence the
increase of TAN value in the fuel blends is due to increase of by-
products of oxidation [13]. According to Fazal et al. [13], increase
in TAN number indicates the presence of corrosive acids, which
accelerates corrosion of metals. A detailed summary of the TAN
value obtained from the previous studies compared to the present
study is reported in Table 3. It can be inferred that the TAN values of
BDE blends exposed to metals in the present study is comparatively
lower than the previous studies reported for neat biodiesel (B100).

3.2.2. Density and viscosity

Fig. 3 (a) shows the changes in density of the fuel blends before
and after exposure to metals at room temperature. It can be noticed
that all the BDE fuel blends exposed to metals have increased in
density, which is due to conversion of metal elements into the fuel
blends at oxygen atmosphere [18]. Among them, copper exposed
B20D70E10 blend has higher increase in density (2.3%) compared
to other fuel blends, exposed to mild steel and aluminum. This is
due to greater oxidation of copper element into the fuel, as evi-
denced in ICP-MS test, and reported in Table 4. However, it is
noticeable that each exposed BDE fuel blends show a density value
within the limit for biodiesel (0.86e0.9 g/cc) as stated by specific
standard EN 14214. Moreover, as shown in Fig. 3 (b), there is slight
variation in the viscosity of BDE fuel blends exposed to metals,
which is noticeable in copper exposed fuels. This indicates copper
particles have greater influence on degradation of biofuels [14].
However, these values are within the permissible limit for biodiesel
(2e6 mm2/s) as specified by ASTM D6751.

3.2.3. Calorific value and flash point

Fig. 2. Total acid number (TAN) of fuel blends upon exposure to metals (a) at room
temperature (b) at 60
C.
The initial calorific value (CV) of B20D75E5 and B20D70E10 was
41.124 MJ/kg and 40.524 MJ/kg, respectively. These values increased
to 44.98 MJ/kg and 44.71 MJ/kg after exposure to copper, whereas
increased to 43.35 MJ/kg and 43.45 MJ/kg after exposure to mild
steel, respectively (as shown in Fig. 4 (a)). Conversely, there is no big
change in the CV of diesel (B0) before and after exposure to metals.
The main reason for increased in CV for BDE fuel is due to change in
the composition of BDE fuel (reduced oxygen content) when
exposed to metals. Fazal et al. [14] reported that the calorific value
of biodiesel increases in the presence of metal elements, and
changes in calorific values may depend on the types of molecules
formed after hydrolysis of esters. There is no change in the flash
point of B0 and B20D75E5 fuel exposed to metals, whereas, there is
a slight change in the flash point (from 16 to 17
C) of B20D70E10
 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9 5

Table 3
TAN value (mg KOH/g) of biodiesel exposed to metals.

Reference Biodiesel (B100) TAN as received TAN after exposed Temp

MS Cu Al

Haseeb et al. [10] palm oil 0.25 e 1.125 e 25e30
C

Fazal et al. [12] palm oil 0.35 e 2.6 e
Fazal et al. [13] palm oil 0.35 e 2.57 1.68
Fazal et al. [14] palm oil 0.25 2.75 3.28 e
This study B20D70E10 0.25 1.15 1.33 0.94
Norouzi et al. [17] rapeseed oil 0.23 e 2.75 1.75 50
e80
C
Fazal et al. [18] palm oil 0.35 1.95 e e
Fazal et al. [11] palm oil 0.35 e 1.8 1.5
This study B20D70E10 0.25 1.59 1.75 1.02

Fig. 3. (a) Density and (b) Kinematic viscosity of fuel blends exposed to metals at room
temperature for 800 h. Fig. 4. (a) Calorific value and (b) Flash point of fuel blends exposed to metals at room
temperature for 800 h.

Table 4
Concentration of metal elements (ppm) in B20D70E10 and B0. exposed to mild steel and copper as shown in Fig. 4 (b).
Elements Blank fuel MS Al Cu B0 (Cu)

