Food Research International 55 (2014) 436–444

Contents lists available at ScienceDirect

Food Research International

j ourna l homepage: www.e lsev ie r .com/ locate / foodres

Effect of diacylglycerol addition on crystallization properties ofpure triacylglycerols

Roberta Claro Silva a,⁎, Fabiana Andreia Schäfer De Martini Soares a, Jessica Mayumi Maruyama a,Natalia Roque Dagostinho a, Ylana Adami Silva a, Guilherme Andrade Calligaris b, Ana Paula Badan Ribeiro c,Lisandro Pavie Cardoso b, Luiz Antonio Gioielli a

a Department of Biochemical and Pharmaceutical Technology, FCF/USP, Av. Prof Lineu Prestes, 580, São Paulo, SP CEP 05508–900, Brazilb Institute of Physics Gleb Wataghin, University of Campinas, Brazilc Department of Food Technology, School of Applied Sciences, University of Campinas, Brazil

⁎ Corresponding author.E-mail address: robertaclaro@usp.br (R.C. Silva).

0963-9969/$ – see front matter © 2013 Elsevier Ltd. All rihttp://dx.doi.org/10.1016/j.foodres.2013.11.037

a b s t r a c t

a r t i c l e i n f o

Article history:Received 18 September 2013Accepted 23 November 2013

Keywords:TriacylglycerolsMicroscopyTripalmitinTristearinPolymorphism

The objective of this study was to investigate the effects of blending triacylglycerols (TAGs) and diacylglycerols(DAGs) on themelting and crystallization properties in a fat system. To this end, differential scanning calorimetry(DSC), X-ray diffraction (XRD) and polarized light microscopy (PLM) methods were used. Different DAGs(diolein—OO, dipalmitin—PP and distearin—SS) were added at 5% to each TAG (triolein—OOO, tripalmitin—PPPand tristearin—SSS). DSC results showed that the addition of DAGs delayed the onset of crystallization of saturat-ed TAGs (PPP and SSS). By contrast, the addition of DAGs to unsaturated TAG (OOO) accelerated the onset of crys-tallization, with the appearance of an extra crystallization peak upon the addition of SS and PP. PLM resultsrevealed that the addition of OO affected the polymorphic transition of the TAGs studied findingswere consistentwith DSC melting curves and XRD results.

© 2013 Elsevier Ltd. All rights reserved.

1. Introduction

Edible oils and fats are essential nutrients in the human diet, playingthe important role of supplying essential fatty acids and energy. Fats aremade up primarily of TAGs, representing approximately 98%, with theremainder of the fat comprising more polar lipids such as diacylglycer-ols (DAGs), monoacylglycerols (MAGs), free fatty acids (FFAs), phos-pholipids, glycolipids, sterols, and other minor components (Metin &Hartel, 2005).

The crystallization behavior of lipids has important implications forthe industrial processing of food products whose physical characteris-tics are largely dependent on fat crystals. Such products include choco-lates, margarines, spreads, confectionery and bakery fats, dairy productsand general purpose shortenings (Sato, 2001). In the majority of foods,crystallization of triacylglycerols (TAG) is the most important, althoughcrystallization of other lipids (i.e., monoacylglycerols, diacylglycerols,phospholipids, etc.) may also be important to product quality.

TAGmolecules are inherently able to pack in different crystalline ar-rangements or polymorphs, whose melting temperatures differ signifi-cantly [19,20]. The polymorphic forms of fats are often simply classifiedinto three categories, α, β′, and β, in increasing order of stability. Thus,the α form is the least stable polymorph, having the lowest meltingpoint and latent heat of fusion whereas the β form is the most stable,

ghts reserved.

having the highest melting point and latent heat. Each polymorphicformhas distinct short spacings (distances between parallel acyl groupsin the TAG), used to distinguish the polymorphic forms based on theirX-ray diffraction patterns.

Minor lipids include compounds of greater polarity andwith amphi-philic structure, such as diacylglycerols (DAGs), monoacylglycerols(MAGs), free fatty acids, phospholipids and esters. These constituentshave been considered molecular agents that affect crystallization. Thepresence ofminor lipids can promote crystallization,whereas an inhibi-tion effect is observed in some systems (Metin & Hartel, 2005; Sato,2001; Smith, Bhaggan, Talbot, & van Malssen, 2011). According toToro-Vazquez, Rangel-Vargas, Dibildox-Alvarado, and Charó-Alonso(2005), these compounds modulate the entire crystallization processfrom nucleation to post-crystallization events including polymorphism,driving force and solid fat content (SFC).

