This article was downloaded by: [University of Victoria]

On: 25 April 2015, At: 16:05

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Food Additives & Contaminants: Part A

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tfac20

Other factors to consider in the formation of

chloropropandiol fatty esters in oil processes
a a a b
Muhamad Roddy Ramli , Wai Lin Siew , Nuzul Amri Ibrahim , Ainie Kuntom & Raznim Arni
b
Abd. Razak
a
Protein & Food Technology Unit, Malaysian Palm Oil Board, Kajang, Malaysia
b
Analytical & Quality Development Unit, Malaysian Palm Oil Board, Kajang, Malaysia
Accepted author version posted online: 23 Mar 2015.Published online: 20 Apr 2015.

Click for updates

To cite this article: Muhamad Roddy Ramli, Wai Lin Siew, Nuzul Amri Ibrahim, Ainie Kuntom & Raznim Arni Abd. Razak
(2015): Other factors to consider in the formation of chloropropandiol fatty esters in oil processes, Food Additives &
Contaminants: Part A, DOI: 10.1080/19440049.2015.1032368

To link to this article: http://dx.doi.org/10.1080/19440049.2015.1032368

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Food Additives & Contaminants: Part A, 2015
http://dx.doi.org/10.1080/19440049.2015.1032368

Other factors to consider in the formation of chloropropandiol fatty esters in oil processes
Muhamad Roddy Ramlia*, Wai Lin Siewa, Nuzul Amri Ibrahima, Ainie Kuntomb and Raznim Arni Abd. Razakb
a
Protein & Food Technology Unit, Malaysian Palm Oil Board, Kajang, Malaysia; bAnalytical & Quality Development Unit, Malaysian
Palm Oil Board, Kajang, Malaysia
(Received 6 November 2014; accepted 17 March 2015)

This paper examines the processing steps of extracting palm oil from fresh fruit bunches in a way that may impact on the
formation of chloropropandiol fatty esters (3-MCPD esters), particularly during refining. Diacylglycerols (DAGs) do not
appear to be a critical factor when crude palm oils are extracted from various qualities of fruit bunches. Highly hydrolysed
oils, in spite of the high free fatty acid (FFA) contents, did not show exceptionally high DAGs, and the oils did not display a
higher formation of 3-MCPD esters upon heat treatment. However, acidity measured in terms of pH appears to have a
strong impact on 3-MCPD ester formation in the crude oil when heated at high temperatures. The differences in the
Downloaded by [University of Victoria] at 16:05 25 April 2015

extraction process of crude palm oil from current commercial processes and that from a modified experimental process
showed clearly the effect of acidity of the oil on the formation of 3-MCPD esters. This paper concludes that the washing or
dilution step in palm oil mills removes the acidity of the vegetative materials and that a well-optimised dilution/washing
step in the extraction process will play an important role in reducing formation of 3-MCPD esters in crude palm oil upon
further heat processing.
Keywords: chloropropandiol fatty esters; palm oil; diacylglycerols; acidity

Introduction (Fronimaki et al. 2002) have higher DAG levels than

In recent years, chloropropanediol fatty acid esters seed oils. Some nut oils such as salfat, shea butter fats
(3-MCPD esters – ME) have been detected in oils and and other exotic fats have high DAG contents as well. In
fats (Zelinková et al. 2006; Weißhaar 2008, 2011). These recent years, ME and glycidyl esters (GE) have been
contaminants are formed during processing of oils and linked to DAGs present in vegetable oils (Hamlet et al.
fats, occurring when lipids such as triacylglycerols 2011; Matthäus et al. 2011). Some presentations and pub-
(TAGs) or glycerols react with chlorides at high tempera- lications (Matthäus et al. 2011; Pudel et al. 2012) pointed
tures. Among food products, high levels have been specifically to DAGs as potential culprits of ME forma-
detected in oils and fats or foods containing a high fat tion, especially observed upon adding DAGs to oils. As all
content. As the levels are generally higher in refined oils, oils and fats have similar components (TAG, DAG, mono-
many researchers (Franke et al. 2009; Hrnčiřík & van acylglycerol (MAG), chlorides), high temperature treat-
Duijn 2011; Nagy et al. 2011; Ramli et al. 2011; Craft ment for physical refining of palm oil has been one of
et al. 2012; Destaillats et al. 2012) have investigated the the decisive factors resulting in a greater amount of chlor-
factors attributing to the formation, and certainly their opropandiols. Chemical refining, however, showed less
studies have contributed to a better understanding of the tendency of ME formation as neutralisation and water
presence of the esters in the oil. washing to remove the residual of soap partially eliminates
Levels of ME in palm oil have been reported (Abd. reactive chlorinated precursors from the crude palm oil
Razak et al. 2012; Yamazaki et al. 2013; MacMahon et al. (CPO). Deodorisation is also conducted at relatively lower
2013). Refined seed oils in general have lower values. temperatures.
Some researchers (Nagy et al. 2011; Freudenstein et al. Chlorides were recently studied (Franke et al. 2009;
2013) reasoned that possible precursors such as higher Nagy et al. 2011) and it would not be a surprise that all oils
diacylglycerol (DAG) contents and chloride levels could and fats may have varying quantities of chlorides, being such
be the contributing factors attributing to the higher forma- ubiquitous materials in nature. Franke et al. (2009) reported
tion of the esters in refined palm oil. In this study, another total chlorine in palm oil and rapeseed oil as 2 and < 2 ppm,
possible factor is examined while investigating DAGs in respectively, and for chlorides < 1 ppm in both oils. Nagy
the oil. DAGs are present in all oils and fats, the amount et al. (2011) discussed the type of chlorides as being both
varying from 1% to as much as 8%. Generally, fruit oils inorganic and organic in nature. While there is a small
such as palm oil (Siew & Ng 1995) and olive oil amount of chlorides present in oils and fats, all oils come

