Formation Mechanisms

Formation Mechanisms

1 Formation Mechanisms Brian D. Craft and Frédéric Destaillats  ■  Nestlé Research Center, Food Science and ­Technology Department, Vers-chez-les-Blan...

651KB Sizes 0 Downloads 6 Views

1 Formation Mechanisms Brian D. Craft and Frédéric Destaillats  ■  Nestlé Research Center, Food Science and ­Technology Department, Vers-chez-les-Blanc, Lausanne, Switzerland

Introduction Since the publication of Zelinkova et al. (2006) and the heightened awareness of fatty esters of monochloropropanediol (MCPD-FE) in refined edible oils, the circumstances surrounding their formation have been subject to large amounts of speculation. For instance, some researchers speculated that precursors for MCPD-FE formation (e.g., chlorine and diacylglycerols) are present in partially refined oils (Franke et al., 2009). Other researchers suspected that the refining process results in the uncontrolled introduction of certain compounds to the oils (e.g., inorganic chlorides in the stripping stream), so it should be the first place to explore mitigation strategies (Pudel et al., 2011). Further, amidst analytical developments in MCPD-FE quantification, another family of compounds was discovered in refined edible oils, namely the fatty esters of glycidol (G-FE). G-FE were found to be partially responsible for inflation of the results of MCPD-FE quantifications due to the generation of artifacts during sample preparation before analysis using indirect methods (Weißhaar and Perz, 2010). Despite these early hurdles, some recent breakthroughs were made by Nagy et al. (2011) and Destaillats et al. (2012a, 2012b) on the formation mechanisms of both MCPD-FE and G-FE during palm oil refining. Within this chapter we will take a focused look at the status of the literature to date as it pertains to the formation pathways of MCPD-FE and G-FE in refined edible oils. Critical topics will be covered, including the most prevalent precursor compounds and detailed formation mechanisms responsible for the generation of these process contaminants during oil production and refining. Because both MCPD-FE and G-FE have been found in the highest average abundance in palm oil, the majority of the research reviewed herein involves crude palm oil production and refining.

MCPD Esters Precursors

Chlorine is ubiquitous in nature. Thus, one can speculate about a wide variety of chlorine sources, whether organic or inorganic, as potential precursor compounds to MCPD-FE formation during edible oil production. Further, a host of lipid types and compositions (e.g., acylglycerols, phospholipids, glycolipids) are available in the raw 7

8  ■  B.D. Craft and F. Destaillats

OH O

HN

OH Cl

HO OH

Elemental composition C42H83O4NCl C42H85O4NCl C42H83O5NCl C42H85O5NCl C42H85O6NCl

Exact m/z 700.60280 702.61807 716.59723 718.61357 734.60809

Figure 1.1  Proposed structure and chemical formulas of an organochlorine family of compounds found in crude palm oil. (Reprinted with permission from Nagy et al., 2011.)

materials used to produce edible oils. Many of these lipids could theoretically interact with chlorine sources and result in the formation of MCPD-FE during oil refining. The critical precursors responsible, however, are mostly dependent on the oil type, quality, and, to a lesser degree, the circumstances of manufacture, as will be described below. Given that refined palm oil specifically has been shown to contain significant levels of MCPD-FE (2.7 mg/kg) (Weißhaar, 2011), it has been exclusively used as a model matrix in the literature. The first question often raised regarding MCPD-FE precursors is the origin of chlorine involved in the MCPD-FE reaction during oil refining and why it is potentially more abundant in crude palm oil (CPO) in comparison to other crude vegetable oils (Matthäus, 2012). Recently, Nagy et al. (2011) demonstrated that many sources of covalently bound inorganic chlorine exist at ppm (mg/kg) levels in crude palm oil, including FeCl3, FeCl2, MgCl2, and CaCl2. Further, a “pool” (n = 300) of organic monochlorinated compounds was also found and it appears to undergo a transformation throughout the stages of oil refining with certain compounds being formed while others decompose over time. In order to elucidate the composition of the more predominant chlorine “donor” compounds, Nagy et al. (2011) used LCMSn in the framework of model experiments. Therein the authors identified a specific family of chlorinated compounds present in both the lipids extracted from handpicked Malaysian palm fruits and commercially procured CPO samples. Figure 1.1, taken from Nagy et al. (2011), shows the proposed structure and chemical formulas of this monochlorinated family of compounds. The authors suggest that given their

