Oxidative Stability of Conjugated Linoleic Acid Rich Soy Oil

ORIGINAL PAPER

Oxidative Stability of Conjugated Linoleic Acid Rich Soy Oil

Ramesh R. Yettella • Chelsey Castrodale •

Andrew Proctor

Received: 6 June 2011 / Revised: 24 September 2011 / Accepted: 14 October 2011

� AOCS 2011

Abstract This study was conducted to determine the

oxidative stability of conjugated linoleic acid rich soy oil

(CLARSO) and the effects of conjugated linoleic acid

(CLA) levels on volatile oxidation products formed during

CLARSO oxidation. CLARSO oxidative stability was

determined by gravimetric analysis, peroxide value, head-

space oxygen analysis and p-anisidine value. Volatile

oxidation compounds were analyzed by solid phase mic-

roextraction–gas chromatography with a flame ionization

detector and a mass spectrometer. CLA oxidation results

were highly dependent on analytical methods used and

oxidation parameters measured. The gravimetric study

showed a CLA concentration effect on oxidation, which

was not seen in the headspace oxygen depletion and per-

oxide value. Volatile oxidation data indicate that CLARSO

had significantly higher (p \ 0.05) levels of pentanal and

trans-2-heptenal than the other oils, but there was no sig-

nificant difference between the amounts of any volatiles

present in 8 and 15% CLARSO. This suggests that oxi-

dation was greater in CLARSO and that CLA concentra-

tion did not affect oxidation.

Keywords Conjugated linoleic acid � Lipid oxidation �Volatile oxidation compounds � SPME–GC � Peroxide

value � Photoisomerization

Abbreviations

CLA Conjugated linoleic acid

LA Linoleic acid

CLARSO CLA-rich soy oil

TAG Triacylglycerol(s)

RBD soy oil Refined, bleached and deodorized soy oil

GC–FID Gas chromatography–flame ionization

detector

PV Peroxide value

p-AV p-Anisidine value

SPME Solid phase micro extraction

CAR/PDMS Carboxen/polydimethylsiloxane

ANOVA Analysis of variance

Introduction

Conjugated linoleic acid (CLA) is a term that refers to a

collection of linoleic acid (LA) positional and geometric

isomers, with conjugated double bonds between carbon 7

and carbon 13. Currently, there is much interest in CLA

due to its anti-atherogenic [1], anti-carcinogenic [2], anti-

diabetic [3], and anti-obesity [4] health benefits. CLA is

found naturally in meat and dairy foods as a product of

ruminant bacterial fermentation. The average daily con-

sumption of CLA is estimated to be 1 g per person per day

[5]. However, about 3 g per person per day is recom-

mended to achieve optimum human health benefits.

Therefore, there is a need for concentrated sources of CLA

that are low in saturated fat.

Jain and Proctor [6] reported a simple method to pro-

duce high levels of CLA-rich soy oil (CLARSO) on a

laboratory scale by converting soy oil LA to CLA using a

UV/visible lamp with 0.15% iodine. However, the photo-

irradiation took 144 h to produce 20% CLA. Jain et al. [7]

further optimized the process to produce CLARSO on a

pilot scale resulting in large quantities of CLA in less time.

R. R. Yettella � C. Castrodale � A. Proctor (&)

Department of Food Science, University of Arkansas,

2650 N. Young Avenue, Fayetteville, AR 72704, USA

e-mail: [email protected]

123

J Am Oil Chem Soc

DOI 10.1007/s11746-011-1962-1

Approximately 75% of total CLAs were trans,trans-iso-

mers, while the remaining were cis,trans and trans,cis-

isomers. In a recent animal study in which obese Zucker

rats were fed 0.5% CLARSO, CLA lowered total and LDL

serum cholesterol and total liver lipids, which are heart

disease risk factors [8]. These benefits were attributed to

trans,trans- isomers, as they were the predominant CLA

isomers in the oil [8]. Because of these health benefits and

the low production cost of CLARSO, the oil could become

an important source of dietary CLA.