*Corresponding author. Email: roddy@mpob.gov.my

© 2015 Malaysian Palm Oil Board. Published by Taylor & Francis.

 2 M.R. Ramli et al.

into contact with chlorides, for example in bleaching clays, Fresh fruit bunches (FFB)
water, phosphoric acid, etc.
Several analytical measurements of ME and GE have
Steam Sterilization Condensate
been reported (Crews et al. 2013; Ermacora & Hrnčiřík
2013; Yamazaki et al. 2013; Ermacora & Hrnčiřík 2014; Sterilized FFB
Zhou et al. 2014). The American Oil Chemists' Society has
also developed three new methods, namely AOCS Cd13-29a Stripping Bunch stalks
(2013a), AOCS Cd13-29b (2013b) and AOCS Cd13-29c
Fruit
(2013c). These methods made it easier to relate the work of
one laboratory with that of another, especially where data on Steam Digestion
different products were reported.
Comprehensive studies have been undertaken to miti-
gate the formation of these esters. The mitigation proce- Pressing
dures can be generally divided into three approaches:
Crude oil Press cake
removal of precursors in the raw material, modification
of the refining process and removal of the esters post- Water Screening
refining. Siew et al. (2012) and Matthäus and Pudel
Downloaded by [University of Victoria] at 16:05 25 April 2015

(2013) discussed potential procedures to reduce the for- Oil

mation of ME based on these three approaches. Craft and Clarification
Nagy (2012) highlighted several ‘suggested practices
(SP)’ from upstream (plantation) to downstream (refining)
to improve the quality and safety of refined palm oil. A
Separator Dirt
summary of the different mitigation strategies possible to
reduce the formation of ME and GE is given in Table 1.
The production of palm oil differs very much from
Vacuum
that of seed oils, in that seed oils are extracted from dry drying
seeds by solvent or by cold press, while because of the
size and nature of the fruit, palm oil is extracted with
the whole fruit bunches in the processing steps. Fresh Crude palm oil (CPO)
fruit bunches (FFBs) are harvested when ripe, usually
20–22 weeks after flower anthesis. The FFBs are trans- Figure 1. Schematic diagram for the production of crude palm
ported in lorries to mills, where they are extracted by a oil.
mechanical press. A bunch consists of a bunch stalk and
spikelets of fruitlets attached to the stalk. The oil extrac- sterilisation, stripping, digestion, pressing, screening,
tion process is described in Figure 1, involving clarification, centrifugation and drying. In some

Table 1. Summary of the possible mitigation strategies in reducing the formation of 3-MCPD and glycidyl esters in refined oil.

Mitigation step Key finding Reference

Refining aids
Glycerol and ethanol 25–35% reduction of MCPD esters Craft et al. (2012)
Ethanol:water (1:1) Approximately 30% reduction of MCPD esters
Diacetin 50% reduction of MCPD esters and related Matthäus and Pudel (2013)
compounds
Carbonates 66% reduction of MCPD esters and related
compounds
Dual deodorisation process (short term at 63% reduction of MCPD esters Matthäus and Pudel (2013)
higher temperature and long term at lower
temperature/vice versa)
Deodorisation using short-path distillation 90% reduction of MCPD esters and related Matthäus and Pudel (2013)
compounds
Neutralisation of degummed oil with
Potassium hydroxide 45% reduction of MCPD esters Matthäus and Pudel (2013)
Sodium hydroxide 35% reduction of MCPD esters
Calcium oxide 45% reduction of MCPD esters Ramli et al. (2011)
Hot water degumming 70% reduction of MCPD esters Ramli et al. (2011)
Washing of crude palm oil before refining 38% reduction of MCPD esters Matthäus and Pudel (2013)
Washing of palm fruit pulp before oil extraction 95% reduction of MCPD esters Craft et al. (2012)
 Food Additives & Contaminants: Part A 3

