Bioresource Technology 100 (2009) 19–24

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Ultrasonic monitoring of glycerol settling during transesterification of soybean oil

A. Bulent Koc *
Harran University, Faculty of Agriculture, Department of Agricultural Machinery, Sanliurfa, Turkey

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

Article history: A low-intensity ultrasonic measurement system was used to monitor the products of transesterification
Received 4 February 2008 of soybean oil in methanol to FAME (biodiesel). The byproducts of the transesterification reaction are
Received in revised form 19 May 2008 methyl esters, glycerol and other products. During the transesterification reaction, the glycerol, having
Accepted 23 May 2008
a higher density than the methyl ester, settles at the bottom of the reaction vessel. The aim of this study
Available online 21 July 2008
was to measure the glycerol deposition rate during transesterification and to assess the reaction rate and
end time. Soybean oil was converted into biodiesel at four temperature levels. The amount of catalyst
Keywords:
(KOH) used in the transesterification reactions was determined by titration. The ultrasonic waveforms
Ultrasonic velocity
Monitoring
captured during the reaction were recorded and analyzed automatically. The ultrasonic system moni-
Transesterification tored the effects of reaction temperatures on the glycerol settling rate and the reaction end times. The
Biodiesel ultrasonic measurement of glycerol settling would be a useful non-destructive method for evaluating
Glycerol the effects of parameters such as catalyst amount, mixing time and temperature on transesterification
reactions.
Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction final products. Other byproducts such as residual alcohol, unre-

Biodiesel is an important alternative fuel. It is biodegradable,
non-toxic, produced from renewable energy sources and less
harmful to the environment than petroleum-based diesel fuel
(Kerschbaum and Rinke, 2004; Marchetti et al., 2007). Biodiesel
can be produced from vegetable oil and animal fat through the
transesterification process. Transesterification is a chemical reac-
tion process during which various types of vegetable oil or fat
are combined with alcohol, usually ethanol or methanol, in the
presence of a catalyst to form fatty esters and glycerol (Gerpen,
2005; Marchetti et al., 2007; Saifuddin and Chua, 2004; Meher et
al., 2006; Demirbas, 2008). After the reaction, the products sepa-
rate and form into two phases. Glycerol, a byproduct of transeste-
rification, forms the first phase and the fatty esters form the second
phase (Saifuddin and Chua, 2004). The fatty esters settle above the
glycerol in the reaction vessel. The formation of the glycerol phase
begins immediately after the reaction starts. The overall transeste-
rification process occurs in a sequence of three reversible and con-
secutive steps (Marchetti et al., 2007; Meher et al., 2006; Singh et
al., 2006). The triglycerides convert into diglycerides, and then the
diglycerides convert into monoglycerides, which convert into glyc-
erol. Diglycerides and monoglycerides are the intermediate prod-
ucts of the reactions, while the fatty esters and glycerol are the
* Present address: Ankara University, Faculty of Agriculture, Department of
Agricultural Machinery, 06140 Diskapi-Ankara, Turkey. Tel.: +90 536 2328791; fax:
+90 312 3183888.
E-mail addresses: a_bulent_koc@yahoo.com, koc@harran.edu.tr
acted mono, di and triglycerides, moisture and free fatty acids
may be present in biodiesel (Knothe, 1999). As vegetable oil or
fat is converted to fatty esters, the height of the glycerol layer in-
creases until all the oil or fat is converted. The rate of glycerol
deposition depends on the oil to fatty ester conversion rate, which
depends on the process variables of reaction temperature, catalyst
type and amount, mixing time and intensity, oil or fat quality and
amount of alcohol (Marchetti et al., 2007). Several research articles
were published regarding the optimization of these parameters
during biodiesel production from neat vegetable oils and used veg-
etable oils (Singh et al., 2006; Vicente et al., 2007; Ghadge and Rah-
eman, 2006; Shieh et al., 2003; Antolin et al., 2002).
Transesterification reduces the density and viscosity of vegeta-
ble oil and makes it possible to use the fatty esters in regular com-
bustion engines. There are several methods used to test the quality
of biodiesel, including gas chromatography (GC), high performance
liquid chromatography (HPLC), 1H NMR and near infrared (NIR)
spectroscopy. An extensive review about the analytical methods
used in production and fuel quality assessments are presented by
Knothe (2001). Most of these methods and chemical analyses re-
quire taking samples to determine the quality parameters and
the analyses can be time-consuming.
Ultrasonic non-destructive evaluation is an alternative method
for monitoring the transesterification process. Providing non-
destructive, non-invasive, robust and rapid measurements,
suitable for automation, at relatively low costs make ultrasonic
evaluation techniques attractive for monitoring and control of
various bioprocesses. Low-intensity ultrasonic inspection is based