Formation Mechanisms  ■   9

structural similarities to phytosphingosines, it is perhaps more likely that the chlorine donors identified are endogenous plant metabolites as opposed to chlorinated contaminants introduced to the oil palm’s direct environment during growth and maturation. Extrapolating from this hypothesis, one might tend to the logic that the raw materials intended for production of each edible vegetable oil have their own reactive-chlorine pool capable of donating chlorine during oil refining and ultimately resulting in MCPD-FE generation. In terms of the most predominant lipid precursors of MCPD-FE in edible oils, Zelinkova et al. (2006) proposed that there may be a link between the content of diacylglycerols (DAGs) in refined edible oils and their MCPD-FE levels. This assertion was likely due to the fact that the highest MCPD-FE levels were observed within the fruit pulp oils analyzed. Fruit pulp oils, such as olive and palm, are known for having high DAG contents compared to seed oils, due to the greater prevalence of lipolytic reactions during harvest (Dijkstra and Segers, 2007). This correlation, however, has been disproved in recent literature (Hrncˇirˇík and van Duijn, 2011; Matthäus et al., 2011). Although DAGs could potentially react with chlorine donors during oil refining and result in the formation of MCPD-FE, they are not the most critical lipid precursors of these process contaminants. Further, lipids such as monoacylglycerols (MAGs), phospholipids, and glycolipids are largely removed during oil degumming and are not present during the later stages of oil refining (Dijkstra and Segers, 2007). Because the bulk of MCPD-FE have been shown to be generated during oil deodorization (Franke et al., 2009; Ramli et al., 2011), the entirety of the aforementioned lipid classes is not expected to be greatly involved in MCPD-FE formation reactions. This of course leaves the triacylglycerols (TAGs) up for consideration. TAGs can represent more than 90–95% (v/v) of refined vegetable oils, whether pressed from nuts, seeds, or fruit pulps. TAGs are, therefore, the most logical critical lipid precursor available for MCPD-FE formation during oil deodorization. The results of in vitro thermal reaction experiments carried out by Destaillats et al. (2012b) appear to confirm this hypothesis. Destaillats et al. (2012b) demonstrated in controlled conditions that TAGs, not DAGs, are preferentially reacting with chlorine donors to form MCPD-FE. Formation Pathways

As previously mentioned, MCPD-FE are formed almost completely during the deodorization unit operation of edible oil refining. As such, the majority of scientific research conducted on MCPD-FE formation pathways has been carried out either in conditions mimicking oil deodorization or within bench-top, pilot scale, or commercial deodorization units. Destaillats et al. (2012b) showed through in vitro experiments

10  ■  B.D. Craft and F. Destaillats

120

Relative abundance (%)

100 80 60 40 20 0 100

120

140

160

180

200

220

240

260

Temperature (ºC) Sum of organochlorines

Sum of MCPDs

Figure 1.2  The simultaneous decomposition of some key organochlorines (n = 8) monitored and the formation of 3-MCPD diesters during the thermal treatment of crude palm oil. (Reprinted with permission from Nagy et al., 2011.)

that 3-MCPD diesters, which are the most predominant form of MCPD-FE in refined oils (Seefelder et al., 2008), can be generated at temperatures as low as 180–200 °C. It follows from this observation that within either type of edible oil refining (chemical or physical), the typical deodorization conditions employed strongly favor MCPD-FE formation. For example, typical chemical refining operations for palm oil involve a deodorization step at around 240 °C, whereas physical refining operations can involve a deodorization step at even higher temperatures (260–270 °C) in order to remove excess free fatty acids (FFA) (Dijkstra and Segers, 2007). In order to determine the origin of chlorine involved in MCPD-FE formation during oil refining, Franke et al. (2009) and Hrncˇirˇík and van Duijn (2011) examined the content of chloride ions present in oils pre- and post-deodorization. They then attempted to correlate these levels with the ultimate levels of MCPD-FE observed in the fully refined oils. Unfortunately, little or no correlation was observed. Pudel et al. (2011) attempted to determine if the chlorine responsible for MCPD-FE formation was originating from the stripping steam applied during oil deodorization, but with a similar negative result. Only recently, Nagy et al. (2011) demonstrated that both inorganic and organic chlorinated compounds are present at ppm (mg/kg) levels