variations, decanter systems are used instead of clarifi- measurements. The FFA, DAG and pH of the oils were
cation. These steps begin with sterilisation, which kills analysed.
enzymes such as lipases and softens the fruits for ease
of removal from the bunches; cooking the fruits in
digesters; pressing, which releases the oil, followed by Crude palm oil (CPO) from evaporator process
either addition of water for ease of separation of oil In this process shown in Figure 2, the oil/water and
from the vegetative materials or through use of a decan- vegetative mixture were subjected to a falling film eva-
ter. Finally the crude oil is separated by centrifugal poration process, where the water inherent in the fruit was
separators or by decanters, then dried and sent to sto- removed. No dilution of water takes place, and removal of
rage tanks. The difference between oil production from vegetative materials takes place by separators. The water
palm and that of other seeds is mainly in the production obtained from the process is collected for the pH test. The
process, where palm oil comes into contact with vege- crude oil is also tested for FFA, DAG, ME formation on
tative materials of the oil palm (bunch stalk, spikelets, the heat test, and its acidity in terms of pH. The main
calyx, vegetative fibres), whereas for seed oil produc- difference from the conventional commercial process is
tion the vegetative materials are removed from the seeds the removal of the water step, and no additional water is
prior to extraction. This difference in production is seen added to the process, except that which is derived from
as a possible explanation for the higher occurrence of steam used in the sterilisation of bunches.
Downloaded by [University of Victoria] at 16:05 25 April 2015

ME observed upon heat treatment of palm oil. This

paper will examine how this is so.
This paper also evaluates the different conditions of Fresh fruit bunches (FFB)
storage of fruits and their effects on the formation of ME.
It also examines the effect of oil acidity on the formation. Steam Sterilization Condensate

Sterilized FFB

Materials and methods Stripping Bunch stalks

Storage of fresh fruit bunches (FFBs) and oil extraction Fruit
FFBs were harvested from MPOB oil palm estate in Steam Digestion
Bangi, Selangor, Malaysia. Spikelets of fruits were cut
from the bunches and separated into four portions. The
first portion was sterilised fresh for oil extraction. The Pressing
second portion was bruised with a hammer and kept for
1 day prior to sterilisation and oil extraction. The third Crude oil Press cake
portion was incubated at 15°C for 24 h prior to sterili- Screening
sation; while the last set was incubated at 5°C for 24 h
prior to sterilisation. The last two treatments were given Oil
cold treatments to increase the speed of hydrolysis,
since it was reported that lipase within the fruits are Desanding Sand
activated at low temperatures (Sambanthamurthi et al.
1991). Sterilisation was carried out in a laboratory auto-
clave for 30 min at 15 psi. The fruits were then Cake 2-Phase separation
detached from the spikelets and the mesocarps removed Oil and water
using a knife. The mesocarps were pressed in a labora-
Evaporator
tory-scale hydraulic press and the collected oil was
filtered to remove fibres and other impurities. The Oil and sludge
crude oil was dried by a rotavapor for 30 min.
Separator
Analysis of ME upon heat test, FFA and DAG levels
was carried out. Quadruple experiments were con-
ducted. The mean of four sets of bunches was recorded.
Vacuum
drying

Samples collected from palm oil mills

Crude palm oil (CPO)
Samples of CPO were collected from commercial palm oil
mills. The oil samples were heat treated at 260°C for ME Figure 2. Modified experimental process.
 4 M.R. Ramli et al.

Palm fibre oil (PFO) 1.2 ml min−1. The column temperature was programmed at
A sample of PFO is obtained from a local refinery that 60°C (1 min) to 190°C (1 min) at a rate of 6°C min−1, and
processes fibre oil from palm fibres by a solvent process. was then accelerated to 280°C at a rate of 30°C (5 min).
The palm fibres are the vegetative material left over from The injector was held at 180°C and 1 µl of the sample was
the extraction of CPO from fruit bunches. These fibres injected in a split-less mode. The quantitative analysis was
contain 5–8% of oil (Abd. Majid et al. 2012). carried out by monitoring characteristic ions at m/z 91, 147
and 196, respectively, for derivatised 3-MCPD; while for 3-
MCPD-d5, characteristic ions were at m/z 93, 150 and 201.
Diacylglycerol (DAG) analysis Qualifier ions were m/z 196 towards m/z 201.
HPLC was carried out using Gilson (Rue Gambetta,
Villers le Bel, France) 303 and 302 pumps and a Waters
(Milford, MA, USA) differential refractometer. The two Heat test
columns were of 25 cm length and 4 mm i.d. with 5 µm Crude oil samples (2 ml) were heated in a covered vial to
Lichrosphere RP18. The mobile phase was acetonitrile: 260°C, the vial being held in a sandbath. The oil was
acetone (65:35 v/v) at a flow rate of 1.0 ml min–1. maintained at 260°C for 1 h, then cooled and analysed
Injection was achieved through a Rheodyne valve fitted for ME. The measurement may not simulate the different
with a 20 µl loop. Samples were injected as 10% w/v aspects of refining such as degumming and bleaching, but
Downloaded by [University of Victoria] at 16:05 25 April 2015

