WO2014124162A1 - Conjugated linoleic acid rich vegetable oil production - Google Patents

Abstract

The invention is generally directed to conjugated linoleic acid (CLA)-rich vegetable oil produced from linoleic rich oils by heterogeneous catalysis. The process includes heterogeneous catalytic vacuum distillation under high temperature conditions to isomerize linoleic acid in triacylglyceride vegetable oils to CLA to produce CLA-rich oils. After processing, the catalyst may be removed to obtain high quality, CLA-rich oils. The CLA-rich oils may then serve as a potent and bioactive nutraceutical and can be incorporated into various food products, such as a CLA-rich dressing, shortening, margarine, chocolate or chips.

Description

[0001] This application claims priority to and the benefit of U.S. Provisional Patent Application No..61/762,678, filed Februar g, 2013, which is incorporated herein by reference In Its entirety.

[0002! Not Applicable,

8ACKGROU @„ F THE INVENTION

i- Mot iYs ffi

[0003] Ibis .invention relates generally to conj gated nolei.c acid (CLA)-nch vegetable oil production Jrom Imaleic rich, oils by heterogeneous^' catalysis, and In particular to a proc ss far producing CLA-rich oil b isoroerizmg llnokie- acid in Ixiacy!glyceride vegetable oils to CLA by lo pressure high temperature catalysis.

2. OggsdglQi he ek Ar .

[0004] CLA is a. group of positional and geometric isomers of oci¾dec¾dienoic^: acid with conjugated double bonds. CLA has anti-carcinogenic, anti-atherogenk_> anti-diabetic and anti- obesity properties, along with the ability to increase lean bod mass and to protect against: immune induced body wasting disease, chronic inflammatory disease, cancer and to provide •other positive .health effects.

[0005_.) CLA is .found naturally in dairy and beef products at levels of approximately 0,3- 0,8% (w/w) of the fat as bovine rumen ier entation products. The current human intake of CLA is, however, approximately ten (1 ) times less than the 3g/dsy minnmnn value resce.mmeii4ed^' as being necessary to produce desirable physiological health effects. Obtaining the estimated optimum dietary CLA levels &om n tural beef and dairy sou ces woaid increase the total fat and saturated fat intake and increase the negative health risks associated ith dietary animal tats. Therefore, a concentrate source of dietary CLA that is lo in saturated tat and cholesterol is desirable.

[0006] Soy oil is the most commonly used vegetable oil In United States, and J contains about 50% Hnoleic acid_* Other vegetable oils high in Hnoleic acid include sunflower (57%), com (55%)₅ cottonseed (50%) and peanut (50%). CLA fatty acid has been produced historically by fermentation and e z me technology. CLA in vegetable oil has also been produced by converting linoleie acid to CLA using iodine by homogeneous photo-catalysis. A drawback of this process is the removal of iodine in order for the resultin CLA-rieh oil to b suit le for human consumption.

[OO07J It is^' therefore desirable to provide CLA-rtoh vegetable oil produced from linoleie rich oils using heterogeneous catalysis,

[0008] it is further desirable to provide a process for producing CLA-rieh oil by iso erization of Hnoleic acid in triaeylglyceride vegetable oils to CLA using low pressure/high temperature catalysis.

[0009] It is still forfher desirable to provide a process for producing a 20% CLA-rich oil that: requires only post-processing cataly st removal.

[00.10] .ft is yet further desirable to provide a process tor. apidly producing CLA-rieh oils from Hnoleic rich oils using heterogeneous catalysis that may utilise a continuous: fixed bed: reactor and/or a continuous stirred tank reactor for a cost effective and energy efficient process, [0011] It is yet further desirable to provide a process for producing CLA-rieh oils from, iinoleie rich oils using. he!erogeaeous catalysis that is an environraesitdly-friendly process m contrast to alkali isonierization.

^'[0012] It is still further desirable to provide- a process for producing CLA-rieh oils fern !inoleic rich oils using heteroge&eous^' catalysis thai does not require an solven or any efeernlcal. other than catalyst.

[0013] It is yet further desirable to provide a process for producing. CLA-rich oils fro Imo!eic rich oils using heterogeneous catalysis in the absenc of iodine a»d with a metal catalyst that can be easily removed^' and reused.

[0014] It is yet farthe desirable to provide a process for producing CLA-rieh oils ftom linoleic rich oils using heterogeneous catalysis that utilizes continuous steam Injection to remove any rancidity vo les n linoleic acid-rich oil.

m M x im mi

[0015] In general, the invention relates to a process for producing conjugated ixt leic* aeid-rich oil The process includes catalyzing an. oil-catalyst mixture of iiaoleie acid-rich oil and a catalytic amount of a transition metal, having a vacant. 3d-orbItal or a. vacant d-or iaI to produc the conjugated Hnoieie acid-rich oil. The process can also include the ste of extracting the metal from, the conjugated linokic acid-rich oil such as via filtration or eerm rogation. Th Unolele acid-rich oil can be- a iriacylglyoeride vegetable oil, such as soy, unflower, com, .cottonseed or peanut oil

[0016] The catalysis of the process can. further include catalyzing the oil-eaialyst mixture in low pressure conditions and high temperature conditions to produce the conjugated ! o!eic acid-rich oil. The oi!-eaiaiys .mixture can he catalyzed for up to approximately 240 minutes to produce the conjugated Ifnoleic acid-rich oil Further, the catalysis can be processed between approximately 1 mm Hg and approximately 2 mm fig at temperature conditions- between approximately 165 ⁶C and approximately 2§2 Furthermore,- the transition metal can he ruthenium, rhodium, molybdenum, palladium, gold, copper, iron, manganese, silver or nickel having a concentration between 0,08% to approximately 8%, such as a ruthenium catalyst or a nickel catalyst,

fOOlTj In addition, the Invention relates to conjugated lino!eie aeid-r ch oil produced by the process described herein. Moreover_* the invention relates to an. enriched food, product p epared using the conjugated liuoleic acid-rich oil from the process described herein. BRIEF DESCRTOO ¾£THE DRA INGS_,

[0018] Figure 1 is a schematic diagram of an. example of a deodonisailoa process in - aspect of the CLA-r!e vegetable oil production, from linoleic rich oils by heterogeneous catalysis disclosed herein;

[0019] Figure 2 is a contour plot predicted by a statistical model showing various processing conditions obtained for optimism CLA yields by IsomerixailoB of soy oil linoleic acid to CLA by rutheniutn catalysis, where: the contour lines are a function of time and. emperaftite along which the amoti n has a constant value of CLA yields (18, 18.5, 19, 19,5 and 20%); the unshaded area, shows the- combination of conditions that can be used to obtain. CLA yields of > 8.5%; and the d a was obtained from a high temperature, long processing time study using a central composite ra atable experimental design to o timise heterogeneou ruthenium catalysis of 250 nil soy oil linolelc acid to CLA with 0.64% ruthenium catalyst over range of temperatures (197-282 ^&C) and times (108492 mm);

[0020] Figure 3 is a graphically Hesitation of the effect of nrtheaium on the CLA yield at 270 °C for 3 hoars with 4% formic acid m oil and steam of 85 % ibrmie- acid;

[0021] Figure 4 is a comparative, graphical illustration of force versus compressive extension. Ibr soy oil margarine, CLA-rich soy oil margarine and commercial margarine, where five repetitions were made for each sample and the standard deviation was within. 5 % of mea load v&!nes In all eases;