Oil rancidity and oxidation are of concern because

CLARSO is a polyunsaturated oil. The oxidative stability

of CLA is a controversial and unresolved issue due to

contradicting reports regarding CLA oxidative stability and

antioxidative capacity. There are several reports that CLA

exhibits antioxidant activity [2, 9–11]. Ha et al. [11]

reported that CLA was more effective in preventing LA

oxidation than a-tocopherol, and almost as effective as

BHT. However, others have found that CLA acts as a

prooxidant [12] or found that anti- and prooxidation vary

between forms of CLA [12]. van den Berg et al. [12]

reported that CLA was a prooxidant when a mixture of

CLA and LA was exposed to air. Chen et al. [13] added

various concentrations of CLA in the methyl ester, free

fatty acid, and triacylglyceride (TAG) forms to canola oil,

and they deduced that CLA acted as a prooxidant in the

methyl ester and free fatty acid forms, but not in the TAG

form. Most researchers agree that CLA oxidizes faster than

LA [11–14], and there is evidence that CLA has a greater

oxidation rate than linolenic and arachidonic acids [14].

Yang et al. [15] investigated the oxidation of different CLA

isomers by comparing groups of four individual cis,cis;

cis,trans, and trans,trans-isomers. They concluded that

trans,trans-isomers were most stable, and that the oxidative

stability of CLA was more affected by the geometric

configuration of the double bond than the position [15].

Jain et al. [7] investigated the effect of iodine as a catalyst

during photoirradiation and the degree of refining on CLA

rich soy oil oxidation. They reported that photoirradiated

soy oil rich in CLA had lower induction time relative to the

control possibly due to the pro-oxidant effect of iodine.

The removal of iodine after processing is important and its

role in affecting oil quality is explored in this paper. Fur-

thermore, CLA oxidation is controversial as reports of

oxidative stability vary. This is probably due to differences

in experimental conditions including CLA isomers inves-

tigated, oxidation storage conditions, and methods of oxi-

dation analysis [16]. Researchers also disagree regarding

the oxidation products of CLA, and on whether CLA oxi-

dizes by the same mechanism as linoleic acid. The most

recent investigation of oxidation products of CLA was by

Garcıa-Martınez et al. [17]. They determined the volatiles

of Tonalin�, a commercial CLA source derived from

safflower oil, and they found heptanal and trans-2-nonenal

to be marker compounds of CLA oxidation [17].

The oxidative stability of CLARSO has not been

reported. Therefore, a study of CLARSO oxidative stability

was conducted. The specific objectives of this research

were to (1) determine the oxidative stability of CLARSO

(2) determine the effects of CLA concentration on the

oxidative stability of CLARSO, and (3) determine the

volatile compounds formed during CLARSO oxidation.

Materials and Methods

Materials

Commercial refined, bleached, and deodorized (RBD) soy

oil was obtained from Riceland Foods (Stuttgart, AR)

containing 52% linoleic and 6% linolenic acid, and it did

not contain any added antioxidants. Resublimed iodine

crystals were used as a catalyst (Alfa Aeser, Ward Hill,

MA). Heptadecanoic acid methyl ester (17:0; Sigma-

Aldrich, St. Louis, MO) was used as a standard for

GC–FID analysis.

Methods

Soy Oil Adsorption Pretreatment

Soy oil in a 1-L beaker was mixed with 5% of Magnesol�

(The Dallas Group of America Inc., Whitehouse, NJ) and

stirred with a magnetic stirrer (Model PC-620, Corning

Inc., Lowell, MA) for 20 min on speed setting 8, adapting

the method of Tokle et al. [18]. The oil was then vacuum

filtered and deaerated with a sonicator for 30 min and

placed in a 1-L beaker wrapped with aluminum foil to

prevent exposure of oil to light.