solution in warm acetone. Identification of TAGs and relates to the high temperature as used in deodorisation. It
DAGs was made by comparison with reference standards. is found to be a reliable indicator of the degree of ME
The DAG values are given in area per cent. The analysis expected for different oil sources.
of DAG was as reported in Yeoh et al. (2014).

pH test of oils
Free fatty acids (FFAs) A total of 100 ml of oil was stirred with a similar amount
FFAs were analysed according to AOCS Official Method of water for 1 h at 70°C. The water was removed by
Ca 5a-40 (1998). separating funnel, and the pH measured with a pH meter.

3-MCPD esters (ME) Statistical analysis

Analysis of ME was carried out according to Abd. Razak Data obtained from the analyses were subjected to one-
et al. (2012), and recorded in mg kg−1 values. The Federal way analysis of variance (ANOVA) with Fisher’s multiple
Institute for Risk Assessment (BfR) Method 008 – comparison to determine the significant differences among
Determination of 3-MCPD fatty acid esters in edible oils the samples defined at 95% confidence interval (p < 0.05).
and solid fats by GC-MS (2009) – was adopted. The inter- All analyses were conducted in duplicate and reported as
nal standard used was deuterated 3-MCPD (3-MCPD-d5). means ± standard deviation (SD) of independent trials.
Oil sample of 100 mg was weighed and dissolved in 0.5 ml The statistical analyses were performed using the
tert-butyl methyl ether (t-BME) containing the internal Minitab software Version 16 statistical package (Minitab
standard (0.4 µg). Sulphuric acid solution in methanol Inc., State College, PA, USA).
(2%, v/v) was added to the sample and the mixture was
incubated for 16 h. Sodium hydrogencarbonate solution
(0.5 ml) was then added to stop the hydrolysis. To separate
Results and discussion
the fatty acid methyl ester from the sample, 1.8 ml of an
aqueous salt solution were added. Derivatisation was DAG and ME formation
achieved by addition of 250 µl of phenylboronic acid solu- Formation and occurrence of ME in foods have been
tion to the mixture. The derivative was extracted and discussed by Hamlet et al. (2011). Mechanisms of forma-
analysed. tion are related to lipids, which are possible precursors of
Detection of the analyte was carried out by means of formation when conditions are optimum, for example
GC-MSD from Agilent Technologies (Santa Clara, CA, reactions occur in the presence of chloride ions at high
USA), equipped with a Series 5975C quadrupole detector reaction temperatures. Thus, it appears easy to find the
and controlled by a programmable GC 7890A. culprits in the case of palm oil, where DAGs are clearly
Chromatographic separation was performed on a fused higher than in other seed oils. Model experiments do show
silica capillary column, HP-5MS inert (30 m length, that DAGs, especially those added to the system, are
0.25 mm i.d., 0.25 µm film thickness). Helium (purity reactive under high temperature conditions (Freudenstein
99.999%) was used as a carrier gas at a constant flow of et al. 2013). However, oils from various sources having
 Food Additives & Contaminants: Part A 5

Table 2. 3-MCPD esters (ME) formed upon heat treatment of of lipase of oil palm fruits was also shown by Cadena
CPO extracted directly from fruit bunches: effect of hydrolysis. et al. (2013).
Sample FFA (wt %) DAG (wt %) ME (mg kg−1) In commercial oils, the FFA limit for traded CPO is
5% and any oil of values of > 10% would have been
Fresh CPO 0.63 ± 0.17a 3.3 ± 0.3aa 1.55 ± 0.64a traded as high FFA oils. Keeping other variables such as
Oil from bruised 1.85 ± 0.71a 3.8 ± 0.2a 1.45 ± 0.80a the temperature of treatment for the oil constant, the
FFB experiment shows that the ME of the oils with higher
High FFA oil 1 6.75 ± 2.06a 5.4 ± 0.9b 0.74 ± 0.23a
High FFA oil 2 59.65 ± 8.02b 4.8 ± 0.5ba 1.01 ± 0.50a FFA and DAG levels derived from same bunch of fruits
did not differ significantly (p ≥ 0.05) from each other. In
Notes: Values are the mean ± SD of duplicate analysis. experiments involving added DAGs from commercial
Values within a column with different superscript letters are significantly
different (p < 0.05). sources, Freudenstein et al. (2013) reported an increase
in ME when 1,3-dipalmitin is added to palm oil. However,
the process by which such 1,3-dipalmitin is produced is
varying FFA and DAG contents point to the contrary of not known. Any impurities or residue (if acidic) left from
the expectations of the above (Lacoste et al. 2010). the production process may cause an effect of increasing
When a FFB is harvested, FFA increases in the oil, the ME formation. The experiments from fruits using
especially if there is considerable bruising, usually caused DAG formed within the fruit illustrated conclusively that
Downloaded by [University of Victoria] at 16:05 25 April 2015