[0022] Figure 5 is a graphical illustration of complex modulus (O*) measured as a function of oscillatory stress for control soy oil margarine, CIA-rich soy oil margarine and commercial margarine, where the test was performed in triplicates and error bars represent standard deviation; [00231 Fiaure 6 is a granhica!. slot of chaise m phase an¾le with apnlied .oscillatory stress measured for control soy oil margarine, CLA-^rich soy oil margarine and cornrnercial margarine, where the test was -performed in triplicates and error bars represent standard deviation;

10024} Figure ? Is DSC thermograms- showing the melting profiles of control soy oil margarine CLA-rich soy oil margarine and commercial margarine;

[0025} Figure- 8 is a graphical illustration of the solid fat content measured -as a function of temperature for control soy oil margarine, CLA-rich soy oil margarine and -eo merciai margariiie .measured, in triplicate and error bars represent standard deviation;

[0026] Figure 9 are control soy oil margarine microscopic images observed under transmitted norma! (9a 1.) and polarized (9a2) light CLA-rich soy oil, margarine microscopic- images observed under transmitted normal (9hl) and polarized (9b2) light, and commercial margarine microscopic images observed in transmitted norma! (el) and polarized (c2) light; -and

|0027] Figure 1.0 is -a graphical illustration, oi puncture force tests (N) of CLA-rich soli oil shortening samples,

[0028^·] Other advantages and features- of the invention will, be apparent" from, t e following description and from the claims_*

DjTA!LEP DESCRIPTION OF THE INVENTj^j

£^'O029J the processes and compositions^' discussed^' herein re me ely Illustrative of specific manners in wh c to make and use this Invention and are not to be interpreted as limiting in scope,

|:003 ] While the processes ani compositions have been described with a certain degree of particularity, it is to be noted tha many^' variations and modifications may be made: m -the details of the sequence, components, concentrations and the ...arrangement of the: processes and compositions without departing ft m the scope of this disclosure. It is understood -that the invention is not limited to the embodiments se forth herein for purposes of exemplification.

[00 I J The in ention is generally directed to a heterogeneous catalytic vacuum distillation process that utilizes high temperature conditions to Isomerize linofele acid in tfiaeylgiyceride vegetable oils to CLA in order to produce CLA-rica- oils. After processing, the catalyst may be removed, by filtration or centniugatlon to obtain Mgh quality, OLA*rieh oils. The heterogeneous catalysis utilizes- a metal catalyst tha can be selected from my suitable transition metal, such as ruthenium, rhodium, silver or nickel. The process is a two-phase system (oil/catalyst) using conditions with km pressure, between · approximately 1 to 2 mm of Eg pressure, high temperature, above about 2(H) ^¾C, and a continuous flow, suc as at about 0,4 l mm steam flow, to produce a 20% CIA-rich, oil in. less than 2- hours. The CLA-fich. oils may then serve a a potent and hioactive nutracenticai In addition, the CLA-rich oils can be incorporated Into various food products, such as a CLA-ricb dressing, shortening_* margarine, chocolate or chips. EXAMPLES

[0032] The CLA-tkli vegetable oil production from linoieie rich oils by heterogeneous- catalysis disclosed herein is f sr illustrated by the following examples, which are provided for the -purpose, of demonstration rather than limitation. Although soy oil was used in the following examples, any linoieie acid-rich oil cars be used.

EXAMPLE 1

[0033] Studies were .conducted . duplicate to screen nickel, ruthenium, silver and rhodium catalysts fo isomerixaflcm of linoieie acid to GLA,- Rhodium on carbon with 5% loading (Sigma-Aldr ch Product number: 20 164), ruthenium (Sigma-AIdrieh Product number: 206180), nickel on silica/alumina . -65 wt % loading (Sigma-Aldrich Product number: 208779). Silver powder instead of silver coated on carbon was used because silver is not commercially available coated oa carbon or other support

£0034] Catalyst concentrations and processing conditions are in Table 1 beiow> Duplicate samples of 250 mL of fully refined oil were processed for 90 minutes at their optimum temperature of each catalyst. Control experiments were also conducted using iodine (0,35%) and tocopherols (0,14%) catalysts dissolved In oil.

Γ¾ ε 1: C mmerckl metsl catalysts sc eening for CLA-ricJi oil production relative to >r»dsieti<ni with iodise and mixed t€Of¾er©ls«

[0035] Heterogeneous catalyzed oils were then filtered and oil tatty acid composition determined by OC fatty acid methyl, ester analysis in. duplicate. Oil quality as determined by measuring peroxide value and free fatty acid valve in duplicate.

(0036] As the data in Table 1 illustrates, ruthenium and rhodium produced higher but statistically similar CLA levels a a concentration of 0.32% than other metal, catalysis. Since rhodium is 10 times more expensive than ruthenium, ruthenium was selected as the most viable metal for CLA production by heterogeneous catalytic deodor&ation process. Ruthenium was further o iimked as an economically viable conjugation catalyst, and preliminary screening showed that, increasing the ruthenium catalyst concentratio increases CLA yields. Around 15% CLA was produced at a catalyst level of 0.64% at 165 "C over 90 mim [0037] Transition metals contain ires and vacant d-orhltals, hicli cm. interact with, pi- bonds of lij oleic acid, and which are- capable of activating a nearby C~M bond leading to bond migration. Such interaction destabilizes the native structure allowing the more stable CLA eonjugaied-system to form a«d be released from the catalyst.

EXAMPLE 2

[0038] To o timize conditions by a central composite rotatable experimental design using a Ruthemtsn catalyst (R« loaded on earbon) was used with 5 levels of 3 variables (catalyst concentration, time, and temperature) (Table 2), The results of Example 1 were used to select the center point ibr experimental design to determine the effect of variables on CLA yields above and below the- center point.

Table 2: Centre! Com osite Rotatable Besign. of Exaai le 2 for Optmi&stbtt ©f CLA Pro h*et-e*t Ove a r s mg* of esiperatares { 0-2 0 ¾}, Times (40-141 mis) mil R« Catalyst€¼»ceairatioss (0,21-0.75%),

[0039} The ruthenium catalyst Is o» a carbon support consisting; of a 1 microns particle with total surface area surface area of 900 per gram. The ruthenium catalyst has surface area of J 3 m~ per gram and moistur content of <S% (Sigma- Aidrich Product number: 20. 180).

[0040] Two hundred arid fifty milliliter samples were processed in duplicate- -for each: set of processing conditions. Oils were then .filtered and oil fatty acid comp s ti n determined b gas chromatography ike fatty acid (FFA) methyl ester analysis duplicate. Oil quality was determined b measuring the peroxide value. (PV) and free tatty add value in. duplicate.

[0041] As the data is Table 2 illustrates, 0,64% ruthenium at 210 ^*C over 120 minutes produced 20% CLA-rich oil, which were the conditions that produced most CLA, Yields increased with temperature, time and catalyst concentration but time and temperature had the g eatest effect on CLA yields,

EXAMPLE 3

[0042] To further o timi e the time and temperature conditions for a catalyst concentration fixed at 0,64% (ruthenium on ca bon), a central composite ro atahie design was utilized with 5 levels of 2 variables (time and temperatare) (Table 3). The results of Example were used to select the higher temperature- and time settings for the experimental design to determine the effect of variables on CLA yields for time arid temperature conditions over the ones used in Example 2. The objective of Example 3 was based on the hypothesis predicted by experimental design of Example 2 that higher temperatures and times Would give .greater CLA yields at a fixed catalyst concentration (0,64%), as farther loading of the catalyst coated on. carbon increased, the viscosity of oil which would increase process cost, and cause filtration problems. Table 3; Cent al omposite Rotatafele Design for Ex m le 3 for Optiiakatloa€LA~Ek¾ Oil Predtaciios O er a High Temperature (197-282 ¾) Range and Logger mim (W -iSl^' mm) ih m Example 2_» with &6 % Ru catalyst

[0043] As the data In Table 3 illustrates, CLA yields at higher temperatures- md longer times with 0.64% mtheniurn catalyst resulted is less CLA produced relative to Example 2. A maximum of 21.05% CLA can be obtained, at 240 ^*C temperature for \ 8 mm.