Diacylglycerols represent theminority class of lipids that are of mostinterest in studies of crystallization of lipids, since they occur at higherconcentrations in practically all fats of vegetal or animal origin(Craven & Lencki, 2011a, 2011b). Therefore, with regard to the actionof DAGs on the velocity of crystal nucleation and growth, the relevanttechnical literature shows that this glycerol class can exert a promotingor inhibitory effect on crystallization, conditioned mainly by the com-patibility of its composition with that of the raw materials. Accordingto Wright, Hartel, Narine, and Marangoni (2002), the ability of DAGsto act as modifiers of the crystalline behavior of TAGs is primarily relat-ed to the similarity in chemical composition between these glycerol

Fig. 1. Crystallization curves obtained by DSC of pure tristearin and with addition of 5%diacylglycerols.

437R.C. Silva et al. / Food Research International 55 (2014) 436–444

classes. Studies reported to date suggest the parameters defining thisdegree of similarity are the isomerism or stereospecificity of the DAGsand the composition in fatty acids, with relation to chain size and satu-ration of the fatty acids.

In a review by Smith et al. (2011) on the effects of minor andadditive lipids on the physical properties of lipid systems, the authorsconcluded several rules. For instance, when the acyl group of theminor lipid is similar to those present in the fat, a greater effect oncrystallization was observed (Elisabettini, Desmedt, Gibon, & Durant,1995; Garbolino, Bartoccini, & Floter, 2005; Smith & Povey, 1997). Inaddition, super-cooling reduced the effects of minor lipids (Cheong,Zhang, Xu, & Xu, 2009; Wright, Hartel, Narine, & Marangoni, 2000).The concentration of the minor lipid directly influenced the crystalliza-tion process of fat (Herrera & Marquez Rocha, 1996).

In relation to crystallization, DAG constituents have been studied invarious fats and oils such as milk fat (Wright et al., 2002), palm oleinTAGs (Siew & Ng, 1996), palm oil (Siew & Ng, 1999; Verstringe,Danthine, Blecker, Depypere, & Dewettinck, 2013), trilaurin (Smith &Povey, 1997) and lard blends with rapeseed oil (Cheong et al., 2009)and are widely reported as inhibitors of the crystallization process.

Studying the crystallization behavior and polymorphism of a purelipid system is of great scientific importance as ameans of gaining anun-derstanding of the phenomena involved, serving as basic knowledge tohelp guide the addition or removal of these compounds in differentrawmaterials. Thepurpose of the present studywas therefore to charac-terize and compare the effects of the addition of 5% pure diacylglycerols(diolein——PP and distearin—SS) to pure triacylglycerols (triolein—OOO,tripalmitin—PPP and tristearin—SSS) using differential scanning calo-rimetry (DSC), polarized light microscopy and X-ray diffraction.

2. Material and methods

2.1. Material

2.1.1. TriacylglycerolsThe samples were acquired from the company Sigma-Aldrich

(United Kingdom).

Glyceryl trioleate N 99% (OOO)

Glyceryl tripalmitate N 99% (PPP)Glyceryl tristearate ~99% (SSS)

2.1.2. DiacylglycerolsThe samples were acquired from the company Nu-Chek Prep. Inc.

(USA).

diolein N 99% (OO)

dipalmitin N 99% (PP)1,3-distearin N 99% (SS)

2.2. Methods

2.2.1. X-ray diffractionThe polymorphic form of the fat crystals in the sample was deter-

mined according to the AOCS Cj 2-95 method (2004). Analyses werecarried out on a Philips diffractometer (PW 1710), using Bragg–Brentano (θ:2θ) geometry with radiation of Cu-Kα (λ = 1.54056 Å,tension of 40 kV and 30 mA). Measures were attained with 0.02° in 2θsteps and an acquisition time of 2 s, using scans of 5–40° (2θ scale).The samples were melted in a microwave oven at approximately80 °C and stabilized at 25 °C for 24 h. Analyses were carried out at25 °C. Polymorphic form identification was performed based on thecharacteristic short spacings of the crystals. Formα presents a single dif-fraction line at 4.15 Å. Form β′ is characterized by two strong diffractionlines at 3.8 Å and 4.2 Å,whereas formβ is associatedwith a series of dif-fraction lines exhibiting a prominent line at 4.6 Å and lines of lesser

intensity at 3.7 Å and 3.8 Å (Ribeiro, Basso, Grimaldi, Gioielli, &Gonçalves, 2009; Rousseau, Marangoni, & Jeffrey, 1998).