Formation Mechanisms  ■   11

in partially refined edible oils. Further, the authors demonstrated that the thermal decomposition of organic chloride–containing compounds in CPO was found to coincide strongly with the evolution of 3-MCPD diesters during the thermal treatment of CPO (see Figure 1.2). Given that 3-MCPD diesters are the predominant class of MCPD-FE in refined oils, the study of Nagy et al. (2011) has proven causality behind the greatest portion of MCPD-FE formed during the deodorization of edible oils. A given organic (or inorganic) chlorinated compound may decompose at a certain temperature, above which the released reactive chlorine can then interact with TAG and result in the formation of MCPD-FE. The fact that the graphic in Figure 1.2 crosses at 180 °C, whereas the formation temperature of MCPD-FE in the standardized in vitro experiments of Destaillats et al. (2012b) was 180–200 °C, is quite a coincidence. In order to uncover whether a certain reactive chlorine intermediate was most responsible for MCPD-FE formation, Nagy et al. (2011) monitored the decomposition of the organochlorines found in crude palm oils via LC-MSn experiments. The authors reported that hydrogen chloride (HCl) is a typical thermal decomposition product of the organochlorine pool monitored. Thus, HCl could prove to be the predominant form of reactive chlorine responsible for MCPD-FE formation during oil deodorization. Several potential formation mechanisms of MCPD-FE have been recently reviewed in the literature (Hamlet et al., 2011; Rahn and Yaylayan, 2011b). Given that many of the past mechanistic studies of MCPD-FE formation were carried out in hydrophilic media, it has long been suggested that TAG underwent hydrolysis to DAG as a first step in the MCPD-FE formation reaction. DAGs then reacted with chlorine donor compounds resulting in the formation of acyloxonium ion intermediates and eventual nucleophilic substitution of chloride ion on the glycerol backbone (Velisek et al., 2002). More recently, however, it was proven that TAGs can act directly as a substrate for MCPD-FE formation (Destaillats et al., 2012b). Further, Rahn and Yaylayan (2011a) proved with infrared (IR) spectroscopy that heating TAG in the presence of Lewis acids can lead to cyclic acyloxonium ion formation. Thus, is it theoretically plausible to assume that the two pathways could be favored in the case of hydrophobic systems like oil deodorization. One pathway involves the reaction of TAG directly with HCl formed by thermal degradation of chlorine donors, nucleophilic substitution of chloride ion on the glycerol backbone to form an MCPD-FE, and finally the release of a fatty acid. The second mechanism first involves the formation of an acyloxonium ion intermediate compound. These two MCPD-FE formation mechanisms were summarized by Destaillats et al. (2012b) and appear in Figure 1.3. Destaillats et al. (2012b) also proved through in vitro experiments that the MCPD-FE formation reaction is regioselective and preferential to the sn-1(3) positions on the glycerol backbone. Because the critical precursors and predominant formation mechanisms have been elucidated in the case of palm oil production and refining, it was then possible

12

O

O

R2

O O

R3

O

Cl

R1

O O O

R1

R1

R2

O

O

O O

+

O

O

O

R3

H

O

O

R2

Cl +

R2

O O

O O

+

O R3

H

R3

O H

R1

O O O

+

Cl

O R1

O

R2

O+

O

O

O

+

R1

O O

O

O

O R2

Cl

R3

H

O

Figure 1.3  Proposed mechanisms for the formation of MCPD-FE from TAG at high temperatures in the presence of HCl evolved from the thermal decomposition of trace organochlorines. Two putative pathways including the formation of reactive cyclo acyloxonium ion intermediate (upper panel) or a direct nucleophilic substitution reaction (lower panel) are displayed. Both pathways result in the formation of an MCPD-FE molecule and the release of a fatty acid. (Reprinted with permission from Destaillats et al., 2012b.)

Acyloxonium ion pathway

R1

O

H

Cyclic acyloxonium ion pathway

Formation Mechanisms  ■   13

to speculate about the potential predominant root causes responsible for the manifestation of MCPD-FE therein. Nagy et al. (2011) demonstrated that chlorinated compounds (n = 300 found) can be monitored in crude palm oils utilizing mass-defect filtering of isotope signatures. Further, the authors then discovered that these same compounds can be utilized to segregate commercial palm oil samples (n = 26) based on their processing stage using multivariate statistical analysis (see Figure 1.4). The grouping of commercial samples from crude to refined-bleached to refined-bleacheddeodorized, as shown in Figure 1.4, suggests that chlorinated compounds undergo a transformation throughout palm oil production and refining. This truth might lend to the logic that chlorine is similarly transformed throughout the agricultural process involved in palm oil growth, maturation, and harvest. Figure 1.5 is a schematic taken from Craft et al. (2012a) and serves as a root-cause analysis of MCPD-FE formation during refined palm oil production. It summarizes potential locations/sources for the influx of chlorine from the environment, accumulation of the chlorine in the palm Crude RB RBD 4 2 0 Factor 3