by the ripe fruit dropping to the ground from a height of DAG alone would not be very reactive unless other pre-
several feet. Inherent and extraneous lipases come into conditions are present. Other researchers who have also
contact with the oil released from ruptured cell walls as shown that DAG is not necessarily the critical factor are
a result of bruising. A rise in FFA is slower if fruit Hrnčiřík and Ermacora (2011) and Lacoste et al. (2010).
bruising is minimised. Table 2 shows results of ME Destaillats et al. (2012) showed that TAG is the main
formed in heat-treated CPO extracted from palm fruits of acylglycerol class involved in MCPD diester formation
varying state of hydrolysis. In the fruit experiments, spi- when oils are heated. DAG was found to form ME as
kelets from a bunch were separately treated as (1) freshly well, but it has relatively lower reactivity than TAG.
extracted, (2) bruised fruits and (3 and 4) hydrolysed
fruits. The oils extracted were heat treated for their poten-
tial to form ME. In these experiments, the degree of Crude oil from evaporation process
hydrolysis of palm oil was affected to show the different In an experiment where CPO is processed in a different
levels of FFA and DAG. In fresh oil, the FFA of the CPO manner from that of commercial mill operation, it was
was only 0.63%, while oil from slightly bruised fruits found that the way palm oil is extracted from FFBs may
resulted in an FFA of 1.85%. Chilling the fruits can play a role in determining the formation of ME. Figure 1
enhance lipolytic hydrolysis by lipase present in the oil shows CPO being extracted in conventional mills. The
palm fruit, where it was observed that palm fruits sub- process takes the fruit bunches through sterilisation, strip-
jected to 5°C chilling resulted in FFA levels of up to 70% ping of spikelets from the bunch, digestion of the fruits,
(Sambanthamurthi et al. 1991, 1995; Cadena et al. 2013). pressing and dilution with water, followed by clarification.
Although FFA levels were high, the DAG levels were The clarification process allows settling of the oil and
unexpectedly low. In our experiments, hydrolysed fruits water phase and the oily phase is passed to separators
caused by chilling at different temperatures (15 and 5°C) for removal of dirt, and the crude oil is finally vacuumed
resulted in high FFA oils as indicated by values of 6.75% dried. In Figure 2 oil extraction is carried out from a
and 59.65%, respectively. Structural changes in mesocarp modified process in a small pilot plant. In the initial few
tissue as a result of chilling resulted in high enzyme steps, processing is generally similar as the conventional
substrate interaction. Even at FFA levels as high as process. After digestion and pressing, the oil together with
59.65%, DAG is still relatively low, being only slightly the fruit liquor is passed to a desander and a two-phase
higher than the low FFA oils. In fact, 8% is the highest separation, removing most of the vegetative material. The
limit observed for DAG in most CPOs. The observations oil and fruit liquor is retained in an evaporator, where the
appeared contradictory to expectations of high DAG to be fruit liquor and water from steam sterilisation of bunches
found in such highly hydrolysed oils. It is hypothesised are removed via a falling film evaporator. The oil obtained
that whatever DAG and MAG formed, they are being is centrifuged and dried as before. The difference between
quickly hydrolysed to FFA. Sambanthamurthi et al. this process and that of the current commercial mill pro-
(1995) discussed evidence of an active endogenous lipase cess is in the use of the evaporator to remove any water
in the oil palm mesocarp which is located in the oil body. present either inherently in the fruits or introduced through
While in vitro experiments found the lipase maximum sterilisation. On the other hand, in current commercial
activity at 18°C, an in vivo assay showed maximum activ- process, a large quantity of water is added to dilute the
ity at 5°C. Further evidence of low-temperature activation oil to allow better separation of the oil from other
 6 M.R. Ramli et al.

Table 3. pH of crude palm oil and other properties: effect on the formation of 3-MCPD esters (ME).