0960-8524/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.biortech.2008.05.037
20 A. B. Koc / Bioresource Technology 100 (2009) 19–24
on propagating high frequency sound waves through a test mate-
rial and measuring the ultrasonic parameters that characterize
the propagating ultrasound (Koc and Vatandas, 2006). Common
ultrasonic parameters are ultrasonic velocity, attenuation, acoustic
impedance and scattering. Among these, ultrasonic velocity and
attenuation are the most frequently measured parameters used
to evaluate material properties. Developing a meaningful relation-
ship between the ultrasonic parameters and physicochemical
properties of the bioproducts is fundamental for a successful
application.
A temperature-controlled automatic ultrasonic measurement
system can be used to monitor the glycerol settling rate during
the transesterification reaction. Monitoring the rate could provide
information to decide when to cease the reaction and separate the
ester from glycerol. Real-time ultrasonic monitoring could also be
used to determine the effects of various parameters on the reaction
speed and time to optimize biodiesel production processes as well
as triglyceride to fatty ester conversion rates and yield. The
purpose of this study was to develop an automated ultrasonic mea-
surement system to monitor glycerol settling during transesterifi-
cation of soybean oil for biodiesel production.
2. Methods
2.1. Materials
The materials used in this study were degummed soybean oil,
methanol (purities >95%) and potassium hydroxide (KOH) (purities
>85%). Transesterification is one of the most common methods of
biodiesel production and is usually conducted in batches. The pro-
cess involves a chemical reaction in which an alcohol and an ester
react to produce a different type of alcohol and a different type of
3.0
2.0
1.0
Amplitude (V)
0.0
to
-1.0
-2.0
-3.0
0.E+00 3.E-05 6.E-05
Fig. 1a. Ultrasonic waveform on a layer of methoxide above vegetable oil. (32 ml methoxide and 98 ml soybean oil; the total height of two liquids in 12 mm diameter vessel
was 101.28 mm, methoxide height was 24.0 mm and oil height was 76.88 mm). to: time of flight of ultrasonic wave in vegetable oil layer, tm: time of flight of ultrasonic wave
in methoxide layer.
ester. Soybean oil with a specific gravity of 910 kg/m3 was mixed
with a methanol and KOH mixture at four temperatures (35, 40,
50 and 60 °C). Soybean oil ester, glycerol and other residue were
the products of the transesterification process.
The amount of KOH (as a catalyst) was determined with titra-
tion. For this study 1 g KOH was mixed with 32 ml of methanol un-
til all the KOH was dissolved. Next, 98 ml of soybean oil was
warmed to the desired temperature in the measurement cell. The
methoxide solution was mixed with soybean oil for 5 min with a
mixer. After mixing, the mixer was removed from the measure-
ment cell and ultrasonic measurements were conducted. The com-
puter program was initiated to take measurements every 5 s. The
ultrasonic waveforms appearing on the oscilloscope screen were
transferred to the computer and the significant ultrasonic spikes
in the waveform were analyzed.
2.2. Ultrasonic measurement system and measurement procedure
The ultrasonic measurement system consisted of a measure-
ment cell, an ultrasonic transducer, a signal generator, an oscillo-
scope and a computer. The ultrasonic measurement cell was
designed to take measurements in pulse-echo mode. A cylindrical
glass vessel of 13.5 cm length and 38.0 mm inner diameter was
used as the measurement cell. An ultrasonic transducer was fitted
into one end of the glass vessel. An o-ring washer provided a leak-
proof surface between the transducer and the glass vessel. The
other end of the transducer was fitted into a 12.0 cm length plastic
cylinder base. Two temperature sensors were placed on one side of
the measurement cell at 1.5 cm and 8.0 cm distances from the
ultrasonic transducer surface. The temperature sensors were in
contact with the reagents in the measurement cell. A flexible resis-
tance heater with 15  10 cm dimensions was wrapped around the
measurement cell to adjust the temperature of the reagents to the
tm
9.E-05 1.E-04 2.E-04 2.E-04
Time (s)
 A. B. Koc / Bioresource Technology 100 (2009) 19–24 21