[0038] An additional study was conducted, to optimize conditions of a nickel .catalyst for isomerixatieti of Imoleic acid to CLA. using a central composite rotata le design .similar to the preceding examples, The nickel catalyst was used on a silica/alumina (powder) with total surface area surface area, of 190 m² per gram (Sigma Aldrich Product number: 208779), The nickel catalyst concentration was fixed at S% with the lime and temperature processin conditions listed ^'below Table 4, Table 4^; CestraS Composite Rotatable Besiga of Example 4 for Q il nkafls ef CLA ^•Prodteetiea Over a Tea ratare Ra«ge (180-248 °C) md Times (188-24® ia) with B% T¾ catalys

[0039] As the data s- Table 4 ilinstraies, 8% ukkel at 240 ¾ over 180 minutes produced approximately 6.70% CLA-rich oil, which were the conditions that produced most CLA. Yields increased with temperature, time and catalyst eonoentsmiors but time and temperature had the greatest effect on CLA yields, While t e about 6,7% CLA yields from Example 4 using a nickel catalyst where lower t an the around .21% CLA yields from E am le 3 thai utilized a ruthenium catalyst, nickel is less expensive than, ruthenium. Therefore nickel is also an economically viable catalyst with a high commercial potential for CLA production by a heterogeneous catalytic- deodorkation proce s,

[0044] In addition to. the traasition metals of ruthenium, rhodium, silver and nickel utilized in the foregoing non-limiting examples, the processes and systems to produce CLA-rich oils disclosed herein can be used, wit other inexpensive (comparative- to nickel) transitional metal catalysts, such as copper, iron, ami manganese having vacant 3d-orhria!s< Moreover, more expensive transition metals (comparative to ruthenium), such as molybdenum, palladium and gold having vacant 4d-orb s and larger atoms than nickel, can similarly be used as catalysts in the processes and systems to produce CLA-rich oils disclosed herein. EXAMPLE 4

[0045] Example optimizes the operating parameters to maximize CLA yields^■■of CLA- rich oil ruthenium heterogeneous catalysis without the us of .sol e ts or .hydrogen sod also determines fatty acid distribution and CLA isomer composition of CLA-rich oil produced, ixier optimised conditions. Processing conditions were optimized to maximize CLA yields usin central composite experimental design studies with variable: time, temperature, and catalyst concentration. In Example: 4,- the inventive process produced. 21%€LA-rich oil in 108, mln at 240 ΐ by a single heterogeneous catalysis step using: a nitheniu (0.64%) under low pressure (1-2 mm Hg) in presence of steam (0,4 rnL rnin steam Sow).

[0046] Refined, bleached, and. deodorized (RBD) so oil was obtained f nt Rieciand Foods (Stuttgart, AR, USA) and used, throughout this Example 4, Raihemum eaiab si (5 loading on carbon) was purchased from Sigma- Atdrich. Chemie-als, inc-, (St. ou _. MO, USA), The lab scale- deodorization unit was modeled by O'Brien scientific glassblowing (Montkello, II, USA) in accordance with the design adapted iroxn Rieeland Foods (Stuttgart, All, USA}* HPLC solvents hexaae and aeetontele were HPLC grade (VWR. International, West Chester, PA, USA), Other chemicals sed were reageni grade.

P∑047] Miml Study wtik Variable Temperatures_* p o essing Times and C iafyst Co ce t tion.- A central composite totatab!e experimental design was employed t ti ise conditions to maximize CLA yields, ^'the design had 5 levels of 3 variables:, catalyst concentration (0.21, 0>32, . 8, 0.64 and 0,75%), time (40;, 60, 90, 120 and. 140 mis) an temperature (9 , 120, 165_:, 210 and 240 ^*€) in duplicate as shown in Table 5 below, Two hundred and fifty (250) ml, soy oil samples were processed in duplicate for each set of processing conditions in a. random order as described below. Ta¾Ie 5: Initial study (Exmn te 4) tisiag Central Com osite Eotatable expedm«st*t esig® to o imis hete egtsBteOHs rui¾ealai» catalysis of 2$0»ssL soy oS! Ik« e acfcl to CL over a raage of temperatures (90-240 ¾ , imes (40-140 mia> and catalyst *«Bce»Sratieas .(0,21- 0.75% .

[0048] The lab scale deodorixation uni was used to produce CL-A-nch oil by heterogeneous catalysis and is shown in Figure i. The assembly was composed of glass, units with standard ground glass spherical joints that could be lubricated with vacuum grease to maintain a vacuum of 1-2 mm. f ig. Steam was generated at a rate of 0.4 mL/min by a 2S0-wait infrared bulb that heated the water in the water reservoir. The 250-rnL round bottom flask placed in the heating mantle contained the mixture oi oil and catalyst to be processed. A iherxnorneter was Inserted through ground glass joint into the body of the glass to continuously monitor temperature and the condensers were cooled with the dry ice-acetone mixture. All the vapors were retained by the first condenser and the second condenser acted as a safety device in case of fa lur of the first condenses The outlet of the condenser was attached to the vacuum pom to a t in a near vacuum of I to 2 mm- Hg pressure throughout the assembly.

[0049] After the oD-eaiaiyst mixt re was add d to the flask, he dr ice and acetone were added to the condensers (sec Figure- J) and water was laced in the water reservoir- flask. The- heating mtle apd the vacuum pump were then turned cm. As the oil tem erature reached 60 ¾ the infrared heating lamp was turned on and the pressure valve at the top of the flask was closed. The oil-catalyst mixture was processed for the predetermined specified time and temperature. The oil was then allowed to cool to 30-40 ¾ in the vacuum. It was then separated f m the catalyst hy c^irifugaiioa at 9000 for 15 min and filtered through a 47 mm 0 circle Whatman mlcrofib-er .filter. Bach processed oil sample was then subjected to the following, analysis.

[0050] Free Fatly Acid Analysis, The percent free fatty acids were determined is duplicate for each sample replicate, as oleic acid.

[DOS 1] Peroxide Vai Analysis, Peroxide values were deteffiiio in..duplicate- for eac sample replicate by the nncto-ffiethpd.

[0052] TAG Fatty Acid and CM o er Analysis by FAMES C?C-/¾D. The fatty acid composition of eaeh. replicate sample was determined m duplicate by GC FAMES analysis. Methyl ester were prepared by a base-catalyzed meuhylation method. One-hundred (100) rng samples were weighed into 2S»niL centrifuge tubes and 500 uL of 1 ¼ heptadecanok acid methyl ester (17:0* Internal standard), 2 ml.- of toluene and 4 raL of 0.5 M sodium -methoxide methanol were added to eaeh centrifuge tube and the tube were purged with nitrogen gas, The cemri-luge tubes were heated to 50 ^aC for 10 mhi and then cooled for 5 min. After the tubes cooled* 200 L of glacial acetic acid was added to each centrifuge tube to prevent the Ibopatio i?

Of sodium hydroxide., five (5) ml of distilled water was added to each centrifoge tube followed by 5 mL of hexane, and the tubes were vortexed. for 2 m The hexane layers were extracted; .iiad dried over anhydrous sodknn sulfate in 7 -ml, glass vials, The extracted layers were then taken from the glass via! and placed m gas cnrotnatograph vials, Methyl esters were asia!y¾ed by GC by using m BP 2560 fused silica capillary column (1 0 m 0,2.5 mm 1 & 0,2 urn film thickness; S peico !nc._s Be!le dnte, PA_:, USA) with a flame joBizatkm. detector (model Varfarj, Walton Creek, CA, USA), Two it samples, prepared, n hexanevw^:ere injected by using ah aulosanipler CF84 0 (Yari^'an) and gas chrom&tograms -were collected by Galaxie chromatography workstation 1.9.3.2 (Vanan). Analysis of variance was used to determine significant differences in. total CLA levels betwee treatments.