CLA-Rich Soy Oil (CLARSO) Production

by Photoirradiation

Soy oil was photoirradiated with a 0.35% iodine catalyst in

a pilot scale system [7]. The system consisted of illumi-

nated laminar flow units, which allowed maximum expo-

sure of the oil for linoleic acid photoisomerization. The

pilot plant unit contained a 10-L capacity reaction vessel

with a temperature coil and a mechanical stirrer to hold

degassed oil for irradiation. Two borosilicate glass plates

(45 9 45 9 1 cm) were fixed in 0.5 cm Teflon coated

grooves for a 1-L volume of oil within a 303 stainless steel

frame to allow laminar flow. A third plate permanently

fixed at a distance of 0.5 cm from the first plate served as a

water jacket to provide cooling of oil during irradiation. A

thermocouple probe was in the steel frame to monitor the

J Am Oil Chem Soc

123

temperature of the oil, which was 48 �C. Three 450 W UV/

visible lamps (Ace Glass Inc., Vineland, NJ) were on the

water-jacketed side of the laminar flow unit to ensure

maximum light exposure. The laminar flow unit and UV/

visible lamps were in a sealed irradiation chamber.

Deaerated soy oil with iodine was pumped into the illu-

minated laminar flow unit and was irradiated for 6 h to

obtain 8% CLARSO, and 12 h to obtain 15% CLA-rich soy

oil.

GC–FID CLA Analysis

Methyl esters were prepared from RBD soy oil and

photoisomerized soy oil by a base-catalyzed method to

reduce conjugated trans,trans-isomers formation during

analysis [19]. One hundred milligrams of photoisomerized

soy oil was weighed into a 25-mL centrifuge tube and

500 lL of 1% heptadecanoic acid methyl ester (17:0,

internal standard), 2 mL of toluene, and 4 mL of 0.5 M

sodium methoxide in methanol were added to the centri-

fuge tube and then purged with nitrogen gas. The centri-

fuge tube was heated to 50 �C for 10–12 min and then

cooled for 5 min. To inhibit formation of sodium hydrox-

ide, which could hydrolyze methyl esters to free fatty acids,

200 lL of glacial acetic acid was added to the centrifuge

tube. Five milliliters of distilled water was added to the

centrifuge tube followed by 5 mL of hexane, and the tube

was vortexed (Model VM-3000, VWR, Thorofare, NJ) for

2 min. The hexane layer was extracted and dried over

anhydrous sodium sulfate in a 7-mL glass vial. Another

5 mL of hexane was added to the centrifuge tube, the tube

was vortexed for another 2 min, and the hexane layer was

dried over anhydrous sodium sulfate prior to methyl ester

analysis.

Methyl esters were analyzed by gas chromatography

(GC) using an SP 2560 fused silica capillary column

(100 m 9 0.25 mm i.d. 9 0.2 lm film thickness; Supelco

Inc., Bellefonte, PA) with a flame ionization detector (FID)

(Model 3800,Varian,Walton Creek, CA). Duplicate 2-lL

samples prepared in hexane were injected by an autosam-

pler CP8400 (Varian) and gas chromatograms were col-

lected by Galaxie Chromatography Workstation 1.9.3.2

(Varian). Commercial CLA methyl ester, methyl linoleate,

and mixed methyl fatty esters (Sigma-Aldrich, St. Louis,

MO) were used as standards. Two determinations each

consisting of duplicate injections were conducted for each

treatment. CLA concentrations were calculated by the

following equation:

Lipid Oxidation Studies

Experimental Design

Three treatments viz. RBD soy oil treated with 0.35%

iodine; 8% CLARSO; 15% CLARSO and control RBD soy

oil were investigated. The CLA concentration in treatments

was confirmed by GC–FID. The initial peroxide values and

p-anisidine values of the oils were determined, and then 7 g

of oil was weighed into 10-mL amber colored glass vials.

Vials were stored in an oven, in the dark at 50 �C for

10 days. Triplicate vials for each treatment were sampled

daily, and the oxidative stability was investigated by

headspace oxygen measurement, peroxide value, and

p-anisidine value.

Gravimetric Analysis of Lipid Oxidation

Five hundred milligrams of the oils was weighed into

aluminum pans (I.d. 9 D, 5.7 9 1.6 cm; VWR Interna-

tional Inc., West Chester, PA) in triplicate, and initial

weights were recorded. Pans were stored in oven at 50 �C

for 10 days. Samples were weighed daily, and results were

expressed as change in weight (mg per 500 mg of oil).

Headspace Oxygen Analysis

Oxygen Analyzer-Quantek Model 905 (Quantek Instru-

ments, Grafton, MA) was used to measure the headspace

oxygen content in each vial. The probe needle was inserted

through the rubber septum, and a stable headspace oxygen

reading was reported.