desired level during the measurements. Oil and methoxide were
poured into the measurement cell before the measurements were
taken. As methoxide and oil are not miscible without agitation, an
electric mixer was placed in the upper end of the measurement cell
to mix the reagents. After the reagents were mixed for 5 min, the
mixer was removed and the upper end of the measurement cell
was closed with a plastic cap.
A 1.0 MHz broadband central frequency and 38.0 mm diameter
immersion ultrasonic transducer was used during the experiments
(Panametrics V392, Waltham, MA, USA). An ultrasonic pulser/re-
ceiver was used to drive the transducer (Panametrics 500 PR, Wal-
tham, MA, USA). Electrical signals were transmitted and received
by the transducer and monitored on a digital oscilloscope screen
(Tektronix TDS 3032, Tektronix Inc., Wilsonville, OR, USA).
The oscilloscope was attached to a PC with a GPIB connection.
The measurements were analyzed and recorded by a computer
with a program developed in G-programming language (Labview
6.1, National Instruments, Mopac Expwy, Austin, TX, USA).
To minimize noise errors, 512 ultrasonic pulses were averaged
for every measurement and monitored on the oscilloscope screen.
Transferring the waveform from the oscilloscope to the computer,
determining the time of flight and recording the results totaled

Echo 1 Echo 2
2

1
Amplitude (V)

-1

-2

-3
0.E+00 2.E-05 4.E-05 6.E-05 8.E-05 1.E-04 1.E-04 1.E-04 2.E-04 2.E-04 2.E-04
Time (s)
Fig. 1b. Ultrasonic waveform measured after 5 min of mixing soybean oil and methoxide. (Echo 1: ultrasound echo reflected back to the transducer from the interface
between the transducer surface and the oil-methoxide mixture, Echo 2: ultrasound echo reflected back to the transducer from the liquid mixture-air interface).

3
Echo 1

2
Echo 3

1 WF17 WF25 WF33 WF41 WF49 WF83

Amplitude (V)

-1

-2

-3
0.0E+00 5.0E-06 1.0E-05 1.5E-05 2.0E-05 2.5E-05 3.0E-05 3.5E-05
Time (s)
Fig. 2. Progress of the ultrasonic waveform during the transesterification reaction of soybean oil with methoxide at temperature of 35 °C. WF17, WF25, WF33, WF41, WF49,
and WF83: waveforms measured at 17th, 25th, 33rd, 41st, 49th and 83rd min.
22 A. B. Koc / Bioresource Technology 100 (2009) 19–24

approximately 1 s. The time of flight of the ultrasonic waves in the ing tube. The liquid was kept at the circulating water temperature
liquid mixture were displayed on the computer screen in real time. for 15 min and then viscosity was measured.
An experiment was conducted to determine the possible effect
of the ultrasonic measurement system on the transesterification
3. Results and discussion
reaction. For this experiment, methoxide solution was poured into
the soybean oil in the ultrasonic measurement vessel. No mechan-
3.1. Ultrasonic waveforms during progressing transesterification
ical mixing was applied to the liquids in the vessel. Ultrasonic mea-
reaction
surements were conducted. Because there was no agitation due to
mixing, methoxide and vegetable oil did not mix during the mea-
When a traveling ultrasonic wave encounters an interface
surement period of 2 h. From this experiment, it was concluded
(boundary between two phases), some of the sound wave propa-
that the low intensity ultrasonic system used in this study did
gates through the second medium, and some is reflected from
not have an effect on the transesterification reaction.
the interface. The amount of reflected and transmitted sound
waves depends on the acoustic impedances of the two media.
2.3. Density and viscosity measurements
The acoustic impedance (Z, kg/m2 s) of a material is the product of
material density (q) and sound velocity (c) which is given in
The densities of the soybean oil and the methyl ester after
Eq. (1) (Hykes et al., 1992).
transesterification at four temperatures were measured using a
hydrometer. The sample liquid was warmed to 40 °C and 100 ml Z ¼ qc ð1Þ
:
of sample was poured into a graduated cylinder and the density
of the liquid was measured. During transesterification reactions, vegetable oil, methanol
The viscosities of the soybean oil and the methyl esters were and KOH are mixed to start the reaction. When methoxide is
measured using a falling ball viscometer (Colora Dropping Ball Vis- poured into oil, they do not mix due to density differences. Meth-
cometer, Type A, Germany). The measuring tube was filled with the oxide and oil layers have different acoustic impedances. Fig. 1a
sample liquid and 40 °C water was circulated around the measur- shows a typical ultrasonic waveform transmitted through the

3.50E-05

35ºC 40ºC 50ºC 60ºC

3.00E-05
Ultrasonic wave time of flight (s)

2.50E-05

2.00E-05

1.50E-05

1.00E-05

5.00E-06 a b c d

0.00E+00
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85
Time (min)
Fig. 3. Ultrasonic time of flight measured during the transesterification of soybean oil with methoxide (1 g KOH, 32 ml methanol 98 ml oil) at four temperatures (35, 40, 50
and 60 °C) (a–d indicate when the glycerol started settling).