[0053] Variable Temperatures, Processing Times mdCaiaiyst Can elation. Table 5 above shows the CLA yields obtained by the variables in ins central composite xotats e design ex eriment The conditions tha produced the most. CL (20%) were a concentration of 0M% ruthenium, a temperatur of 210 °C and a processing time of 120 mm. reatments a femperafures < 1 0 ^*C rodnesd less than 3% CLA, Independent of time and catalyst concentration. There was significant increase in the CLA content by 3 to ?¾ in treatments: wit temperatures of 165 C compared, t treatments with, temperatures < 120 C, The oil CLA increased in the treat e ts with the temperatures > 21.0 ^"C. The increase .from 120 to 210 ^'€ ^•resulted In an. S tol.7 % increase in CLA, showing that catalytic isomerizaiion increased with temperature, indepeodeni of time and concentration. The effect of time in increasing CLA levels became more significant at higher temperatures. The statistical model (Table S) showed thai there were no Interaction effects between he variables on the CLA yields, which also predicted Increase In CLA yields with temperature, time and catalyst concentration. However, tem erature had. the greatest linear effect on CLA yields. This was followed by time and catalyst concentration effect, respectively.

[0054] TAG Fatty Acid and CLA Isomer Analysis by FAMES GC-FID, Table 6 below shows the total fatty acid, composition produced, m CLA-rich. oil processed during the isliisi study with variable temperatures, processing times and. catalyst concentration. There are no significant differences amongst treatments within palmitic, stearic and oleic acid content However, there was a significant redaction In Hnoleie and linofenic acid, which were isonierized during processing. These differences were greatest at higher temperature tre tmen s with, most CLA production. Tram._s trans CLA composition also signifkantiy increased with increasing temperature, This occurs because the higher temperatures favors formation of the thenHodynarnlcai!y stable (ram., trims CLA isomers..

Table 6: Total fatty acid oss ossr os of th* proesssed oils obtained » *¾ initial. -$t¾iiy (Example 4) aslp Central Composite R»tata&le expers ssmta design to optimize het rogeneeiiS ruthenium catalysis of 250~*»L soy oil iisofcie add to CLA over a rssge of teia erateres (90-240 tim s (40-140 mm) and catalyst eo»ce»tra^om (0,21-0.75%).

[0055] Free Fatly Acid The fee- fatty acid levels of all treatments were less than 0,06% (Table 5). These low levels were aintained due to- volatilization of any FFA .formed: during the. process.

[0056] Peroxid Value, The peroxide values of all the treatments were below 1 raeq/kg (Table 5), which Is within, acceptable limits of good quality edible oil. Although oxidation can occur rapidly when, oils containing metal catalyst are exposed to air at high temperatures, peroxide ^'values were minimked by -maintaining hig vacuum in the processing unit and llo ng oil to coo! under vacuum in. the assembly before filtration The- process was conducted, under conditions similar to deodoriigalion processing so volatile oxidation: products were readily removed to maintain, low oxidation levels,

[005?] As noted above_* the ef&ct of higher temperatures and longer^' processing times, ^•which were not studied I» Example 4,. eeded to be: subsequently investigated in an attempt to further increase CLA yields. The Example 4 stndy with variable temperatures, processing times and catalyst concentration showed that higher temperatures had the most signiiiearit effect on maximizing CLA yields* followed by longer processing; times. Hence, a high temperature, long, processing time study (Example 5 and. Table 6) was then, conducted to determine the effects- of longer times and higher temperatures^' on CLA yields. However, the catalyst concentratio was kept constant at 0.64% as higher concentrations of the catalyst loaded on carbon increased the viscosity of oil which would increase process cost and cause filtration problems.

EXAMPLE S

10 $$} The central composite rotaiable experimental design of .Example 4 produced a statistical model that predicted that higher temperatures, longer times and greate catalyst amounts would produce greater CLA yields (Table 5 above). The catalyst concentration, however, could not ^'be increased, over 0.64% because of an unmanageable increase in oil viscosity, Therefore, a subsequent optimization study (Example 5} was peribrmed at higher temperatures and longer processing times but with a fixed catalyst concentration,

[0059] High Temperature, long Processing Time Situfy. A central composite rotatable: design with. 5 levels of 2 variables, time- (108_* 120. 150. 180 and 192 min) nd temperature (19?. 210, 240, 270 and 282 ^*C>, was used. Two hundred and fifty (250) mL soy oil samples were processed. In duplicate, in random order , for each set of processing conditions^ as sho n in Table 7.

Table 7; High temperature, ioag processing time study (Example 5) usin a Central. Composite^■Rotat&bi.e experimental des gn to o timize fceteH*ge»eo«s ra.thfisa.ltti». catalysis of 2S0mL so oil Uaelek acid to CLA !i¾ % rat&eaium catalyst ove a -raage- »f temperstam (197*382 ¾) aad k¾es (!.0 -!92 rate).

[0060] Each duplicate sample w s analyzed in duplicate for .free fatty acids, peroxide value and TAG fatly acid composition b FAMES by GC~FJD, Analysis of variance was used, to determine significant differences in total CLA levels between treatments.. Contour rofile analysis generated by the central composit design modal was used to predict the range of time and temperatures to obtain the optimal CLA levels.

[0061] CL4 homers Determination by Silver-fan HPLC The positional tram, tram CLA and the other CLA isomers were quantified by silver-ion HPLC chromatography, as GC-FID FAMES fatty acid analysis does not distinguish betwee positional ira jr m CLA isomers.

[0062! $ilver*½n chromatographic separation, quantification and identification as performed in duplicates for each re licate sample. Two ChromSpher 5 Lipids analytical silver- impregnated columns each 4.6 mm id. x 250 mm stainless steel 5 urn particle size (Chrompaek, Bridgewater, X USA) in series were used. The column temperature was kept constant at 23 ¾ by using a temperature control module (Waters^' Corporation, Milford, MA, OSA), Ahetit fery μ$. of irradiated sample FAMES in 20 ni hexane was injected using the Waters- 717 plus auidsampler and a Waters model 600 system, equipped with a quaternary pump (Water Delta-. 600), umping at rat of 1 L mim A simple- solvent system consisting of 0.1% acetonitrile in hexane was nsed, The column efflnent was: connected to a...photediode array (PDA) detector (Waters Model 2996}* measuring ahsorbanee at 233 .am. The data outpiit from PD was integrated by Waters Empower™ 2 Software.

[0063] Statistical Analysis.. Statistical software IMP 10 (SAS Institute, Cary, HQ was used throughout Example 3 to randomize the treatment runs, analyze the results -of -the central com o ite desi gn, generate a contour profile of the data and to conduct a three factor regression analysis and analysis of variance (P~O,0S). The means of me treatments were compared with Tnkey's .significance test for P™0.05,

[0064] High Temperature, long Proces i g lim Hestdta,- Table 6 s ows that there is an Increase CLA. yields produced relati e to the Initial study of Example 4- Treatment 2, with levels of 240 ¾ and 108 mm, produced the highest CLA yield of 21.05%, which was significantly higher than other treatments. The statistical model produced from central- composite rotat& ie design showed that, at. higher temperatures and longer times there are significant interaction effects between, the time and temperature. There is a significant linear -and quadratic time effect and a significant linear es^er ure effect (P™0,05),.