Peroxide Value (PV)

PV was obtained using a small-scale adaptation to AOCS

Official Method Cd 8-5 [20].

p-Anisidine Value (PA)

The p-anisidine value was determined using the AOCS

Official Method Cd 18-90 [21].

Isomer conc: ¼ internal standard conc: ð5 mgÞ � peak area � relative response factor

internal standard peak area

� �

J Am Oil Chem Soc

123

Determination of Major Volatiles Produced During

CLARSO Oxidation

Four treatments of RBD soy oil were investigated: a con-

trol RBD soy oil, RBD soy oil with added 0.35% iodine,

8% CLA-rich soy oil (8% CLARSO), and 15% CLA-rich

soy (15% CLARSO). Two milliliter of oil was pipetted into

10-mL glass amber vials, which were crimp capped with

rubber septa. Vials were stored horizontally in an oven, in

the dark at 75 �C for 5 days. Headspace volatiles were

analyzed daily by gas chromatography–mass spectrometry

(GC–MS) in duplicate.

GC–MS Analysis of Volatile Oxidation Products

An internal standard of 1 ppm 1, 2- dichlorobenzene was

added to each vial prior to oven storage. Headspace

volatiles of duplicate samples were extracted by solid

phase microextraction (SPME) using a carboxen/poly-

dimethylsiloxane (CAR/PDMS) fiber (Supelco Inc., Bel-

lafonte, PA). Vial septa were pierced with the SPME

needle and volatiles were absorbed onto the CAR/PDMS

fiber at 60 �C for 1 min with agitation. Volatiles were

analyzed using a Varian STAR 3400 CX GC (Varian,

Walnut Creek, CA), and volatiles were desorbed into the

injection port at 220 �C for 1 min. The sample was

injected in splitless mode and split after 0.75 min. The

GC was equipped with a Varian 8200 CX autosampler

(Varian) a Varian Autotherm Heater (Varian), and a

Factor Four VF-5 ms capillary column (30 m 9 0.25 mm

I.d. 0.25 lm film thickness, Varian). The initial GC oven

temperature of 42 �C was held for 1 min, and then the

temperature was increased from 42 to 250 �C at a rate of

10 �C/min. A Varian (Saturn 2000) mass selective

detector with a scan range of 40–350 mg at 1 scan/s was

used to identify the volatile components. The mass

selective interface and ionization source temperature was

250 �C, the manifold temperature was 80 �C, and the

transfer-line temperature was 200 �C. All mass spectra

were obtained at 70 eV. The compounds were identified

by gas chromatographic retention times of the standard

compounds.

Statistical Analysis

All parameters were determined in triplicate for each

sample. The data was analyzed by analysis of variance

(ANOVA) and mean comparisons were done by a Stu-

dent’s t test (p \ 0.05) using JMP 9.0 (SAS Institute

Inc., Cary, NC). In addition, the gravimetric data was

analyzed by linear and quadratic contrasts between

concentrations of CLA using JMP 9.0 (SAS Institute

Inc., Cary, NC).

Results and Discussion

CLA Concentration of CLARSO by GC–FID Analysis

The fatty acid composition of CLARSO irradiated for 12 h

is presented in Table 1. The oil irradiated for 12 h con-

tained 15.2% ± 0.2 CLA and the oil irradiated 6 h con-

tained 8.15% ± 0 CLA.