2.2.2. Differential scanning calorimetry (DSC)The DSC curves were obtained by the differential scanning calo-

rimetry (DSC) cell on a DSC 4000 Perkin Elmer (Perkin ElmerCorp., Norwalk, CT, USA) under a dynamic atmosphere of He(20 mL/min), a cooling rate of −10 °C/min, at temperatures rangingfrom 80 to −60 °C with an isothermal time of 10 min at 80 °C, usingsealed aluminum capsules containing a sample mass of 5 to 10 mg.The melting curves were obtained from −60 to 80 °C (5 °C/min) withan isothermal time of 30 min at −60 °C. The temperature and heat ofmelting were calibrated with indium (initial temperature of 156.6 °C).Curves were processed by Pyris software and crystallization curveswere analyzed for onset of crystallization (Tonset °C), peak crystallizationtemperatures (Tpeak °C) and crystallization enthalpies (ΔHc J/g) (Ribeiroet al., 2009).

2.2.3. Polarized light microscopyThe samples were heated to 70 °C in an oven and held at this tem-

perature for 30 min to destroy the crystalline memory. One drop ofmelted fat was withdrawn using a capillary tube and then mounted ona slide with a similarly pre-heated coverslip. The prepared slide wasplaced on the temperature controller of the microscope. The samplewas first subjected to 100 °C for 15 min to ensure total melting of thecrystals. The sample was then cooled to 15 °C (rate of 10 °C/min)using a Linkam temperature controller,model EP-120 (Surrey, England).The crystallization imageswere captured every 2 s from45 °C (PPP) and55 °C (SSS) until complete crystallization of the sample. Subsequently,the samples were maintained at 15 °C for 40 min and melting thenevaluated. The samples were heated to 100 °C (rate 5 °C/min). Meltingimages were obtained every 2 s from 35–40 °C (PPP) and 50–53 °C(SSS), until completemelting of the sample. The diameter of the crystalsafter complete crystallization was determined for each sample usingImage-Pro Plus version 7.0 software (Media Cybernetics, USA). Theimageswere used to determine the beginning and end of crystallization,induction time and crystallization time.

Fig. 2. Melting curves obtained by DSC of pure tristearin and with addition of 5%diacylglycerols.

Fig. 3. Crystallization curves obtained by DSC of pure tripalmitin and with addition of 5%diacylglycerols.

438 R.C. Silva et al. / Food Research International 55 (2014) 436–444

3. Results and discussion

3.1. Crystallization and melting profiles

The crystallization andmelting profiles by DSC of pure SSS and withthe addition of diacylglycerols are depicted in Figs. 1 and 2. The crystal-lization temperatures were in the 44.7 to 50.7 °C range, indicating thatcrystallization of SSS occurred in theα form as a result of rapid cooling.Based on the crystallization curves and results obtained in the thermalanalysis, the 5% addition of the diacylglycerols SS, PP and OO to theSSS changed the crystallization process slightly. The addition of DAGsdecelerated the process of crystallization of SSS, leading to lower tem-peratures (onset, peak and endset) compared to pure SSS, i.e. SSS withadded DAGs required lower temperatures to attain complete crystalli-zation leading to increased crystallization enthalpies. The greatest vari-ation in crystallization enthalpy of SSS occurred upon the addition of SS,probably owing to the similarity of the fatty acid present. This indicatesan effect of the diacylglycerols on the energy released during the crys-tallization of the SSS, yet without accompanying significant changes incrystallization temperature.

On the melting curve of pure SSS, the first endothermic peak visibleat 52.1 °C (Fig. 1) represents the melting of the α polymorphic formwhile the second peak at 70.3 °C represents melting of the β form. Themelting temperatures of these polymorphic forms were consistentwith values cited in the literature (Berry, 2009). The events betweenthe two endothermic peaks reflect a series of crystallization, meltingand polymorphic transition events: transition from α to β′, crystalliza-tion in the β′ form, melting of the β′ form, transition of the β′ to βform and crystallization of the β form, during constant heating.