[0065] As illustrated in Figure 2, the reduced statistical model generated a contour profile to determine the range of times and temperatures thai produced the optimal CL levels. Figure 2 shows the contour Urns as a function of time and temperatute along which the fVoictipii has a constant va ue of CLA yields. Figure 1 shows the contour lines at CLA yields of Γ8_» 18.5, 19₅ 19,5 and 20 %, The unshaded area shows the combinations of time and temperature necessary o obtain >1 §.5% C'LA. yields.

[0066] Free Fatty Acid The free fatty acid levels of all treatments were ver low (Tabl 5} because free fatty acids volatilises from the oil under high temperature mil vacu m deodorization conditions,

[0067] Peroxide Value. The peroxide values of ah the treatments were below I acq/kg (Table 5) because of the vacuum conditions which limited oxidation. These :meas»reraeuts are within the acceptable limits of good quality edible oil.

[0068] TAG Fatty Acid and CIA Isomer Analysis hy FAMES GC-Fi Table 8 Mo show the total fatty acid composition produced in. CLA-rich oil processed from the higher temperature-longer time study, respectively, A significant increase in. palmitic acid, was observed at temperatures > 270 ¾. There were no significant differences in stearic acid con ent between treatments. There was a significant increase in oleic acid content at 282 ^"C, There were significant reductions in linoleic and linoknic acid at temperatures >210 '€.

^'Table 8: Total fatty ¾cid com os tion of t¾e processed oils obtidised in. m high i&siperatere, Mag rocessing time sf». _y tts!ng a Cent l osaposlte Rotafable experi ental design to optimize: heterogeneous r¾t¾enlssa catalysis of 250i»L soy oil i!nolek id to LA. with 0,64% ratbeaiaa* catalyst over a rasge of teas erata-res 037* ¾€> sod times (108- 192

[0069] la this Example 5, it was found that the high temperature, long processing times stud lead to & sigaiikant increase in oleic, stearic and palmitic acid composition sad a significan decrease in ilnoleic- acid and linoknic acid eon^osition. compared to the initial study with variable- temperatures, processing times a d catalyst concentration, with no significant increase \n saturate fatty acids. The fatty acid compositio of the oil was mainly affected by higher temperatures and longer processing times. [0070) CLA Isomers Determination by Sifoer- n HPIC. Sllvet-ian chromatography of CLA-rich so oil produced at the optimized parameters showed 1.6 CLA Isomers in 3 geometrical isomer g ou s (Table 9),

Table 9:. Compeeitioi* of CLA isomers Is 21.05% CLA-rich oil obtained by processing I¾SI> oil at 24» ¾ for MB mla with 0.6 */» reifeeaSam eatslysi silver low chromatography.

[0071 j The tra rans CLA isomers comprised of 50% of total CLA. Of all ir&m ram CLA isomers, t-9,t-ll and t~l0_st-I2 CLA were predominant and each comprised a out 18%. The imns_feis cfs,tram CLA isomers constituted 47% of total CLA. The- cis_>cb CLA coosihuied 2% of total CLA. The first group consisted of 6 positional trans ns CLA isomers, the second roup -of 8 positional jt&n$/tr m.,cis isomers and the last group of 2 positional€is,cis isomers. The control soy oil F MES chfomaiogram did not show any CLA peaks because- conjugated CLA was absent (data not shown), EXAMPLE 6

[0072] The central, composite rotatabie ex eriments!, design of Example 6 decreased the catalyst concentration below 0,64% ith the addition of formic acid and tocopherols- in order to overcome the oil viscosity seen in Example 4. Example- 6 ased non-adsorbed unbleached soy oil that was nly partially refined, /.<*.,, oil. thai had not been, absorbed to remove residues -from prior refining. Similar to prior examples, processing conditions were optimized to maximize CLA yields using central composite experimental design, studies with, variable time, catalyst concentration, formic acid, concentration in both, the oil and steam and variable tocopherol concentrati ons (Table 10 below). The process of Example 6 produced similar CLA yields to that obtained by fully refeed soy oil used ¼ the prior examples.

Table 10. Central com osite rentable desi n for Exam l for opiim&jiiioa of 2SfesL my oil finolele acid to CLA over higher tempe atures* !oager rocessin times, mi

rutfeeaiam catalyst eose atrai ss with the d litioa formic acid to the oil 4fot the

Ύ7

** Acetic Acid

[0073] The catalyst residue dissolved is the oil could subsequently be removed by the adsorption processing to .reduce residual catalyst in the oil alter being subjected to the process. For example, R.u catalyst levels should be <0.5ppm to be similar to Ki level allowed alter hydrogenation. la addition, the adsorbent binding the Ra could possibl be used as a cata yst.

[0074] Formic acid is a. volatile organic acid that was added to both the steam and the oil in Example 6 to r mote Ra catalytic activity by Ru solubilization, in. the low pressure system of the present process, foimic acid is lost rapidly and is added continuously with the steam. Alternatively, formic acid can be applied to the oil only during the beginning of the process to activate the catalyst, and subsequently applied with the steam for a predetermined amount of time, such as up to about 30 mins or between about 30 ibs and about 60 rains,

[0075] In addition to for ic acid, tocopherols were utilised in the processing of Example 6, One percent (1%) was previously found to be effective with iodine catalysis. Since tocopherols are a co-product of oil processing, there addition to the processing would neithe be cost prohibitive nor need to be purchased t¾r an. industrial system,

[0076] Figure 3 illustrates the effect of ruthenium on. the CLA yield at .270 X. for 3 hours with 4% .fbrmk acid In. oil and steam of 85 % formic acid. Tables 11, 1 and 13 below illustrate the various effects of formic acid i oil and steam on CLA yield, and as can be seen, CLA yield Increases 4,6 times in. the presence of formic acid In oil, whereas steam formic acid results in less than I times increase.

Table !t: Effect of formic add in ml and stream w CLA yield af OJ.28% ofratfceata, at.

270 °C for a 3 ¾rs re ction

Table 12: Effect »f ismperatnre n CLA yield at 0.128% of r the ium, 4 formic acid is oH a¾d 85% formic acid of steam for a 3 fers reaetloa.

Te pera a e (°C) CLA. field (%) ¹

270 25.8 ± 1.1

•240 1 19.4 ± 3.6 Table 13: Effect of reaction: time m CLA yield, at 278 8.128% of ruthenium, 4% formk acid m oil a»d 85% forssk ac d ftf tfenv, ^*

EXAMPLE 7

[0077] The objective of Example 7 was to produce and characterize€LA-dc so oil rn¾rgarine relative to a soy oil control and commercial margarine, CLA-rieh soy oil was used to prepare margarine. The samples were characterized fpr firmness, theology, thermal behavior, solid fat- content (SFC) and mierostruetare and compared with a soy oil control, and commercial margarine, The CLA-rich oil margarine firmness and theological properties w re: similar to commercial margarine and -provided a better texture .relative to the soy oil control, margarine,

[0078] RBD soy oil was obtained from Riceknd Foods (Stuttgart, AR, USA) and used throughout the study. Palm stearin, mono- and diacylglyeerides were provided by Archer Daniels Midland (Decatur, 1L, USA},. Commercial Great Value margarine was purchased from almart store (Benionv!ile, AR, USA), Lecithin was purchased from Spectrum chemical, n anaiaeturing corporation (Mew Brunswick, J, OS A), Other chemicals used were reagent grade,