Sample FFA (wt %) DAG (wt %) pH ME (mg kg−1)

Commercial sample number 1 2.66 ± 0.03a 5.5 ± 0.2a 6.2 0.84 ± 0.05
Commercial sample number 2 2.66 ± 0.04a 5.3 ± 0.1a 5.5 0.99 ± 0.03
Commercial sample number 3 3.66 ± 0.06b 5.4 ± 0.1a 7.1 < LOD
CPO, from evaporation process 3.04 ± 0.02c 5.6 ± 0.2a 4.0 12.96 ± 0.76
Notes: Values are the mean ± SD of duplicate analysis.
Values within a column with different superscript letters are significantly different (p < 0.05).
LOD = 0.25 m g kg−1.

vegetative material. The evaporator water from this experi- bunch. Some mills have installed additional solvent plants
mental process is found to have a very high acidity with a to extract further the oil retained in the fibres. PFOs are sold
pH 3. Interestingly, the CPO obtained from the process has for industrial uses, animal feeds or for biodiesel production.
also high acidity with pH 4, while in comparison samples They contain higher quantities of carotenes (1400–1600
from commercial processes showed a pH of about 5–7 ppm) and tocols (1700–2600 ppm) (Subramaniam et al.
(Table 3). It has been generally observed that palm oil 2013) and are potentially good raw materials for extraction
Downloaded by [University of Victoria] at 16:05 25 April 2015

mill effluent (POME), which is the effluent waters removed of these components. However, the possibility of such oils
during the oil extraction process, is acidic in nature (Igwe & being added to normal palm oil is considered as it enhances
Onyegbado 2007; Irenosen et al. 2014). The sample of the oil extraction rate (OER) of palm oil production at the
CPO with pH 7 was found to have a lower ME upon heat particular mill. With a residual oil recovery system (RORS)
treatment. The results pointed to an acidic nature in CPO, installed in a mill, oil recovery from the mesocarp ranging
which is generally mainly reduced during its extraction from 0.15% to 0.45% per tonne of FFB has been reported
process when there is much water used in the clarification (Subramaniam et al. 2013). The effects of blending a small
step. The acidity is due to its contact with vegetative proportion of PFO to normal oils were examined. In this
materials of the palm fruit bunch, which are generally experiment, the PFO was added in small proportions to
acidic in nature. In conventional mills, the acidity in the CPO. Results in Figure 3 show that PFO had a significant
fruits is further diluted by water just before the clarification (p < 0.05) effect on the formation of ME. DAG level was
process, and therefore is washed out. Any further washing considered high (7.9%) compared with normal CPO. A
carried out by some mills will have removed the acidity blend of 5% PFO into CPO would not contribute much to
completely and also remove chlorides to a minimum level. the DAG level. On the other hand, the pH of the PFO is
Oil obtained from the experimental process (Figure 2), found to be a low value of 4.6. Blending this oil to CPO
when heated to 260°C for 1 h, showed higher contents of would be detrimental to ME formation. The results con-
ME than any commercial CPO as shown in Table 3. The firmed that acidity is an important factor contributing to the
high ME being formed would be caused by a higher ease of formation of ME when the oil is heated to a high
reactivity of the lipids, resulting from the acidity present. temperature.
Reactivity is enhanced when acidic conditions are present
(Hamlet et al. 2011) as indicated in mechanisms of forma-
tion. Also, work by Ramli et al. (2011) has indicated the
effect of acidity from the degumming agent and bleaching
clays, providing further evidence of the effect of acidity on
the formation of ME.
The screening and clarification steps in palm oil mills
may need to be optimised for better reduction of acidity
level or, alternatively, an additional water wash step could
be introduced into the process. In areas where soils are
acidic, harvested fruit bunches that fall to the ground
could also pick up acidic soil. It would be a good idea
to wash the fruit bunches on conveyor belts to remove the
additional acidity.

Effect of oils from palm fibre

Figure 3. Effect of PFO in the formation of 3-MCPD esters
Palm fibres are the by-products of extraction of CPO. These (ME) in crude palm oil. Mean of duplicate analysis ± SD. Points
are the vegetative materials from the mesocarp and empty with different letters are significantly different (p < 0.05).
 Food Additives & Contaminants: Part A 7

Conclusions interspecific hybrid (E.O. x E.G.). J Sci Food Agric.

93:674–680.
While DAG may appear to be a culpable precursor of ME,
Craft BD, Nagy K. 2012. Mitigation of MCPD-ester and glycidyl
their role in several situations as given in fruit experiments ester levels during the production of refined palm oil. Lipid
give some doubts as to their impact on formation. This Technol. 24:155–157.
paper presents some evidence for the case against making Craft BD, Nagy K, Sandoz L, Destaillats F. 2012. Factors
DAG content of palm oil the inherent critical factor for the impacting the formation of monochloropropanediol
(MCPD) fatty acid diesters during palm (Elaeis guineensis)
formation of ME. It is noted that the acidity of CPO is the oil production. Food Addit Contam: Part A. 29:354–361.
more important factor to consider; it should be removed Crews C, Chiodini A, Granvogl M, Hamlet C, Hrnčiřík K,
during processing steps as a way towards reducing the Kuhlmann J, Lampen A, Scholz G, Weisshaar R, Wenzl T,
formation of these esters. The chlorides present can also Jasti PR, Seefelder W. 2013. Analytical approaches for
be removed in the process. The addition of water during MCPD esters and glycidyl esters in food and biological
samples: a review and future perspectives. Food Addit
the process has already been included in most mills and Contam: Part A. 30:11–45.
only needs further optimisation to enhance the better Destaillats F, Craft BD, Sandoz L, Nagy K. 2012. Formation
reduction of acidity. The possible inclusion of other mechanisms of Monochloropropanediol (MCPD) fatty acid
types of extracted oils via solvents or additional pressings diesters in refined palm (Elaeis guineensis) oil and related
should be thoroughly discouraged. fractions. Food Addit Contam: Part A. 29:29–37.
Ermacora A, Hrnčiřík K. 2013. A novel method for simultaneous
Downloaded by [University of Victoria] at 16:05 25 April 2015