2 1

–2

0

–4

–1 –3

–2

Factor 2

–2 –1 Factor 1

0

–3 1

Figure 1.4  Principle component analysis of chlorine-containing compounds (n = 300) present in crude, partially, and fully refined palm oil samples (n = 26). The grouping of oil samples based on refining stage [i.e., from crude to refinedbleached (RB), to refined-bleached-deodorized (RBD)] suggests that these compounds undergo a transformation during oil processing. (Reprinted with permission from Nagy et al., 2011.)

14  ■  B.D. Craft and F. Destaillats

Palm Oil Production Formation of lipophilic organochlorines during fruit bunch sterilization Biosynthesis of hydrophilic organochlorines in palm fruits

(4)

palm oil triacyglycerols at high temperatures

Oil Palm Growth and Maturation

Accumulation of inorganic chloride in the oil palm plant

Reaction of liposoluble

(5) organochlorines with

(3)

MPCD deister formation

(2) (1)

Influx of inorganic chloride from the environment (e.g., KCl, NH4Cl, MgCl2, FeCl3, FeCl2)

Soil, Fertilizer, and Irrigation Figure 1.5  Root-cause analysis of the factors involved in the formation of MCPD-FE during refined palm oil production, including (1) chlorine influx from the environment, (2) accumulation of inorganic chloride in the plant, (3) bioconversion of inorganic chlorides to organochlorines in palm fruits, (4) formation of liposoluble organochlorines during fruit bunch sterilization, and (5) reaction of liposoluble organochlorines with TAG in palm oil during oil deodorization. (Reprinted with permission from Craft et al., 2012a.)

plant and fruits, and transformation of chlorine into more liposoluble forms during CPO production, followed by the resultant formation of MCPD-FE during oil refining. A similar type of root-cause analysis could prove beneficial in the case of other refined vegetable oil crops and the assessment of their potential for production of MCPD-FE after harvest and refining.

Glycidyl Esters Precursors

Given the history behind the discovery of G-FE in edible oils, researchers often pooled G-FE with MCPD-FE (often termed MCPD esters and related compounds)

Formation Mechanisms  ■   15

(Pudel et al., 2011). This fact led to the assumption that these compound families were very closely related and potentially subject to interconversion. As such, both MCPD-FE and G-FE were thought to share the same precursors. This assumption has been disproved recently. Craft et al. (2012b) reported a strong positive correlation between DAG levels alone and the amount of G-FE contained in refined palm oils. Because fruit pulp oils like palm oil naturally contain higher levels of DAG (≥3–4%) (D’Alonzo et al., 1982), this may suggest possible causality as to why their deodorized counterparts also contain higher levels of G-FE in comparison to seed oil crops, as demonstrated by Weißhaar and Perz (2010). The strong correlation between DAG contents and G-FE formation has essentially been validated on an industrial scale. Watkins (2009) recently reported that high-DAG oils marketed for health and wellness had to be taken off the market purportedly due to “high levels” of G-FE. Further, Masukawa et al. (2010) demonstrated that commercially refined oils rich in DAG (87%) can contain more than 10-fold greater G-FE levels relative to oils with lower DAG (3.9–6.8%) contents. The only other proven lipid precursor to G-FE is MAG. Destaillats et al. (2012a) showed that both MAG and DAG can result in formation of G-FE upon thermal treatment, although formation from DAG is most favored. The fact that MAG levels in refinedbleached oils are often quite low (mean < 0.1%) (Goh and Tims, 1985), however, renders this reaction route less significant in the case of edible vegetable oils. Formation Pathways