[007^1 CLA- Rich Soy Oil Production and Fatty Acid Analysis, CIA-rich soy oil was produced by the heterogeneous catalytic me h d of the prior xam les discussed above. The fatty acid composition, of soy and CLA-rich. oil were determined m triplicate using the base catalyzed derivati¾atio» technique to produce FAME and subsequent GC-F!D analysis.,

| 80j Margarine Preparation- Optimization of hard-stock content in fa phase for margarine production. Bardstoek. optimization was determined with 30, 40, 50 % palm stearin (hard stock;} in 18,85 % CLA-rich soy oli in. duplicate. These blends were used to produce mar arines in duplicate. The oil phase consisted of (w/w, ^'%}: 79.37 % fat, 0,42 % mono- and di~ acvlglycerides, 0.2 % soy lecithin and 0.0048 % beta-carotene. The aqueous phase cons sted of (w/w. %}: 17.65 % distilled water and 2.25 % table salt. Both phases we e vigorously mixed, for 10 using & Black & Decker M 300 omogei K r (New Britain, C _? USA) for emulsiilcatlon, lie resulting: emulsion w s then crystallized using half-pint Hamilton Beac ice-cream .maker (Glen Allen, ¥ A. USA). The bowl w s .maintained at around 10 *C during the crystallisation. The resulting crystallized■ emulsion: was en tempered, at room tempe atur for 4 h and then worked vigoroirsiy with a. hand mixer to obtain a smooth and consistent texture. Th : resulting paste-type margarine samples (300 g each) were placed in sealed tubs ami .stored t 5 prior to analysis.. Microscopic analysis showed that the margarine prepared with 50 % palm steari (hard, stock) had least air incorporation, and was used In. subsequent studies.

fOO^ij Margarine roducts ft fit? characterization. The 50 % palm oil stearin Oiardstock) was used throughout in this Exam le 7. GLA-rioh soy oil margarine and a control margarine prepared with .conventional soy oil. were produced in. duplicates,

[0082] Margarim C ar wieriui n. CLA-ricb soy oil margarine and .soy oil control margarines were compared with a commercial Great Valac margarine (Ws!mart Bentor iiie, AR, USA),

[0083] Firmmss Analysis, Measurements were performed, on. five replicates: selected randomly from the two margarine batches. Samples were placed in S c n x 3 cm. plastic cu s and stored at 5 *C in thermostatic cabinet for an hour befo e: analysis. Firmness was : detemt!ned as force required or penetration using a 5942 Instron TA 500 Texture An lyser (Lloyd Instrument. BognOr Regis, West Sussex, ΌΚ). A 1 1 mm diameter cylindrical probe penetrated die sample to a depth of 10 mm at a rate of 10 mm/nnn with 0.1 N trigger value at 5 °C. 3!

[ 0 4] Mheahg . The theology was deiemuned in triplicate, for all mar arine sam les. An advanced rheorneter AR 2000 (TA instruments, New Castle, 0¾ USA) using a paralle , plate with a cross hatched geometr (diameter 40 mm) was used. The geometry gap of the rheorneter was set. at 1,000 μι». Oscillatory stress sweeps were carried out to gain insights into the rlKiologieal behavior of margarine samples. Oscillation procedure was performed in three steps; Firstly, a conditioning step at 5 "C as Initial temperature for a uration of IS in followed by a stress sweep fronj 0.1 to 1.0,000 (Pa), to determine the linear v coeiasiie region at a fee ueney of ! Hz. Finally, a postexperiment step was performed a 5 °C. Complex modulus (G*) and phase angle ^'(§} were measured as a ^'function of oscillatory stress, O* is the measure of the total resistance of sample to deformation and phase angle (which, is a ratio of loss and storage moduli) is a. measure of 'relative viseons or elastic behavior of the sample. The phase angle ^'values indicate the degree of viscosity and elasticity. Small angles indicate, solid-like behavior, wit increasing angles indicative of an increasingly liquid behavior.

[0085] Thermal Behavior, The DSC experiments were performed with a QI0CM1 T¾ero DSC (TA Instruments, New Castle, DH, USA) on triplicate margarine samples. The DSC was calibrated with indium (enthalpy and temperature), azoberrsene {temperature}, and undeeane (temperature) before analyses. ^"Nitrogen was used to purge th system -and an empty pan was used as a reference. The margarines were sampled (5-10 mg) in hermetic alnniinuni pans .and^' sealed. The melting profile was recorded by applying a time-temperature program as follows: initial temperature at 5 °C (followed by loading o crip), holding for 5 mm and heating at 5 "C/min to SO X The integration of obtained melting curves was erformed using the Universal Analysis software ^'(TA Instru e ts, New ^'Castle, USA). [0086] Solid Fat Content (SFC) Determination., SFC~proftle_: was measured by 23 MHz. !H KMI¾. Ma an instrument (Oxford Instruments; Oxfordshire, OK) m. triplicate. Prior io analysis, the fat phases of the margarines were obtained after destahiiizaiioo at 65 ⁸C followed by filtration over sodium sulfate. Molten fat was placed in R tubes and submitted^' -to the tempe a ure treatments of the official method AOCS Cd ldb-93. The SFC was determined feorn 5 "C at 5 ^aC intervals until completely melted following 30 min incubations at each specific temperature.

J00S7}. Micmsimetu e. Microscopic analyses were conducted, by the use of Leitz Diaplao. microscope (Leica Microsystems CMS G bH, ePdar, Germany) eq ipped- with a Lffikam £E 94 t mperature control system (Linkanv Surrey. Germany) in triplicate. Since, the undiluted samples were too dense to be viewed clearly under the microscope; slide preparation as done by smearing: th sample on the slide followed by dilution with a. drop of oil for bette visual assessment. The slides were placed on the. temperature -controlled plate to visualize die •miemstrucnae under^' transmitted polarized and normal light at 20 ¾. Images were recorded with an Olympus* Color View camera and processed with -Cell D software (Olympus, A selaar Belgktm)>

fO SS] Dropki Size Distribution. Triplicate water droplet size analysi of the samples was performed by pulsed field gradient nuclear magnetic resonance pfg-N E.) o bench-top aran Ultra spectrometer (Oxford instruments, UK) operating at a frequency of 23.4 H in combination with the droplet size application. Samples were analyzed at 5 X;- to minimize inter- droplet water difi sion. To suppress the NM^'R contribution of the fat phase, p%-NMi. experiments wer conducted using, an. Inversion recovery stimulated echo pulse sequence. In the performed experiments, the diffusion time (A) was set to 0.2 s: the gradient strength was feed at 1.74 Τ ΪΒ, while the- gradient duration (8) was varied in I? steps from 400 to 4.500 ms. By measuring the echo attenuation ratio of the NMR signal as a inaction of the gradient duration., it Is poss ble to determine the hindered diffusion behavior and hence the droplet si¾e distribution_* Polydispersity index (FDI) was then calculated front the mean and standard deviation values, il^¬ ls the-rneasnre of non-uniformity of water droplet ske distribution m the margar ne. O089] Statistical Analysis, JMP 10 (SAS Institute Inc., Cary, NC> was used to perform analysis of variance (ANOVA) and Student's i test to compare means (i⁵<0.05).