Fe 0 6.1 0 0 0
3.2.4. Oxidation products
Cu 0 0 0 10.3 1.8
Al 0 0 4.7 0 0 Fig. 5 (a) shows the oxidation product level (carbonyl groups) in
fuel blends. The unit used for reporting oxidation product level is
absorbance per 0.1 mm (Abs./0.1 mm), which relates directly to the
6 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9
Fig. 5. (a) Oxidation product level and (b) Water content in the fuel blends before and
after the immersion tests at room temperature for 800 h.
peak intensities observed in the FTIR spectrum. The absorption of
the carbonyl groups in oxidation was noticed from FTIR spectrum
between 1800 and 1670 cm1 according to ASTM D7414. The
oxidation products level in both BDE fuels increased upon exposure
to metals, which is high in copper exposed BDE blends, whereas in
B0, a slight variation was noticed. The main ethanol oxidation
products are ethanol acetaldehyde, ketone, carbonic acid and acetic
acid. The oxidation of biodiesel can produce different types of acids,
such as octanoic acid, myristic acid, palmitic acid etc. In addition to
Fig. 6. Color changes of B20D70E10 fuel blend upon exposure to metals.
these acids, different types of other short chain esters, aldehydes,
and ketones are found to be produced. Apart from increasing the
corrosiveness of biodiesel, these oxidation products may also
degrade the fuel properties by reacting with esters, and thereby
increasing total acid number [18,19].
3.2.5. Water content
Fig. 5 (b) shows the water content in the received fuels and in
the fuel after immersion tests at room temperature. Water content
is the main component to make the fuel more aggressive. The
received B20D75E5 and B20D70E10 blends had low water content
(0.011 and 0.015 vol. %, respectively), as shown in Fig. 5 (b); how-
ever, the water content increased after immersion, which indicates,
BDE fuel blends absorbed more water compared to B0. The
maximum permissible limit of water content is 0.05 (vol. %) for both
diesel and biodiesel according to ASTM standard D975 and D6751,
respectively. In this present study, water content in BDE blends
exposed to metals exceeded the maximum limit. The copper
exposed B20E70E10 fuel blend contains maximum water content of
0.18 (vol. %). However, the maximum acceptable moisture content
for fuel grade ethanol as stated by ASTM E203 is 1%. In summary,
the percentage of water content obtained in the present study is
lower than previous studies reported for B100 [10,11,14,18].
3.2.6. Color changes
Fig. 6 shows the changes in color of B20D70E10 blend upon
exposure to metals at room temperature (25e30
C) and 60
C. The
copper exposed fuel blends show a comprehensive change in color
(pale green color) compared to as received state (light cloudy yel-
low). These results suggest that for copper and copper based alloy,
copper carbonate (CuCO3) of pale green color is the dominant
compound formed in biodiesel [12]. The color variation is an indi-
cator of the presence of copper elements in the fuels, as evidenced
by ICP-MS test, and reported in Table 4. Aluminum exposed BDE
blend does not show any color change on its surface, whereas mild
steel exposed BDE fuel blend changed to dark reddish yellow color
at 60
C as in Fig. 6, which is due to the formation of iron oxides
(Fe3O3).
3.3. Morphology and chemical structure
3.3.1. Morphology of corrosion products
Fig. 7 shows the morphology of metal corrosion products in the
metal surface exposed to fuel blends at room temperature
(25e30
C) and 60
C. There are surface damages due to corrosion
on the surface of metals exposed to BDE blends, compared to the
surface of metals as originally received (as shown in Fig. 7). In both
BDE fuel blends, the surface damages are almost similar in all three
metals at room temperature. At 60
C, the surface damages are
more in mild steel and copper compared to aluminum. A protective
 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9 7

Fig. 7. Optical photograph (100  ) showing the morphology of corrosion products in the surface of metals after immersion at room temperature and 60
C.

layer covered the aluminum surface, which protected the
aluminum from corrosion.
After immersion test, the formation of small pits on the surface
of copper in black color was found at room temperature. At 60
C,
the size of pits increased, which indicates the copper has corroded
due to formation of oxygen rich cupric oxide (CuO) of black color as
the consequence of higher dissolved oxygen in biodiesel [10], and
this is higher in B20D70E10 as shown in Fig. 7. According to Fazal
et al. [11], copper in oxygen atmosphere forms oxygen rich CuO/
CuCO3 (outer layer) followed by Cu2O (inner layer). On BDE exposed
copper surface, pits are more likely to form by replacing oxygen
ions from Cu2O through destruction of CuO layer from copper