Table 1
Glycerol settling start and end times for the experiments conducted at four temperatures

Experiment temperature (°C) Glycerol deposition start time (min) Glycerol deposition end time (min) Point of max. glycerol deposition (min)
35 14.91 65.25 27.41
40 12.25 44.41 19.5
50 4.58 19.50 7.66
60 4.16 21.08 6.08
 A. B. Koc / Bioresource Technology 100 (2009) 19–24 23

1.0E-05

9.0E-06 35ºC 40ºC 50ºC 60ºC

8.0E-06
Rate of time of flight change

7.0E-06

6.0E-06

5.0E-06

4.0E-06

3.0E-06

2.0E-06

1.0E-06

0.0E+00
0 10 20 30 40 50 60
Time (min)
Fig. 4. First derivatives of the ultrasound time of flight change with time for Echo 3 for four experiments. The peaks on each curve indicate the time when the glycerol settling
rate reached a maximum for that experiment.

methoxide layer above vegetable oil before they are mixed. Fig. 1b
Table 2
shows the ultrasonic waveform measured immediately after Density and viscosity values of the soybean oil and methyl esters measured at 40 °C
mixing.
The ultrasonic waveform in Fig. 1a changes with the progress- Sample liquid Density (g/ml) Viscosity (cP)

ing transesterification. Echo 1 in Fig. 1a is the ultrasound echo re- Soybean oil 0.9056 25.25
flected back to the transducer from the interface between the Methyl ester produced at 35 °C 0.8648 3.061
Methyl ester produced at 40 °C 0.8624 2.74
transducer surface and the oil-methoxide mixture; Echo 2 is the
Methyl ester produced at 50 °C 0.8645 2.86
ultrasound echo reflected back to the transducer from the liquid Methyl ester produced at 60 °C 0.8650 2.83
mixture-air interface. As the reaction occurs, glycerol deposits at
the bottom of the measurement cell and soybean oil methyl ester
remains above the glycerol. As a result, an interface between the
glycerol and methyl ester is formed. A third echo (Echo 3) reflected Ultrasonic velocity in glycerol was measured using the same
back to the transducer from the glycerol-methyl ester interface is ultrasonic measurement system. The average density of the glyc-
formed on the waveform. Echo 3 moves from left to right as shown erol settled at the bottom of the measurement vessel for all the
in Fig. 2. As the height of the glycerol increases, the distance (twice experiments was 1.098 g/ml at 30 °C. The ultrasonic velocity in
the glycerol height) traveled by the ultrasonic wave and the time of glycerol decreased with increasing temperature. The ultrasonic
flight increases. velocity in glycerol was 1679.2 m/s at 35 °C and 1595.6 m/s at
The ultrasonic measurements began immediately after the mix- 60 °C.
ing of oil and methoxide. Time of flight for Echo 3 was measured at
each interval and recorded to a file. The points when the ultrasonic 3.2. Results of density and viscosity measurements
times of flight for Echo 3 began to increase were considered to be
the glycerol settling start time. The glycerol deposition start times The neat soybean oil and methyl ester produced by transeste-
decreased with increasing temperature as shown in Fig. 3. The rification at four temperatures were measured at 40 °C using a
shortest glycerol deposition start time (4.16 min) was measured hydrometer and a falling ball viscometer. The results of these
for the experiment conducted at 60 °C. The longest glycerol deposi- measurements are shown in Table 2. The density of the soybean
tion start time of 14.66 min was measured for the experiment con- oil was 0.9056 g/ml and after transesterification at four temper-
ducted at 35 °C (Table 1). The first derivatives of the time of flight atures, the average density of methyl ester was measured as
for Echo 3 change with time were used to determine the glycerol 0.864 g/ml. The lowest density was measured for methyl ester
settling end times. The first point when the derivative of the curve produced at 40 °C as 0.8624 g/ml and the highest was measured
reached zero was considered to be the end of the glycerol settling for the methyl ester produced at 60 °C. The viscosity of soybean
(Fig. 4). Glycerol settling end times are also shown in Table 1. oil at 40 °C was measured to be 25.25 cP. The viscosity of methyl
For all experiments, the reaction speed (glycerol deposition esters was about nine times lower than soybean oil. The average
rate) increased at a decreasing rate of acceleration. After a certain viscosity of the methyl esters produced at four temperatures was