The formation of G-FE from DAG during oil deodorization was shown to be significant at temperatures at or above 230 °C (Hrncˇirˇík and van Duijn, 2011). Craft et al. (2012b) confirmed this temperature of G-FE formation by deodorizing refinedbleached palm oil from 180–240 °C at increments of 20 °C. G-FE values were reported as the sum of the glycidyl-palmitate (16:0 G-FE), stearate (18:0 G-FE), oleate (18:1 G-FE), linoleate (18:2 G-FE), and linolenate (18:3 G-FE) species (see Figure 1.6). As seen in Figure 1.6, the formation of G-FE proceeds at an exponential rate above 220 °C. To further demonstrate the link between DAG and G-FE formation during edible oil refining, Craft et al. (2012b) spiked refined cottonseed oil (~100% TAG) with a DAG standard in concentrations ranging from 1% to 5% and subjected it to thermal treatment at 235 °C for two hours within both an ampoule system and a bench-top deodorization unit (see Figure 1.7 on page 17). As seen in Figure 1.7, G-FE formation appears to increase exponentially when DAG contents exceed 3–4% of total lipids. Further, the marked reduction in the levels of G-FE found in the deodorization trials, when compared to the ampoule system, suggest that a large amount of G-FE are being stripped into the deodorizer distillate. This is logical given the typical range of molecular weights of G-FE species found in palm oil (i.e., 284.4 g/mol for glycidyl myristate to 340.5 g/mol for glycidyl stearate). This could mean that the relative ­concentration of

16  ■  B.D. Craft and F. Destaillats

2.0 1.8

Glycidyl esters content (ppm)

1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

Nondeodorized

180 °C

200 °C

220 °C

240 °C

Figure 1.6  Formation of G-FE during the thermal treatment of refined-bleached palm oil heated at different temperatures for two hours in a bench-top deodorization unit. These G-FE totals are a sum of the glycidyl-palmitate (16:0 G-FE), stearate (18:0 G-FE), oleate (18:1 G-FE), linoleate (18:2 G-FE), and linolenate (18:3 G-FE) species. (Reprinted with permission from Craft et al., 2012b.)

G-FE manifested in refined edible oils is dependent on the efficiency of the deodorizer units in the oil factories (e.g., vacuum pressure, stripping medium). Destaillats et al. (2012a) proposed that G-FE can be formed from the thermal treatment of DAG proceeding through an intramolecular rearrangement mechanism followed by the elimination of a fatty acid (see Figure 1.8 on page 18). The authors also proposed a similar mechanism for the formation of G-FE from the thermal treatment of sn-1(3)-MAG resulting in the elimination of a water molecule. Unexpectedly, while conducting G-FE formation experiments and subsequent quantifications, Destaillats et al. (2012a) discovered some very prominent isomers of G-FE, namely oxopropyl esters. The formation of oxopropyl esters was found to occur at temperatures as low as 150 °C and plateaus around 200 °C, at which point G-FE proceed to be formed at an increasing rate (exponentially at temperatures > 230 °C). Figure 1.9 (on page 18), taken from Destaillats et al. (2012a), summarizes this phenomenon of simultaneous oxopropyl ester and G-FE production with increased thermal treatment. The authors proposed that oxopropyl esters are formed from DAG through a similar intramolecular

Formation Mechanisms  ■   17

Relative abundance to internal standard

5 4 3 2 1 0 0

1

2 3 4 Level of diacylglycerol in oil (%) Heated in ampoules

5

Deodorized

Figure 1.7  Influence of DAG concentration (standardized at 1–5%) on the formation of G-FE during the thermal treatment of refined cottonseed oil at 235 °C for two hours. The greater presence of G-FE in the ampoule experiments (dark bars) when compared to the bench-top deodorizer trials (light bars) suggests that a significant portion of G-FE are stripped off in the deodorizer. (Reprinted with permission from Craft et al., 2012a.)

rearrangement reaction as G-FE, but this reaction proceeds through fatty acid loss followed by the formation of an enol intermediate that then undergoes tautomerization. Craft et al. (2012b) compiled analytical data on the occurrence of G-FE in a variety of commercially refined-bleached-deodorized (RBD) palm oils used in their experiments to determine if relationships were present. Figure 1.10 (on page 19), taken from Craft et al. (2012b), shows the relationship between G-FE and DAG levels in palm oil and palm olein samples (n = 15; panel A). Analysis of a subgroup of samples obtained from the same refinery (n = 6; Figure 1.10, panel B) show a strong positive correlation (R2 ~0.8). Craft et al. (2012b) also developed a predictive model that directly correlates the level of FFA in CPO to the DAG contents of fully refined palm oil. For example, a DAG concentration of 3% in refined palm oil was found to be equivalent to 1.2–1.3% FFA in the initial CPO. Making such a correlation is practical given that CPO is often purchased based on the level of FFA as a general indicator of quality (i.e., the lower the FFA, the higher the quality). From this information the answer seems clear: The higher quality the CPO used to produce refined palm oil, the