Gravimetric Analysis of Lipid Oxidation

Figure 1 shows the effect of CLA concentration on weight

change of CLARSO samples stored at 50 �C for 10 days. The

control (RBD soy oil) did not reach induction time during the

incubation period and 0% CLA (non-irradiated RBD oil with

iodine) had a longer induction time (9 days) relative to the

CLA samples. This suggests that products of photoirradiation

are responsible for oxidation rather than the addition of

iodine alone. This could be explained by the presence of

iodine radicals produced during irradiation. Oxygen levels

are low during irradiation [7] and oxidation is very limited

during the photoisomerization process [6]. However, the

iodine radicals formed as result of UV light exposure may be

responsible for oxidation upon exposure to air if not removed

after processing. In addition, the 15% CLA did have a shorter

induction time (5 days) than the 8% CLA oil (7 days) and a

more rapid subsequent weight gain despite having the same

amount of iodine content. This suggests that increasing CLA

levels may increase oil oxidation during storage. Figure 2

shows the effect of CLARSO CLA concentration on daily

weight change of the oil. This CLA effect is a statistically

significant linear effect on the shaded days, which means that

on day 9 and 10 as CLA concentration increased from 0 to

15%, sample weight increased suggesting that oxidation also

increased. There are no significant differences between

samples until day 9 and 10. This is on or after induction time

of all three samples, when weight increase occurs at different

rates (Fig. 2).

Headspace Oxygen Analysis

Figure 3 shows the headspace oxygen depletion of the four

treatments of oil stored at 50 �C for 10 days. With the

Table 1 Fatty acid composition

of CLA rich soy oil

Values are the average of trip-

licate samples

CLA Conjugated linoleic acid

Fatty acid Percentage

C16:0 11.5

C17:0 4.5

C18:0 4.2

C18:1 23.7

C18:2 35.0

C18:3 5.9

CLA 15.2

J Am Oil Chem Soc

123

exception of the control, headspace oxygen depletion

occurred rapidly over the first 3 days in all samples. The

0% CLA oil, which contains iodine but was not irradiated,

showed the greatest oxygen depletion and fast rate of

oxidation. Oxygen depletion in the CLA oils was not quite

as rapid over the first 3 days as in 0% CLA samples and

was independent of CLA concentration. Although CLA

had no significant effect on oxygen depletion, all oils

containing the 0.35% iodine, had more rapid oxygen

depletion than the control oil. Iodine-containing samples

and the control RBD oil lost 8% oxygen in 3 and 10 days,

respectively.

The trends in oxidation measurement by oxygen

depletion are in contrast to those obtained by gravimetric

analysis and this confirms the reports that CLA oxidation

data are highly dependent on the analytical methods used

and oxidation parameters measured [16]. The gravimetric

study show a CLA concentration effect which is not seen in

the headspace oxygen determination but rather indicates

the presence of iodine as the main factor affecting

Fig. 1 Effect of conjugated

linoleic acid (CLA) on the

change in weight of conjugated

linoleic acid rich soy oil

(CLARSO) stored at 50 �C for

10 days. Control is RBD soy

oil. Error bars represent

standard deviation (n = 3)

Day

-5

5

15

25

-5

5

15

25

-5

5

15

25

-5

5

15

25

Day 00 Day 01 Day 02



Day 09 Day 10

0% CLA 8% CLA 15% CLA 0% CLA 8% CLA 15% CLA 0% CLA 8% CLA 15% CLA

Fig. 2 Daily conjugated

linoleic acid (CLA) effect on

change in weight of conjugated



10 days. Shaded days indicate a

significant (p \ 0.05) linear

CLA effect

J Am Oil Chem Soc

123

oxidation. The 7-g samples of CLA oil with limited oxygen

in a sealed glass vial oxidized faster (3 days) than 500-mg

samples in an open vessel with unlimited oxygen (7–8 days).

Although both measurements were indirect determinations

of primary oxidation products, these measurements could

not be compared as experimental conditions differed.

Peroxide Value (PV)

Peroxide values are shown in Fig. 4. The control RBD soy oil

sample produced a steady increase in PV that paralleled

oxygen depletion in the same sample (Fig. 3). However, no

increase in PV was found in CLA and iodine-containing

samples. This is supported by a report that hydroperoxides

are not products of CLA oxidation [11]. However, the 0%

CLA oil also did not produce hydroperoxides (Fig. 4) indi-

cating the possible presence of an iodine issue. Therefore,

added iodine may have affected PV determination as it par-

ticipates in the chemistry of iodometric titration during PV

determination. The following is a proposed hypothesis for

interference of iodine in the PV determination. When iodine

is added to soy oil in the production of CLARSO, it is in the

molecular iodine (I2) form. Then, during irradiation, or any

heat or light exposure, it becomes iodide (I-).