SSS in the presence of 5% SS and PP exhibited a higher enthalpy ofmelting of the α polymorphic form compared to pure SSS, whereasthe addition of OO led to a lower enthalpy of melting (Fig. 2). By con-trast, the addition of SS and PP led to a lower melting of enthalpy ofthe β form compared to pure SSS, whereas the addition of OO resultedin a virtually unchanged melting of enthalpy of the β form. Oh,McCurdy, Clark, and Swason (2005) also reported that the presence ofOO reduced themelting of enthalpy for theα form of SSS and led to un-changed enthalpy of the β form. Exothermic enthalpy, corresponding torecrystallization,was also affected by thepresence of the diacylglycerols.

In the presence of 5% PP, the enthalpy of recrystallization remainedpractically constant relative to pure SSS, whereas the enthalpy ofcrystallization decreased in the presence of OO and SS.

The lower endothermic peak corresponding to the α form of SSS inthe presence of OO implies that the transition of the α to the β formwas promoted by the addition of OO during constant heating. Garti,Wellner, and Sarig (1982) reported that emulsifiers containing unsatu-rated fatty acids promoted polymorphic transitions probably owing tothe angulation of the chain of unsaturated acid, which results in greatermobility of SSS, reducing the interaction between the emulsifier andSSS. The results indicate that PP and SS co-crystallize with SSS, stabiliz-ing the α form and, retarding transition to the more stable β form. Theaddition of SS resulted in greater reduction into the β form than theaddition of PP, corroborating the findings of Oh et al. (2005). Theseresults confirm that the effects of diacylglycerols in the transition fromα to β depend on the chemical structure of the diacylglycerol, such assaturation and chain length of the fatty acid. Solid diacylglycerolscontaining saturated fatty acids retarded transition from α to β of SSS,whereas the liquid diacylglycerols containing unsaturated fatty acidspromoted transition from α to β.

The crystallization and melting curves obtained by DSC for pure PPPand with added diacylglycerols are depicted in Figs. 3 and 4. The crystal-lization temperatures were in the 33.9 to 41.9 °C range, indicating thatcrystallization of PPP occurred in the α form as a result of rapid cooling.Based on the crystallization curves and results obtained on the thermalanalysis, 5% addition of the diacylglycerols SS, PP and OO produced thesame effects observed for SSS. The greatest change in crystallization en-thalpy of PPP occurred with the addition of PP, probably owing to thesimilarity of the fatty acid. This indicates that there is an effect of the di-acylglycerols on the energy released during crystallization of PPP, yetwithout accompanying significant change in crystallization temperature.

On the melting curve for PPP, the first endothermic peak visible at43.2 °C represents the melting of the α polymorphic form while thesecond peak at 63.8 °C represents melting of the ß form. The meltingtemperatures of these polymorphic forms were consistent with thosecited in the literature (Berry, 2009). The events between the two endo-thermic peaks reflect a series of crystallization, melting and polymor-phic transition events: transition from α to β′, crystallization into theβ′ form, melting of the β′ form, and transition from β′ to the β form,during constant heating.

PPP in the presence of 5% SS and PP exhibited a higher enthalpy ofmelting of the α polymorphic form compared to pure PPP, whereasthe addition of OO did not promote melting of the α form. By contrast,the addition of SS led to a lower melting of enthalpy of the β form

Fig. 4. Melting curves obtained by DSC of pure tripalmitin and with addition of 5%diacylglycerols.

Fig. 6. Melting curves obtained by DSC of pure tripalmitin and with addition of 5%diacylglycerols.

439R.C. Silva et al. / Food Research International 55 (2014) 436–444

compared to pure PPP, whereas the addition of PP increased themeltingof enthalpy and the addition of OO resulted in a virtually constantmelting of enthalpy of the β form.

Exothermic enthalpy, corresponding to recrystallization, was alsoaffected by the presence of the diacylglycerols. In the presence of 5%OO, the enthalpy of recrystallization decreased slightly compared topure PPP, whereas the enthalpy of crystallization increased in thepresence of PP and SS.