O R1

O

R3

OO H

∆ O R3

O O

O

R1

R3



O O +

O

R1



O O + H

+

R3

O O H

O R3

O O

O

+

O

O

R1



O O H

H R3

Figure 1.8  Proposed mechanism for the formation of G-FE from DAG at high temperatures. The DAG molecule undergoes an intermolecular rearrangement followed by the loss of a fatty acid. (Reprinted with permission from Destaillats et al., 2012a.)

Glycidyl vaccenate relative abundance

1

0.1

0.01

0.001 140

160

180

200

220

240

260

280

Temperature (˚C) Sum of isomers

Oxopropyl ester

Glycidyl ester

Figure 1.9  Relative formation of oxopropyl esters and G-FE from DAG at different temperatures. Note that the ordinate axis is log scale and both individual and sum abundance of isomers are shown. (Reprinted with permission from Destaillats et al., 2012a.) 18

Formation Mechanisms  ■   19

A 20.0

Glycidyl esters (ppm)

18.0 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 3.8

4.3

4.8

5.3

5.8

6.3

6.8

7.3

7.8

DAG level (g/100g of oil) B 18.5

Glycidyl esters (ppm)

17.5 16.5 15.5 14.5 13.5 12.5 11.5 10.5

R2 = 0.79

9.5 3.8

4.3

4.8

5.3

5.8

DAG level (g/100g of oil)

Figure 1.10  Relationship between G-FE and DAG levels in palm oil and palm olein samples (n = 15, panel A). Analysis of a subgroup of samples obtained from the same refinery (n = 6, panel B) shows a strong positive correlation (R2 ~0.8). (Reprinted with permission from Craft et al., 2012b.)

20  ■  B.D. Craft and F. Destaillats

lower the G-FE levels will be in their refined-bleached counterparts. This relationship would likely also prove true for other raw materials used to create refined edible oils.

Conclusions and Perspectives In summary, the most predominant formation mechanisms of MCPD-FE and G-FE in refined palm oil were revealed based on quantifications of known matrix components and the execution of well-defined experimental investigations. Once the formation mechanisms were proven, critical precursors and detailed formation pathways could be elucidated. These factors will prove essential in allowing for targeted and efficient mitigation procedures by oil producers.

References Craft, B. D.; Nagy, K.; Sandoz, L.; Destaillats, F. Factors Impacting the Formation of Monochloropropanediol (MCPD) Fatty Acid Diesters during Palm (Elaeis guineensis) Oil Production. Food Addit. Contam. 2012a, 29, 354–361. Craft, B. D.; Nagy, K.; Seefelder, W.; Dubois, M.; Destaillats, F. Glycidyl Esters in Refined Palm (Elaeis guineensis) Oil and Related Fractions. Part II: Practical recommendations for effective mitigation. Food Chem. 2012b, 132, 73–79. D’Alonzo, R. P.; Kozarek, W. J.; Wade, R. L. Glyceride Composition of Processed Fats and Oils as Determined by Glass Capillary Gas Chromatography. J. Am. Oil Chem. Soc. 1982, 59, 292–295. Destaillats, F.; Craft, B. D.; Dubois, M.; Nagy, K. Glycidyl Esters in Refined Palm (Elaeis guineensis) Oil and Related Fractions. Part I: Formation Mechanism. Food Chem. 2012a, 131, 1391–1398. Destaillats, F.; Craft, B. D.; Sandoz, L.; Nagy, K. Formation Mechanism of Monochloropropanediol (MCPD) Fatty Acid Diesters in Refined Palm (Elaeis guineensis) Oil and Related Fractions. Food Addit. Contam. 2012b, 29, 29–37. Dijkstra, A. J.; Segers, J. C. The Lipid Handbook: with CD-ROM; CRC Press: Boca Raton, FL, 2007; pp 143–262. Franke, K.; Strijowski, U.; Fleck, G.; Pudel, F. Influence of Chemical Refining Process and Oil Type on Bound 3-Chloro-1,2-propanediol Contents in Palm Oil and Rapeseed Oil. LWT—Food Sci. Technol. 2009, 42, 1751–1754. Goh, E. M.; Timms, R. E. Determination of Mono- and Diglycerides in Palm Oil, Olein and Stearin. J. Am. Oil Chem. Soc. 1985, 62, 730–734. Hamlet, C. G.; Asuncion, L.; Velisek, J.; Dolezal, M.; Zelinkova, Z.; Crews, C. Formation and Occurrence of Esters of 3-Chloropropane-1,2-diol (3-CPD) in Foods: What We Know and What We Assume. Eur. J. Lipid Sci. Technol. 2011, 113, 279–303. Hrnčiřík, K.; van Duijn, G. An Initial Study on the Formation of 3-MCPD Esters during Oil Refining. Eur. J. Lipid Sci. Technol. 2011, 113, 374–379.