The chemical equation for the PV [22] reaction is as

follows:

ROOH þ KþI� ! ROH þ KþOH� þ I2

I2 þ starch þ 2Na2S2O3 ! 2NaI þ starch þ Na2S4O6

Peroxides (ROOH) react with KI and result in the

release of I2. The solution is titrated with a few drops of

sodium thiosulfate (Na2S2O3), and then a starch indicator is

added, which produces a blue color. The titration with

Na2S2O3 continues until the blue color disappears. The

principle of the PV determination is the reduction of I2

to I-:

I2 þ 2S2O2�3 ! S4O2�

6 þ 2I�:

With added I- already present in the oil, less Na2S2O3 is

necessary to reduce I2 to I-, and the PV is therefore lower.

p-Anisidine Value (PA)

Secondary oxidation products were measured as p-anisi-

dine values (Fig. 5). There was no trend and no significant

difference in the data between samples over time, with

values between 0.5 and 7.5. CLA concentration and storage

time did not have any significant effect on PA. The PA

standard for fresh oil was less than 2, and a highly oxidized

soy oil was reported to have a PA of about 50 [23]. The

control RBD soy oil had minimal PA (Fig. 5). This may be

due to lack of peroxides degradation. Furthermore, PA only

measures 2-alkenals and 2,4-dienals, and it is possible that

other secondary oxidation products were present [21].

Identification of Volatile Oxidation Products in RBD

Soy Oils During Storage

Table 2 shows the concentration of major volatile com-

pounds after 5 days of storage and their sensory threshold

values. The following volatiles were identified in order of

prominence in the oils: hexanal, pentanal, heptanal, trans-

2-heptanal, trans-2-octenal, octanal, and trans-2-nonenal,


linoleic acid (CLA) on

headspace oxygen depletion of

conjugated linoleic acid rich soy

oil (CLARSO) stored at 50 �C

for 10 days. Control is RBD soy



J Am Oil Chem Soc

123

which are the products of oleic and linoleic acid oxidation

[24]. Linoleic acid is the most abundant fatty acid in soy

oil, followed by oleic acid. These fatty acids are the most

common fatty acids in the oils.

The major volatiles were pentanal and hexanal, which

are produced from the breakdown of linoleic acid

13-hydroperoxide [24], thus linoleic acid seems to be

oxidizing faster than CLA and other fatty acids. trans-2-

Octenal and trans-2-heptenal are minor linoleic acid

products, which are most likely produced by photooxida-

tion [24] and are probably formed during photoirradiation

or general light exposure since their concentrations were


linoleic acid (CLA) on peroxide

value of conjugated linoleic acid

rich soy oil (CLARSO) stored at

50 �C for 10 days. Control is

RBD soy oil. Error barsrepresent standard deviation

(n = 3)


linoleic acid (CLA) on

p-anisidine value of conjugated



10 days. Control is RBD soy



J Am Oil Chem Soc

123

high in CLARSO samples (p \ 0.05). Octanal is a break-

down autoxidation product of the 11-hydroperoxide of

oleic acid. Heptanal and trans-2-nonenal may have been

formed by photooxidation of oleic acid in CLA oil during

photoisomerization [24].

CLARSO samples had significantly higher (p \ 0.05)

levels of pentanal and trans-2-heptenal than the other oils,

but there was no significant difference between the

amounts of any volatiles present in 8 and 15% CLARSO

samples. This suggests that oxidation is greater in

CLARSO samples relative to RBD soy oil with and with-

out iodine, but is not affected by CLA concentration.

Possibly CLARSO exposure to UV light during processing

played a role in the formation of this compound because

trans-2-heptenal is common in photooxidation [24]. Gar-

cıa-Martınez et al. [17] reported that significantly high

quantities of heptanal and trans-2-nonenal were found in

CLA rich oil. Heptanal was present in all oils containing

CLA. However, the control also had heptanal, but this

was significantly lower than the CLA-containing oils

(p \ 0.05). The 8 and 15% CLARSO samples had trans-2-

nonenal, but the amounts were not significantly different

(p \ 0.05).