The absence of an endothermic peak corresponding to theα form ofPPP in the presence of OO implies that, akin to SSS, the transition fromthe α to β form was promoted by the addition of OO during constantheating. The results indicate that PP and SS co-crystallize with PPP,stabilizing the α form and retarding the transition to the more stableβ form, as occurred for SSS. These results confirm that the effects ofdiacylglycerols on the transition from α to β depend on the chemicalstructure of the diacylglycerol, such as saturation and chain length ofthe fatty acid. Solid diacylglycerols containing saturated fatty acids

Fig. 5. Crystallization curves obtained by DSC of pure triolein and with addition of 5%diacylglycerols.

retard the transition from α to β of PPP, whereas the liquid diacylglyc-erols containing unsaturated fatty acids promote the transition from αto β.

The crystallization and melting curves for pure OOO and withadded diacylglycerols are depicted in Figs. 5 and 6. The crystalliza-tion temperature of OOO was −53.3 °C, indicating that crystalliza-tion occurred in the α form as a result of rapid cooling. The 5%addition of the diacylglycerols OO, PP and SS induced a shift in thetemperature of crystallization of OOO to the −35.8 to −49.1 °Crange, indicating that the addition of these DAGs accelerated thecrystallization process of OOO. The greatest variation in crystalliza-tion enthalpy of OOO occurred with the addition of OO, probablyowing to the similarity of the fatty acid. Crystallization peaks werealso observed at the temperatures 37.8 and 40.2 °C, due to thepresence of the saturated diacylglycerols PP and SS, respectively.The literature states a melting point for the α form of OOO ofbetween −32 and −37 °C (Berry, 2009; Kodali, Atkinson, Redgrave,& SMALL, 1987).

On the melting curve for OOO, only a single endothermic peak isvisible at 1.3 °C, representing the melting of the β polymorphic form.On the other hand, the presence of the diacylglycerols OO, PP and SSled first to the melting of the β-prime form (peaks within the −12.2to −13.7 °C range), crystallization of the β form (exothermic peaks ataround −10 °C) and subsequent melting of the β form (peaks in the1.3 to 2.7 °C range). These melting temperatures for the β-prime andβ forms of OOO were similar to the values reported by Berry (2009)and Kodali et al. (1987).

OOO in the presence of 5%OOexhibited a higher enthalpy ofmeltingof theβpolymorphic form compared to pure OOO,whereas the additionof 5% PP and SS promoted a slightly lower enthalpy of melting of the βform.

3.2. Polarized light microscopy

During the process of crystallization of SSS, the first crystals formedat around53 °C, corresponding relatively closelywith the crystallizationtemperature obtained by DSC.

The crystallized area (%) obtained from the microscope images ofSSS was used to build the crystallization curves shown in Fig. 7. Theseresults are consistentwith those of the DSC analysis, showing a deceler-ation in the crystallization process upon addition of DAGs. Mirroringresults of the DSC analysis, the addition of a chemically similar diacyl-glycerol had a lesser effect on this deceleration.

The images of the crystalline structure during melting of pure SSSand with added diacylglycerols can be seen in Fig. 8. For the melting

Fig. 7. Crystallization curves obtained by image analysis under polarized light microscopy of pure tristearin and with addition of 5% diacylglycerols.

440 R.C. Silva et al. / Food Research International 55 (2014) 436–444

process, images in three temperature ranges are given: 50–53 °C,corresponding to crystallization into the α form; 57–60 °C, melting ofthe α form; and 68–71 °C, recrystallization into the β form. These tem-peratures correspond with the values observed on melting curves byDSC. Given that the transmission of heat in the aluminum calorimetercapsule and glass microscope slide differs, the temperatures observed

Fig. 8. Images of crystalline structures during melting

for melting of the polymorphic forms are not identical. The structureseen for the α form (Fig. 8) closely resembles that described by Oh,McCurdy, Clark, and Swanson (2002) in which the α form exhibitssmall spherulite structures in the 9–38 μm range. The heating rate of5 °C/min in the melting process for SSS was sufficiently high to causeloss of the spherulite structure of the α form and recrystallization into

of pure tristearin and with addition of 5% DAGs.

441R.C. Silva et al. / Food Research International 55 (2014) 436–444

the β-prime form was not observed. Recrystallization into the β formoccurred, exhibiting dense packing and undefined morphology, sincethere was insufficient time for structuring and growth of the spheru-lites. The voids observed result in contraction of the solid phase duringrecrystallization to the β form (Kellens, Meeusen, & Reynaers, 1992).