Formation Mechanisms  ■   21

Masukawa, Y.; Shiro, H.; Nakamura, S.; Kondo, N.; Jin, N.; Suzuki, N.; Ooi, N.; Kudo, N. A New Analytical Method for the Quantification of Glycidol Fatty Acid Esters in Edible Oils. J. Oleo. Sci. 2010, 59, 81–88. Matthäus, B. Organic or Not Organic—That Is the Question: How the Knowledge about the Origin of Chlorinated Compounds Can Help to Reduce Formation of 3-MCPD Esters. Eur. J. Lipid Sci. Technol. 2012, 114, 1333–1334. Matthäus, B.; Pudel, F.; Fehling, P.; Vosmann, K.; Freudenstein, A. Strategies for the Reduction of 3-MCPD Esters and Related Compounds in Vegetable Oils. Eur. J. Lipid Sci. Technol. 2011, 113, 380–386. Nagy, K.; Sandoz, L.; Craft, B. D.; Destaillats, F. Mass-Defect Filtering of Isotope Signatures to Reveal the Source of Chlorinated Palm Oil Contaminants. Food Addit. Contam. 2011, 28, 1492–1500. Pudel, F.; Benecke, P.; Fehling, P.; Freudenstein, A.; Matthäus, B.; Schwaf, A. On the Necessity of Edible Oil Refining and Possible Sources of 3-MCPD and Glycidyl Esters. Eur. J. Lipid Sci. Technol. 2011, 113, 368–373. Rahn, A. K. K.; Yaylayan, V. A. Monitoring Cyclic Acyloxonium Ion Formation in Palmitin Systems Using Infrared Spectroscopy and Isotope Labelling Technique. Eur. J. Lipid Sci. Technol. 2011a, 113, 330–334. Rahn, A. K. K.; Yaylayan, V. A. What Do We Know about the Molecular Mechanism of 3-MCPD Ester Formation? Eur. J. Lipid Sci. Technol. 2011b, 113, 323–329. Ramli, M. R.; Siew, W. L.; Ibrahim, N. A.; Hussein, R.; Kuntom, A.; Razak, R. A. A.; Nesaretnam, K. Effects of Degumming and Bleaching on 3-MCPD Esters Formation during Physical Refining. J. Am. Oil Chem. Soc. 2011, 88, 1839–1844. Seefelder, W.; Varga, N.; Studer, A.; Williamson, G.; Scanlan, F. P.; Stadler, R. H. Esters of 3-Chloro-1,2-propanediol (3-MCPD) in Vegetable Oils: Significance in the Formation of 3-MCPD. Food Addit. Contam. 2008, 25, 391–400. Velisek, J.; Dolezal, M.; Crews, C.; Dvorak, T. Optical Isomers of Chloropropanediols: Mechanisms of Their Formation and Decomposition in Protein Hydrolysates. Czech J. Food Sci. 2002, 20, 161–170. Watkins, C. Kao Suspends DAG Oil Shipments. Inform 2009, 20, 689–690. Weißhaar, R. Fatty Acid Esters of 3-MCPD: Overview of Occurrence and Exposure Estimates. Eur. J. Lipid Sci. Technol. 2011, 113, 304–308. Weißhaar, R.; Perz, R. Fatty Acid Esters of Glycidol in Refined Fats and Oils. Eur. J. Lipid Sci. Technol. 2010, 112, 158–165. Zelinkova, Z.; Svejkovska, B.; Velisek, J.; Dolezal, M. Fatty Acid Esters of 3-Chloropropane-1,2-diol in Edible Oils. Food Addit. Contam. 2006, 23, 1290–1298.