Though the volatile products are present in small con-

centrations, they are detectable amounts and therefore

contribute to off flavor and odor development. This is

evident by comparing the volatile concentrations with the

sensory thresholds listed in Table 2. Hexanal and pentanal

concentrations are well above their respective threshold

values of 0.0045 and 0.012 ppm [25]. This indicates that

off flavors and odors are likely present in the oils. Minor

volatile components, such as octanal and trans-2-nonenal

also have very low threshold values of 0.007 [25] and

0.001 [26] ppm, respectively. Even though these products

were not formed in all of the oils, they still contributed to

off flavors when they were present, even in very small

amounts.

Conclusions

This study showed that CLA oxidation results are highly

dependent on analytical methods used and oxidation

parameters measured. The gravimetric study indicated a

CLA concentration effect on oxidation, which is not seen

in the oxygen depletion and peroxide value studies. Vola-

tile oxidation data indicate that CLARSO samples had

significantly higher (p \ 0.05) levels of pentanal and trans-

2-heptenal relative to RBD soy oil with and without iodine.

However, there were no significant differences in other

volatiles in CLARSO samples. This suggests that oxidation

was greater in CLARSO, but is not affected by CLA

concentration. Iodine radicals involved in the catalytic

formation of CLA may act as pro-oxidants. Thus, iodine

should be removed from CLARSO to improve its oxidative

stability.

Acknowledgments We are grateful to the Arkansas Soybean Pro-

motion Board for financial support and Riceland Foods Inc. (AR) for

providing us with the RBD soy oil.

References

1. Lee KN, Kritchevsky D, Pariza MW (1994) Conjugated linoleic

acid and atherosclerosis in rabbits. Atherosclerosis 108:19–25

2. Ip C, Singh M, Thompson HJ, Scimeca JA (1994) Conjugated

linoleic acid suppresses mammary carcinogenesis and prolifera-

tive activity of the mammary gland in the rat. Cancer Res

54:1212–1215

3. Houseknecht KL, Heuvel JPV, Moya-Camarena SY, Portocarrero

CP, Peck LW, Nickel KP, Belury MA (1998) Dietary conjugated

linoleic acid normalizes impaired glucose tolerance in the Zucker

diabetic fatty fa/fa rat. Biochem Biophys Res Commun 244:678–682

4. West DB, Delany JP, Camet PM, Blohm F, Truett AA, Scimeca J

(1998) Effects of conjugated linoleic acid on body fat and energy

metabolism in the mouse. Am J Physiol 275:R667–R672

5. Ha YL, Grimm NK, Pariza MW (1989) Newly recognized anti-

carcinogenic fatty acids: identification and quantification in nat-

ural and processed cheeses. J Agric Food Chem 37:75–81

Table 2 Concentration of volatiles formed in RBD soy oils with added iodine and CLA, and CLARSO, after 5 days of storage at 75�C

Oil Volatiles concentration (ppm)

Hexanal Pentanal Heptanal trans-2-

Heptenal

trans-2-

Octenal

Octanal trans-2-

Nonenal

Control 2.39 ± 1.40a 1.04 ± 0.05a 0.44 ± 0.62a 0.54 ± 0.77a, b 0.51 ± 0.72a 0.48 ± 0.68a ND

0% CLA 2.24 ± 0.27a 1.30 ± 0.13b ND ND 0.98 ± 0.02a, b ND ND

8% CLARSO 2.98 ± 0.22a 1.39 ± 0.06b, c 1.26 ± 0.06b 0.99 ± 0.02b, c 1.33 ± 0.03b 0.97 ± 0.01a 0.89 ± 0.02a

15% CLARSO 2.81 ± 0.39a 1.49 ± 0.09c 1.27 ± 0.22b 1.59 ± 0.02c 1.28 ± 0.07b 1.14 ± 0.21a 0.53 ± 0.76a

Sensory threshold valuesA 0.0045 0.012 0.003 0.013 0.003 0.0007 0.001

Means with different letters within the same column are significantly different at p \ 0.05

Control RBD soy oil, RBD soy oil refined bleached deodorized soy oil, CLA conjugated linoleic acid, CLARSO conjugated linoleic acid rich soy

oil, ND not detectedA Sensory threshold values were obtained from Buttery and Ling [25] and Buttery et al. [26]