monitoring of 2-MCPD, 3-MCPD and glycidyl esters in oils

Disclosure statement and fats. J Am Oil Chem Soc. 90:1–8.
Ermacora A, Hrnčiřík K. 2014. Development of an analytical
No potential conflict of interest was reported by the authors. method for the simultaneous analysis of MCPD esters and
glycidyl esters in oil-based foodstuffs. Food Addit Contam:
Part A. 31:985–994.
Funding Federal Institute for Risk Assessment. 2009. BfR
The authors thank the Director-General for financially supporting Method_82_FC-008-01: Determination of 3-MCPD fatty
the study and for subsequent permission to publish this research acid esters in edible oils and solid fats by GC-MS – An
effort. indirect determination by detection of free 3-MCPD released
from 3-MCPD-esters by acid hydrolysis and by derivatiza-
tion with phenylboronic acid. Dahlem: BfR Press.
Franke K, Strijowski U, Fleck G, Pudel F. 2009. Influence of
References chemical refining process and oil type on bound 3-chloro-1,
Abd. Majid R, Mohammad AW, Choo YM. 2012. Properties of 2-propanediol contents in palm oil and rapeseed oil. Food
residual palm pressed fibre oil. J Oil Palm Res. 24:1310– Sci Tech. 42:1751–1754.
1317. Freudenstein A, Weking J, Matthäus B. 2013. Influence of pre-
Abd. Razak RA, Kuntom A, Siew WL, Ibrahim NA, Ramli MR, cursors on the formation of 3-MCPD and glycidyl esters in a
Hussein R, Nesaretnam K. 2012. Detection and monitoring model oil under simulated deodorization conditions. Eur J
of 3-monochloropropane-1, 2-diol (3-MCPD) esters in cook- Lipid Sci Tech. 115:286–294.
ing oils. Food Control. 25:355–360. Fronimaki P, Spyros A, Christophoridou S, Dais P. 2002.
American Oil Chemists’ Society. 2013a. Official methods Cd 29a- Determination of the diglyceride content in Greek virgin
13, Sampling and analysis of commercial fats and oils, 2- and olive oils and some commercial olive oils by employing
31
3-MCPD fatty acid esters and glycidyl fatty acid esters in P NMR Spectroscopy. J Agric Food Chem.
edible oils and fats by acid transesterification. Urbana, IL: 50:2207–2213.
AOCS Press. Hamlet CG, Asuncion L, Velíšek J, Doležal M, Zelinková Z,
American Oil Chemists’ Society. 2013b. Official methods Cd Crews C. 2011. Formation and occurrence of esters of 3-chlor-
29b-13, Sampling and analysis of commercial fats and oils, opropane-1,2 diol (3-CPD) in foods: what we know and what
Determination of bound monochloropropanediol- (MCPD) we assume. Eur J Lipid Sci Tech. 113:279–303.
and bound 2, 3-epoxy-1-propanol (glycidol-) by gas chro- Hrnčiřík K, Ermacora A. 2011. Formation of 3-MCPD esters in
matography/mass spectrometry (GC/MS). Urbana, IL: palm oil: Effect of partial acylglycerols. Informative Meeting
AOCS Press. of 3-MCPD and glycidyl esters in Foodstuffs, organized by
American Oil Chemists’ Society. 2013c. Official methods Cd OVID and BLL, Jan 18. Berlin: Association of Oilseed
29c-13, Sampling and analysis of commercial fats and oils, Processing.
Fatty-acid-bound 3-chloropropane-1,2-diol (3-MCPD) and Hrnčiřík K, van Duijn G. 2011. An initial study on the formation
2,3- epoxi-propane-1-ol (glycidol), determination in oils of 3-MCPD esters during oil refining. Eur J Lipid Sci Tech.
and fats by GC/MS (Differential measurement). Urbana, 113:374–379.
IL: AOCS Press. Igwe JC, Onyegbado CC. 2007. A review of palm oil mill
American Oil Chemists’ Society (AOCS). 1998. Official meth- effluent (POME) water treatment. Global J Environ Res.
ods and recommended practices of the American Oil 1:54–62.
Chemists’ Society. 5th ed. Urbana, IL: AOCS Press. Irenosen OG, Oluyemi AS, Korede AO, Samuel AS. 2014.
Cadena T, Prada F, Perea A, Romero HM. 2013. Lipase activity, Integration of physical, chemical and biological methods
mesocarp oil content and iodine value in oil palm fruits of for the treatment of palm oil mill effluent. Sci J Anal
Elaeis Guineensis (E.G.), Elaeis Oleifera (E.O.), and the Chem. 2:7–10.
 8 M.R. Ramli et al.