[0090] CIA-Rich Oil Fatty Acid Analysis. Table 14 below show the felly acid composition ofCLA-rich. soy oil and conventional soy oil, used, to produce CLA-rieh soy oil and control soy oil margarines. The soy oil has a typical soy oil y acid profile with 55,86 %

I oleic acid as a dominant fatty acid. However, CLA~rieh soy oil comprises of 13,25 % cisjram/imn$,cis CLA isomers and. 5,6 % are tram, tram CLA isomers to produce s iota! of

18.85 % CLA-rich soy oil with a 35.71 % reductio in linoieic acid. About 16,8 g of the CLA* rich soy oil will provide 3 ,2 g CLA that is required to provide clinical benefits,

Ta e 14: Fatty acid com osition of mv oil rsd CLA-rieh aov oil measure I» triplicate s <SC~FI» FAME,

Fatty add Soy oil C%) CLA-rwb sw (%}

P&lrnifie acid 13.48 * 0,14 13,27 ά 0.10

Stearic acid 4.68 ± 0.12 4.98 * 0.06

Oleic acid 24.11. ± 0.22 24,24 * 0.26

Lino!elc acid 55.86 ± 0.24 35,71 * 0.86

Lino!cnlc acid 4.S4 ± 0.H 1.62 i 0.03:

imns x/ds.tram CLA 1 3.25 0.12

tm m CLA 5,60 ± 0.03 [0091] Ftmims Analysis, Figure 4 shows the comparison plot of average force measured versus compressive ex ension for CLA-rich oil margarine, control soy oil marga ne; and comm rci l margarine. The results were reprod cible, as the standard deviations were th 5 % of the mean, force values or each margarine sample. The CLA-rieh soy oil and control soy oil margarine load curves were significantly different from mat of commercial margarine, with ahseri.ee of a defined fracture force indicating a more plastic structu . However, the hardness of CLA-rich soy oil margarine was sigmfeanfly greater than that of the control soy oil margarine (P<Q ) and similar to tir commercial margarine. Hardness is one of the most important ffiacrostr ctnra! pro erties of fa systems and. s widely used, to characterize margarine functionality. Therefore, ClA-tich oil may contribute to the hardness tha consumer desire in- a commercial margarine, possibly by strengthening the lipid crystal network. This could ^'be possibly due to the presence of linear trans, trans CIA. fatty ¾eids which are more readily incorporated than LA Into crystal networks.

[0092] The^' firmness of CLA-rich soy oil margarine and commercial margarine are quits similar. However, the commercial margarine force-extension curve differed by having a prominent fracture force followed by a declme in slope of load curve. This indicates that the commercial margarine has certain fxaetorabtiity leading to a breakdown in structure as the deformation is induced, which is not present i the others margarine curves,

[0093] Rh ofagy. Figure 5 shows the complex modulus (<?*) versus oscillator stress relationship for each margarine. The linear viscoelastlc region was narrower for control soy oil margarine (1-10 Fa) relative to CLA-rich soy oil margarine and. commercial margarine (10- 1 ,000 Fa), The critical value of oscillator stress (<100 Fa) required to deform the SAY l margarine sfmcmre was smaller than that for CLA-rich soy oil and commercial margarine (> 1,000 Pa). This means that the microstructure of CLA-rich soy oil and commercial margarine were more resistant to defoliat on under oscillatory stress, than the control soy oil mar arin . The CLA-ric soy oil and commercial margarine rheoiogv curves are similar and are consistent with the similar firmness -me suremen s for these samples,

[0094] Figure 6 shows the plots of phase angle with applied oscillatory stress. The results obtained from CLA-rich so oil margarine and commercial margarines were comparable. There was a gradual Increase in the phase angle (5-15*). until oscillatory stress of around 5,000 Pa. However, on further additional stress there was a sudden increase in the phase .angle to 3S^a. indicating that both CLA-rich soy oil margarine and commercial margarine maintain their elastic solid structure ven at fairly high stress amplitude of 8,000 Pa. in contrast, control soy oil margarine showed a gradual increase in the phase angle from: 10 to .20* up to 400 Fa of oscillatory stress. On further Increase, in amplitude of stress, the phase angle suddenly increased to -8 *_> indicating rapid transformation to a much mors viscous lkpid ike substance. This suggests that control soy oil margarine has less elastic fat behavio at any given oscillatory stress- tha CLA-rich. soy oil. and commercial margarine. Therefore, larger amplitude of oscillator ^•stress is required to transform the elastic solid character o -CLA-rich soy oil and. ommercial margarine to a viscous li uid.

10095] Thermal Behmior. Figure ? shows the melting behavio of margarine samples monitored by differential scanning caiorirneiry. The results were highly reproducible a the triplicate curves from each replication were identical. The melting profiles of GLA-rieh oil and soy oil margarines were similar. Both samples have a very prominent melting fraction around about 58 C; 57.6 X for CLA-rich soy oil margarine and 58.2 for the control soy oil margarine. This similarity is to be expected since the same SFC was used,. However, ihe ^'commercial marganne showed two broad, shallow peaks at 133 and 29 ^th The meltin profile of commercial margarine was very dlfisrerit, indicating a different formu os- tha the other samples. Despite of the similar melting profiles of the CLA-ric soy oil and control soy oil- margarines* CLA»rich soy oil margarine made Much better eon butio to -firmness and theological properties that closely resembled the commercial oduc This s gests that the CLA components in the soy oil significantly co.ritribu.ted ¾ margarine texture, w ich eookl be related to. the tr m, trans CLA. content

|00 6 | $F€ D ienm uioft Figure 8 shows the S.FC curves of control soy oil margarine_* CLA-rich soy oil margarine and commercial, margarine as a function of tempetaiure. Comme c l margarine had significantly lower SFC as compared to control, soy oil and CLA~rleb soy oil margarines at all. temperatures (ft <QM}> CLA-rteh oil and: control so oil margarines were similar and had a more gradual decrease of S.FC with increase in temperature than the commercial margarine. These results are consistent with the corresponding DSC molting profiles.

[0097] There w a slight, but significant, difference between the CLA and soy oil control margarine samples between 5 and 15 ° , CLA-ricIi oil has slightly but significantly greater SFC in that range of temperatures (5-15 °€) (P<0,O5). This could partly explain the increased ^'hardness (see Figure 4) and significantly different theological properties (Figures 5 and 6¾ which were both measured at 5 ^l-C. However, at temperatures over 15 *C there was no significant difference in SFC content between the soy" oil. md CIA-rich oil margarine (i -05). This data shows that SFC nd. melting profile are not the sole factor influencing the texture and strengt of a fat crystal network and indicate that presence of CLA in margarine may contribute^" to the hardness. However, SFC and hardness do not always have a linear correlation. [0098] Microstmct m Microscopic images, of control s oil margarine,: CLA-rich soy oil margarine and commercial margarine are shown, m Figure 9. There was significant air-- incorporation in the control soy oil and CLA-rich soy oil margarine, as shows by the thick, black outlined circles in normal light microscopy, which w re minimal in. commercial .margarine. The size of water droplets, seen as white solid particles in polarized %ht microscopy, were larger in. CLA-rich soy oil and control, soy oil margarine (r ~ S-20 μηι) than the commercial margarine fr < 5 m)>

[0099] Dropi&i Ske Distribution. Table 15 below shows the average volume weighted water droplet size and PDl of the control soy oil margarine, CLA-rich soy oil. margariae and. commercial margarine. The water droplet size of commercial margarine is approximately five*-, to sixfold smaller than control soy oil and -CLA-rieh soy oil margarine. As seen .from Table I S, the control soy oil margarine has highest PDl (0.12), followed by CLA-rich soy oil margarine (0,06).. while commercial margarine has least PDl (0.01). The droplet size distribution data corresponds with the icrostrocture shown in Figure 9,

Table ISt ^'Water droplet size and oiydls erslty in ex of control soy oil margarine, CLA- rich soy oil marg r ne m$ commercial msz &ri .