J Am Oil Chem Soc

123

6. Jain VP, Proctor A (2006) Photocatalytic production and pro-

cessing of conjugated linoleic acid-rich soy oil. J Agric Food

Chem 54:5590–5596

7. Jain VP, Proctor A, Lall R (2008) Pilot-scale production of

conjugated linoleic acid-rich soy oil by photoirradiation. J Food

Sci 73:E183–E192

8. Gilbert W, Gadang V, Proctor A, Jain VP, DevareddyL (2011)Trans-trans conjugated linoleic acid enriched soybean oil reduces fatty liver

and lowers serum cholesterol in obese Zucker rats. (in review)

9. Nicolosi RJ, Rogers EJ, Kritchevsky D, Scimeca JA, Huth PJ

(1997) Dietary conjugated linoleic acid reduces plasma lipopro-

teins and early aortic atherosclerosis in hypercholesterolemic

hamsters. Artery 22:266–277

10. Ip C, Chin SF, Scimeca JA, Pariza MW (1991) Mammary cancer

prevention by conjugated dienoic derivative of linoleic acid.

Cancer Res 51:6118–6124

11. Ha YL, Storkson J, Pariza MW (1990) Inhibition of benzo(a)-

pyrene-induced mouse forestomach neoplasia by conjugated

dienoic derivatives of linoleic acid. Cancer Res 50:1097–1101

12. van den Berg JJM, Cook N, Tribble D (1995) Reinvestigation of

the antioxidant properties of conjugated linoleic acid. Lipids

30:599–605

13. Chen Z, Chan P, Kwan K, Zhang A (1997) Reassessment of the

antioxidant activity of conjugated linoleic acids. J Am Oil Chem

Soc 74:749–753

14. Zhang A, Chen Z (1997) Oxidative stability of conjugated lino-

leic acids relative to other polyunsaturated fatty acids. J Am Oil

Chem Soc 74:1611–1613

15. Yang L, Leung LK, Huang Y, Chen Z (2000) Oxidative stability of

conjugated linoleic acid isomers. J Agric Food Chem 48:3072–3076

16. Luna P, de la Fuente M, Salvador D, Marquez-Ruiz G (2007)

Differences in oxidation kinetics between conjugated and non-

conjugated methyl linoleate. Lipids 42:1085–1092

17. Garcıa-Martınez MC, Marquez-Ruiz G, Fontecha J, Gordon MH

(2009) Volatile oxidation compounds in conjugated linoleic acid-

rich oil. Food Chem 113:926–931

18. Tokle T, Jain VP, Proctor A (2009) Effect of minor constituents

on soy oil conjugated linoleic acid production. J Agric Food

Chem 57:8989–8997

19. Ma DWL, Wierzbicki AA, Field CJ, Clandinin MT (1999)

Conjugated linoleic acid in Canadian dairy and beef products.

J Agric Food Chem 47:1956–1960

20. Crowe T, White P (2001) Adaptation of the AOCS Official

method for measuring hydroperoxides from small-scale oil

samples. J Am Oil Chem Soc 78:1267–1269

21. American Oil Chemists’ Society Official Method Cd 18-90

(2004) Official methods and recommended practices of the

AOCS, 5th edn. AOCS, Champaign

22. Pike OA (2003) Fat characterization. In: Nielsen SS (ed) Food

analysis, 3rd edn. Springer Science, New York, pp 227–246

23. Anwar F, Shahid Chatha SA, Ijaz Hussain A (2007) Assessment

of oxidative deterioration of soybean oil at ambient and sunlight

storage. Grasas y Aceites 58:390–395

24. Frankel EN (1998) Hydroperoxide decomposition. In: Frankel

EN (ed) Lipid oxidation, 1st edn. The Oily Press Ltd, Dundee,

pp 54–77

25. Buttery RG, Ling LC (1995) Volatile flavor components of corn

tortillas and related products. J Agric Food Chem 43:1878–1882

26. Buttery RG, Orts WJ, Takeoka GR, Nam Y (1999) Volatile flavor

components of rice cakes. J Agric Food Chem 4:4353–4356

J Am Oil Chem Soc

123

Documents

Oxidative Stability of Conjugated Linoleic Acid Rich Soy Oil