Fig. 8. FTIR spectra of copper exposure to B20D70E10 fuel blend (A) before exposure; (B) after exposure at room temperature; (C) after exposure at 60
C.
8 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9
Fig. 9. FTIR spectra of mild steel upon exposure to B20D70E10 fuel blend (A) before exposure; (B) after exposure at room temperature; (C) after exposure at 60
C.
surface. The higher percentage of carbon (black color) and oxygen
are available on the oxide layer as observed. The concentration of
oxygen increases with the increase of immersion time. The pres-
ence of higher oxygen suggests greater concentration of oxygen-
ated copper compounds observed on the metal surface. According
to Norouzi et al. [17], the oxide layer created on the surface is due to
exposure of metals to oxygen. Moreover, the carbon content
significantly increased on the surface where corrosion had caused
severe damage. This increase in carbon content could be expressed
as organic deposits, such as oxide layers and fatty acid salts. Ac-
cording to Fazal et al. [14] oxidation of biodiesel can produce
different types of fatty acids, which can accelerate the corrosion
rate by forming different metal oxides.
3.3.2. Chemical structure of corrosion products
The chemical structure (FTIR spectra) of copper exposed to
B20D70E10 fuel blend is shown in Fig. 8. More number of peaks was
observed after immersion at room temperature and 60
C
compared to before immersion. The development of new peaks
after immersion is due to formation of corrosion products. The
chemical structure was completely changed after immersion,
which is significant at 60
C. As shown in Fig. 8 (B) & (C), the peaks
at the regions 580 cm1 and 610 cm1 are indicating the presence
of CuO and Cu2O, respectively and the hydroxyl (OH) group in the
region 3620 to 3680 cm1 indicates the presence of Cu (OH) 2. The
peaks at the region 1670 to 1750 cm1 are likely to indicate the
formation of carboxylate anion (COOe), suggests the film to be
composed of metal carboxylate corrosion compounds, e.g. CuCO3
[12]. Carboxylates are the salts and esters of carboxylic acids, which
are most likely derived from the original ester group in biodiesel.
The peaks at the regions 700, 1470 and 2100 cm1 indicates the
presence of alkene.
Fig. 9 shows the change of chemical structure of mild steel
exposed to B20D70E10 fuel. As shown in Fig. 9 (B) & (C), the peak
after immersion in the region 590 cm1 indicates the presence of
Fe3O4, whereas the small peaks in the region 1020 cm1 and
890 cm1 are due to g-FeOOH and a-FeOOH, respectively [24]. The
new peaks after immersion in the regions 1318 to 1710 cm1 are
indicating the formation of carboxylic acid and esters in the fuel,
which produced the carboxylate corrosion product Fe2O2CO3. The
free hydroxyl (OH) groups in the regions 3601 to 3788 cm1 indi-
cate the presence of Fe (OH)3. In overview, the corrosion of metals
in B20D75E5 and B20D70E10 blends are higher than that of B0,
whereas comparatively lower than B100. In addition, the degra-
dation of fuel properties of BDE is somewhat lower than B100,
which is due to the BDE blends exhibiting better properties of water
tolerance than the biodiesel [3]. However, the TAN value is found to
be problematic from the investigated BDE fuel properties, as the
values are slightly higher than the ASTM specific standard for
biodiesel.
4. Conclusion
Corrosiveness of B20D75E5 fuel is less than B20D70E10 fuel. At
60
C, the corrosion of metals in BDE blends is higher than at room
temperature. Corrosion rate of metals in BDE blends is in the order:
aluminum < mild steel < copper at both temperature conditions.
Degradation of fuel properties in BDE blends, such as TAN, density,
viscosity, calorific value, flash point and color changes are higher
than diesel (B0). In addition, more amounts of oxidation products,
water content and metal elements were observed in the BDEs after
exposure to metals. Metal elements of copper enhance the
 S.K. Thangavelu et al. / Renewable Energy 94 (2016) 1e9
oxidation of BDE blends.
Acknowledgments
The authors are grateful to Swinburne University of Technology,
Sarawak, Malaysia and Universiti Malaysia Sarawak, Malaysia for
facilities and support. Research Management Centre, Universiti
Teknologi Malaysia, Malaysia is also acknowledged for their sup-
port. We also thank our colleagues for proof reading the