Lacoste F, Joffre F, Soulet B, Griffon H. 2010. 3-MCPD esters In: Kader JC, Mazliak P, editor. Plant Lipid Metabolism.
in edible oils: analytical aspects, 101st AOCS Dordrecht: Kluwer Academic Publications.
Annual Meeting and Expo; May 16-19; Urbana, IL: AOCS Siew WL, Kuntom A, Ibrahim NA, Ramli MR, Abd. Razak RA.
Press. 2012. The possible mitigation procedures for the reduction
MacMahon S, Begley TH, Diachenko GW. 2013. Occurrence of of chloropropanol esters and related compounds. Palm Oil
3-MCPD and glycidyl esters in edible oils in the United Develop. 57:21–27.
States. Food Addit Contam: Part A. 30:2081–2092. Siew WL, Ng WL. 1995. Diglyceride content and composition
Matthäus B, Pudel F. 2013. Mitigation of 3-MCPD and glycidyl as indicators of palm oil quality. J Sci Food Agric. 69:73–79.
esters within the production chain of vegetable oils espe- Subramaniam V, Menon NR, Sin H, Choo YM. 2013. The
cially palm oil. Lipid Technol. 25:151–155. development of a residual oil recovery system to increase
Matthäus B, Pudel F, Fehling P, Vosmann K, Freudenstein A. the revenue of a palm oil mill. J Oil Palm Res. 25:116–122.
2011. Strategies for the reduction of 3-MCPD esters and Weißhaar R. 2008. 3-MCPD-esters in edible fats and oils – a new
related compounds in vegetable oils. Eur J Lipid Sci Tech. and worldwide problem. Eur J Lipid Sci Tech. 110:671–672.
113:380–386. Weißhaar R. 2011. Fatty acid esters of 3-MCPD: Overview of
Nagy K, Sandoz L, Craft B, Destaillats F. 2011. Mass-defect occurrence and exposure estimates. Eur J Lipid Sci Tech.
filtering of isotope signatures to reveal the source of chlori- 113:304–308.
nated palm oil contaminants. Food Addit Contam: Part A. Yamazaki K, Ogiso M, Isagawa S, Urushiyama T, Ukena T,
28:1492–1500. Kibune N. 2013. A new, direct analytical method using
Pudel F, Matthäus B, Freudenstein A, Rudolph T. 2012. LS-MS/MS for fatty acid esters of 3-chloro-1,2-propanediol
Mitigation of 3-MCPD and glycidyl esters in refined palm (3-MCPD esters) in edible oils. Food Addit Contam: Part A.
Downloaded by [University of Victoria] at 16:05 25 April 2015

oils, 103rd AOCS Annual Meeting and Expo, April 29-May 30:52–68.
2. Urbana, IL: AOCS Press. Yeoh CM, Phuah ET, Tang TK, Siew WL, Abdullah LC, Thomas
Ramli MR, Siew WL, Ibrahim NA, Hussein R, Kuntom A, Abd Choong SY. 2014. Molecular distillation and characterization
Razak RA, Nesaretnam K. 2011. Effects of degumming and of diacylglycerol-enriched palm olein. Eur J Lipid Sci Tech.
bleaching on 3-MCPD esters formation during physical 116:1654–1663.
refining. J Am Oil Chem Soc. 88:1839–1844. Zelinková Z, Svejkovská B, Velíšek J, Doležal M. 2006. Fatty
Sambanthamurthi R, Chong CL, Oo KC, Yeo KH, Rajan P. 1991. acid esters of 3-chloropropane-1,2-diol in edible oils. Food
Chilling-induced lipid hydrolysis in the oil palm (Elaeis Addit Contam: Part A. 23:1290–1298.
guineensis) mesocarp. J Exp Bot. 42:1199–1205. Zhou H, Jin Q, Wang X, Xu X. 2014. Direct measurement of 3-
Sambanthamurthi R, Oo KC, Parman SH. 1995. Factors affecting chloropropane-1,2-diol fatty acid esters in oils and fats by
lipase activity in the oil palm (Elaeis Guineensis) mesocarp. HPLC method. Food Control. 36:111–118.