point where the glycerol deposition rate reached a maximum, the 2.87 cP. The highest viscosity was measured for the methyl ester
deposition rate increased at a slower pace until the glycerol depo- produced at 35 °C. This is an expected result because as the
sition stopped. The times at which the maximum glycerol settling transesterification reaction temperature decreases, the oil to es-
occurred for each experiment are shown in Table 1. Increasing the ter conversion speed also decreases and the conversion takes
temperature reduced the glycerol settling time. longer time to complete.
24 A. B. Koc / Bioresource Technology 100 (2009) 19–24
4. Conclusions
Low intensity ultrasound was used to monitor soybean oil in
methanol transesterification reactions. The ultrasonic time of flight
during the reaction was determined from the ultrasonic wave-
forms automatically. During the transesterification reactions, glyc-
erol deposits at the bottom of the reaction cell. An interface is
formed between the deposited glycerol and the methyl ester,
above the glycerol. Due to the acoustic impedance differences be-
tween glycerol and methyl ester, some of the ultrasonic waves are
reflected from the interface and the remainder propagates through
the reacting solution and reflects back from the liquid-air interface.
The deposition rate of glycerol, an indicator of reaction rate, was
monitored and the transesterification end points were determined
with an on-line pulse-echo ultrasonic measurement system. The
ultrasonic system provided a rapid and non-invasive method for
monitoring the transesterification process. The ultrasonic mea-
surement system could be used to determine the effects of param-
eters such as catalyst type and amount, reaction temperature,
mixing time, alcohol type and amount on the reaction rate, reac-
tion end times and yield in the transesterification of various vege-
table oils.
References
Antolin, G., Tianaut, F.V., Briceno, Y., Castano, V., Perez, C., Ramirez, A.I., 2002.
Optimization of biodiesel production by sunflower oil transesterification.
Bioresource Technology 83, 111–114.
Demirbas, A., 2008. Comparison of transesterification methods for production of
biodiesel from vegetable oils and fats. Energy Conversion and Management 49
(1), 125–130.
Gerpen, J.V., 2005. Biodiesel processing and production. Fuel Processing Technology
86, 1097–1107.
Ghadge, S.V., Raheman, H., 2006. Process optimization for biodiesel production from
mahua (Madhuca indica) oil using response surface methodology. Bioresource
Technology 97, 379–384.
Hykes, D.L., Hedrick, W.R., Starchman, D.E., 1992. Ultrasound Physics and
Instrumentation, second ed. Springer, Berlin (pp. 5–23).
Kerschbaum, S., Rinke, G., 2004. Measurement of the temperature dependent
viscosity of biodiesel fuels. Fuel 83, 287–291.
Knothe, G., 1999. Monitoring a progressing transesterification reaction by fiber-
optic near infrared spectroscopy with correlation to 1H nuclear magnetic
resonance spectroscopy. JAOCS 77 (5), 489–493.
Knothe, G., 2001. Analytical methods used in the production and fuel quality
assessment of biodiesel. Transactions of the ASAE 44 (2), 193–200.
Koc, A.B., Vatandas, M., 2006. Ultrasonic velocity measurements on some liquids
under thermal cycle: Ultrasonic velocity hysteresis. Food Research International
39, 1076–1083.
Marchetti, J.M., Miguel, V.U., Errazu, A.F., 2007. Possible methods for biodiesel
production. Renew Sustainable Energy Review 11, 1300–1311.
Meher, L.C., Sagar, D.V., Naik, S.N., 2006. Technical aspects of biodiesel production
by transesterification – a review. Renew Sustainable Energy Review 10 (3),
248–268.
Saifuddin, N., Chua, K.H., 2004. Production of ethyl ester (Biodiesel) from used
frying oil: optimization of transesterification process using microwave
irradiation. Malaysian Journal of Chemistry 6 (1), 77–82.
Singh, A., He, B., Thompson, J., Gerpen, J.V., 2006. Process optimization of biodiesel
production using alkaline catalysts. Applied Engineering in Agriculture 22 (4),
597–600.
Shieh, C.-J., Liao, H.-F., Lee, C.-C., 2003. Optimization of lipase-catalyzed biodiesel by
response surface methodology. Bioresource Technology 88, 103–106.
Vicente, G., Martinez, M., Aracil, J., 2007. Optimization of integrated biodiesel
production. Part 1. A study of the biodiesel purity and yield. Bioresource
Technology 98 (9), 1724–1733.