manuscript.
References
[1] A. Demirbas, Competitive liquid biofuels from biomass, Appl. Energy 88
(2011) 17e28.
[2] P. Kwanchareon, A. Luengnaruemitchai, S. Jai-In, Solubility of a diesele bio-
dieseleethanol blend, its fuel properties, and its emission characteristics from
diesel engine, Fuel 86 (2007) 1053e1061.
[3] S.A. Shahir, H.H. Masjuki, M.A. Kalam, A. Imran, I.M. Rizwanul Fattah, A. Sanjid,
Feasibility of dieselebiodieseleethanol/ethanol blend as existing CI engine
fuel: an assessment of properties, material compatibility, safety and com-
bustion, Renew. Sust. Energy Rev. 32 (2014) 379e395.
[4] G. Labeckas, S. Slavinskas, M. Ma zeika, The effect of ethanoledieselebiodiesel
blends on combustion, performance and emissions of a direct injection diesel
engine, Energy Convers. Manage 79 (2014) 698e720.
s, A. Todoruţ, D. Ba
[5] I. Baraba ldean, Performance and emission characteristics of
a CI engine fuelled with dieselebiodieseleethanol blends, Fuel 89 (2010)
3827e3832.
[6] N. Yilmaz, F.M. Vigil, A.B. Donaldson, T. Darabseh, Investigation of CI engine
emissions in biodieseleethanolediesel blends as a function of ethanol con-
centration, Fuel 115 (2014) 790e793.
[7] S.A. Shahir, H.H. Masjuki, M.A. Kalam, A. Imran, A.M. Ashraful, Performance
and emission assessment of dieselebiodieseleethanol/ethanol blend as a fuel
in diesel engines: a review, Renew. Sust. Energy Rev. 48 (2015) 62e78.
[8] M.A. Fazal, A.S.M.A. Haseeb, H.H. Masjuki, A critical review on the tribological
compatibility of automotive materials in palm biodiesel, Energy Convers.
Manage 79 (2014) 180e186.
[9] A.S.M.A. Haseeb, M.A. Fazal, M.I. Jahirul, H.H. Masjuki, Compatibility of auto-
motive materials in biodiesel: a review, Fuel 90 (2011) 922e931.
9
[10] A.S.M.A. Haseeb, H.H. Masjuki, L.J. Ann, M.A. Fazal, Corrosion characteristics of
copper and leaded bronze in palm biodiesel, Fuel Process. Technol. 91 (2010)
329e334.
[11] M.A. Fazal, A.S.M.A. Haseeb, H.H. Masjuki, Comparative corrosive character-
istics of petroleum diesel and palm biodiesel for automotive materials, Fuel
Process. Technol. 91 (2010) 1308e1315.
[12] M.A. Fazal, A.S.M.A. Haseeb, H.H. Masjuki, Corrosion mechanism of copper in
palm biodiesel, Corro. Sci. 67 (2013) 50e59.
[13] M.A. Fazal, A.S.M.A. Haseeb, H.H. Masjuki, Degradation of automotive mate-
rials in palm biodiesel, Energy 40 (2012) 76e83.
[14] M.A. Fazal, M.R. Jakeria, A.S.M.A. Haseeb, Effect of copper and mild steel on the
stability of palm biodiesel properties: a comparative study, Ind. Crop. Prod. 58
(2014) 8e14.
[15] E. Hu, Y. Xu, X. Hu, L. Pan, S. Jiang, Corrosion behaviors of metals in biodiesel
from rapeseed oil and methanol, Renew. Energ 37 (2012) 371e378.
[16] D.L. Cursaru, G. Branoiu, I. Ramada, F. Miculescu, Degradation of automotive
materials upon exposure to sunflower biodiesel, Ind. Crop. Prod. 54 (2014)
149e158.
[17] S. Norouzi, F. Eslami, M.L. Wysynski, A. Tsolakis, Corrosion effects of RME in
blends with ULSD on aluminium and copper, Fuel Process. Technol. 104
(2012) 204e210.
[18] M.A. Fazal, A.S.M.A. Haseeb, H.H. Masjuki, Effect of temperature on the
corrosion behavior of mild steel upon exposure to palm biodiesel, Energy 36
(2011) 3328e3334.
[19] K.V. Chew, A.S.M.A. Haseeb, H.H. Masjuki, M.A. Fazal, M. Gupta, Corrosion of
magnesium and aluminum in palm biodiesel: a comparative evaluation, En-
ergy 57 (2013) 478e483.
[20] S.K. Thangavelu, C. Piraiarasi, A.S. Ahmed, F.N. Ani, Corrosion behavior of
copper in biodieseledieseleethanol (BDE), Adv. Mat. Res. 1098 (2015) 44e50.
[21] S.K. Thangavelu, C. Piraiarasi, A.S. Ahmed, F.N. Ani, Compatibility of elastomers
in biodieseledieselebioethanol blend (BDE), Adv. Mat. Res. 1098 (2015)
51e57.
[22] L.M. Baena, M. Go mez, J.A. Caldero n, Aggressiveness of a 20% ethanole80%
gasoline mixture on autoparts: I behavior of metallic materials and evaluation
of their electrochemical properties, Fuel 95 (2012) 320e328.
[23] R. Sanchez, C. Sanchez, C.P. Lienemann, J.L. Todoli, Metal and metalloid
determination in biodiesel and bioethanol, J. Anal. At. Spectrom. 30 (2015)
64e101.
[24] A.V. Ramesh Kumar, R.K. Nigam, S.S. Monga, G.N. Mathur, Effect of inhibitors
on the nature of corrosion products of mild steel by Mossbauer and FTIR
spectroscopy, Anti-Corrosion Methods Mater. 49 (2002) 111e117. http://dx.
doi.org/10.1108/00035590210419353.