[0100] ^'The better water droplet distribution and smaller water droplet size of the commercial margarine reflects^' the improved industrial preparation techniques, liomogen.izat.ion and scraped surface heat exchangers* used at commercial scale than the samples produced with the hand blender and ice-cream maker used to prepare- experimental margarines. [0101 J As can be concluded from Example 7, CLA-rieh soy oil provides- -atx improved^' margarine texture than control soy oil The CLA-rieh soy oil mar arine hardness and rheologica! properties obtained at 5 °C were comparable to commerciai margarine but SFC, drople siz distribution and melting behavior of CLA-rieh oil margarine were similar to control soy oil margarine, except at 5-15 *€_> which explains iinnness and theology differences.. This suggests thai hardness and. rheologleal properties of margarine ate not solely dependent on SPC and meiting behavior. Lipid composition, polymorphism and microstrueture differences tn CLA-rie s oil relative to the control soy oil- may play an important role on the texture and rheoiogical properiies ot msrgatine.

[0102] A 7 g typical serving of the experimental CLA-rieh oil margarine will provide 0.6 g CLA, T ms five servings will provide about 3.2 g day of CLA arid 185 calories day _s which is well within the maximum recommended 7D0~$8G fat calories/day, However, there was more hard Cat in the Example 7 samples than would b used in. a commercial product. Therefore, industrial production would he able to increase the CLA content in a serving: at he expe se, of hard fat due to improve processing.

EXAMPLE 8

[0103] The objective of Example 8 was to produce and characterize CLA-rieh. say oil shortening relative to a commercial, shortening, namely Criseo*. Similar to Example 7 for margarine above, CLA-rieh soy oil was use to prepare shortening this Example 8. The samples were characterised for hardness using puncture force test, compared wit a commercial, shortening. The CLA-rieh. oil. shortening hardness was similar to commercial shortening.

[0104} As shown Table 16, .for the method of Example 8, palm stearin and soy -oil were heated to 70-77 ⁸C while stirring constantly. The mixture was poured into a stainless-steel eaker and immediately placed in 10 °C water bath while stirring constantly with a hand-mixer on the lowest speed setting. The mixture was cooled to a predetermine cooling temperature, namel , 21 *C₅ I S °C or 15 °C. The resulting shortening was placed in. plastic centrifuge tabes and stored at room temperature for at least one week to temper. Puncture forc tests were carried out to explore CLA ranges and harnesses, and were performed with TA-XT2I texture^■analyzer 2mm diameter cylinder probe for CLA shortening formulation and Crisco*. a commerci l shortening.

Ta¾I« Hi Shorteaiag WwemKtattom.

f01O5^'J As Shown in Figure 10, the commercial shortening had a psncture force of 0.5 N. Using regression analysis, best-tit line fo samples cooled to each of the three temperatures was determined; 2 ^'PC; y * -0.1 ^'!OSx + 7.8418, with the intersection with commercial control at x ^« $6.0% CLARSO; WV: y =^» -0.0442.x + 4.0194, with the intersection with commercial control at x - 79,0% CLARSO: and 15*0: y = -0,Q265x + 2,936% with the intersection, with commercial control at. X ~ 90.9% CLARSO, Based on this, the ideal CLA shortening formulations to make a shortening product having similar hardness properties to the commercial control,€riseo are below in Table 17.

Table 17; Ideal CLA shortenin fenaulatieBS to make 'Crisco Ideaticai* (0,5 N) product..

[0106] In conclosion, the shortening formulations of this Example 8 gave the same puacmre force as the commercial, control, shortening.

f 0.107} Whereas, the processes and compositions have been described in relation to he drawings and claims, it should he understood that other and farther modifications,, apart from those shown or suggested herein, may he made within the scope of this invention.

Claims

4 !

\ L A process .for producing conjugated linolele acid-rich oil, said process comprising the

2 steps of:

3 heterogeneonsly catalyzing a ijno!eie acid-rich oil and a catalytic amou.it· of a transition metal having a vacant 3d- or 4d~orb a! to produce said conjugated imoleie

5 acid-rich oil.

1 2. The process of Claim 1 further comprising the step of extracting said transition metal from said conjugated linojeie acid-rich il

! 3, The- process of Claim I wherein said linoleic acid-rich oil is a inaeylglyeeride vegetable

2 oil ί 4. l¾e process of Claim 3 wherein said inaeylglyeeride vegetable oil is -selected from the

2 group consisting of soy, sunflower, com, cottonseed or peanut oil.

1 5, The process of Claim I wherein said transition metal is selected from the grou consisting of ruthenium., rhodium, molybdenum, palladium, gold, sil ver, copper, iron, manganese or nickel.

1 6, The process of Claim 5 wherein said, transition metal is ruthenium, or nickel

1 7. The process of Claim I wherein said catalysis step farther comprises catalyzing said linoleic acid-rich oil and said transition: metal in pressure conditions between approximately ! 3 mm Hg and approximatel 2 .mm Hg for between approsimaieiy 1.08 .minutes and approximately 240 minutes to produce said conjugated Imoleie acid-rkh oil,

1 . The -process of Claim 7 wherein said catalysis step further comprises catalysing said

2 Hnoiek cid-rich oil and said transition metal at temperature conditions etween approximately

3 165 °C and approximately 282 ^δ€.

1 9: The process of Claim I wherein said catalysis step, further comprises catalyzing said

2 linoleic aoid^rich oil and said transition metal, at substantially vacuum pressure fm up ¾

3 approximately 240 minutes to produce said conjugated iinolelc acid-rich, oil

\ 10, The process of Claim. 9 wherein said catalysis step further comprises catalysing said:

2 linofeic acid-rich oil and said transition metal at temperature- conditions between approximately

3 165 °C and approximately 282 °C for between approxfniateiy 60 mirmtes and approximately 240

4 minutes to produce said conjugated linoleie acid-rieh oil. ί 11 , The process of Claim 10 wherein said catalysis step further comprises catalyzin said

2 linoleie acid-rich oil and said- transition metal between approximately 1 mm Eg and

3 a proximatel 2 mm Hg at a temperature of between approximately 210 ^< C and approximately

4 270 °C for between approximately 12 ) minutes and approximately 180 minutes. 1.2. The process of Claim. 10 whereip said catal tic amount of said, transition metal comprises

2 between.0,ø§% to approximately %% of said transition, metal. 1 13» The process of Claim 12 wherem said catalytic amount of said transition metal is between approximately 0.08% to approximately 0.75% of a ruthenium catalyst or approximately of a

3 nickel catalyst,

1 1 The rocess of Claim I wherein said catalysis step further comprises catalyzing said

2 linolcic acid-rich oil and said transition metal in pressure conditions between approximately I

3 mm Hg and. a proximately 2 Eg and te peratore- conditions between approximately 165 °C

4 and approximately 282 ^SC to produce said conjugated !inoleic acid-rich oil ϊ 15, The process of Claim 1 wherein, said catalysis ste further comprises catalysing, said liijoleie acid-rich oil and said transition meial for between approximately 60 minutes and

3 approximate I y 240 mi nines ,

\ 16. The process of Claim 15 wherein said catalysis step further comprises heterogeneous catalysing via vacuum distillation, of said linolelc acid-rich, oil and said transition racial at a

3 temperature of between approximately 210 ^eC and approximately 270 °C and at a substantially vacuum pressure tb.r between approximately 120 minutes and approximately .180 minutes.. 17, The process of Claim 14 wherein said catalytic amount of said transition metal comprises between approximately 0.08% and approximately 8¾ of said transition metal.

18. The process of Claim ! 7 wherein, said catalytic ammmt of said transition metal is between approximately 0.08% to approximately 0.75% of a ruthenium, catalyst or approximately 8% of a nickel catalyst:.

1 , The process of Claim I wherein said catalytic amount of said transition metal comprises between 0.08% to approximately S¾ of said transition metal.

.20. The process of Claim 1 wherein said catalytic amount of. said transition metal is between approximately 0.08% to approximately 0.75% of a ruthenium catalyst or approximately -8% of a nickel catalyst.