A concomitant result of the crystallization is contraction of the crys-tallized fat, an essential step in the production of chocolates. As formanysubstances, the solid phase of fat has a higher density than the liquidphase. However, saturated triacylglycerols, such as PPP or SSS, expandupon crystallizing, with sufficiently violent force to shatter the recipientused to hold them (Timms, 1995). Cebula, McClements, and Povey(1990) showed that the expansion of the crystallized fat results fromthe high proportion of voids caused by defects in the crystallinestructure.

The addition of DAGs affected the crystallization and the resultantmelting of SSS. The addition of SS and PP did not substantially changecrystallization in the α form whose crystals measured 22–57 μm and15–34 μm, respectively. However, an effect on melting of the α formwas observed, being stabilized by the diacylglycerols containing satu-rated fatty acids, as occurred in the calorimetric analyses, retarding tran-sition to the more stable β form. The effect of OO differed however,where crystallization became more homogenous with crystals measur-ing around 15–17 μm. In the presence of OO, the crystalline structure ofthe α form was maintained and stabilized, where polymorphic transi-tion and consequent recrystallization were not visible. The OO promot-ed the transition of the α into β form with almost no change to themicrostructure during constant heating.

The crystallized area (%) obtained from the microscope images ofPPP were used to build the crystallization curves shown in Fig. 9.These results are consistent with those of the DSC analysis, showing adeceleration in the crystallization process upon the addition of DAGs.During the process of crystallization of PPP, the formation of crystalsof the α form first took place at around 38 °C, corresponding closelywith the crystallization temperature obtained by DSC. The images ofthe crystalline structure during melting of pure PPP and with addeddiacylglycerols can be seen in Fig. 10. For the melting process, imagesin three temperature ranges are given: 35–40 °C, corresponding tocrystallization in the α form; 42–47 °C, melting of the α form; and49–54 °C, recrystallization for the β form.

The structure seen for the α form (Fig. 10) closely resembles thatdescribed by Kellens et al. (1992) in which the α form is characterizedby the bright spherulite structure. The spherulites are firmly packedtogether, resulting inwell-defined, straight edges. The size of the spher-ulites of the α form depends strongly on nucleation and can varyconsiderably between different points in the same sample. The α form

Fig. 9. Crystallization curves obtained by image analysis under polarized ligh

of PPP exhibited crystals in the 11–74 μm size range, confirming thatthe size of these spherulites can vary considerably.

The heating rate of 5 °C/min in the melting process for PPP wassufficiently low to allow recrystallization to the β form with almost nochange in microstructure. During the transition to β, small breaks andvoids were observed in the crystalline phase, as a result of the superiordensity of the β form over the α form (Kellens et al., 1992). The βspherulites exhibit greater surface rugosity than the smoother surfaceof the α spherulites.

The addition of DAGs affected the crystallization and resultant melt-ing of PPP. The addition of SS and PP slightly changed crystallization inthe α form: the crystals had a narrower range of size variation, withdimensions of 16–50 μm and 16–30 μm, respectively. There was alsoan effect on melting of the α form, being stabilized by the diacylglycer-ols containing saturated fatty acids, as occurred in the calorimetric anal-yses, retarding transition to the more stable β form. On the other hand,the effect of OO was direct. Although the crystals of the α form had asimilarly narrower size range (17–34 μm), the crystalline structurewas stabilized in the presence of OO, where polymorphic transitionand consequent recrystallization were not observed. OO promoted thetransition from theα to theβ form during constant heating, as observedin the calorimetric analyses.

The images obtained from the microscope were used to determinethe beginning and end of crystallization, induction time and crystalliza-tion time of SSS and PPP with added DAG (Table 1). These resultsshowed that the influence of DAG on the crystallization of the triacyl-glycerol is dependent on the acyl groups present in the additive.Where the acyl chain of the DAGwas similar to TAG acyl chain, the crys-tallization time decreased but there was no influence on the inductiontime (Table 1). A similar behavior was found by Foubert, Vanhoutte,and Dewettinck (2004) using milk fat with added DAGs and MAGs.Furthermore, the addition of DAGs which had different chemicalcomposition to TAGs promoted an increase in both induction andcrystallization times.

It was not possible to observe the images of the crystalline structuresof pure OOO and OOOwith added diacylglycerols, owing to the absenceof crystallization of these samples up to the temperature of 15 °C(limit of use of microscope).

3.3. X-ray diffraction

Diffractograms of TAGs with addition of DAGs are shown in Figs. 11and 12. The OOO and OOO + DAGs samples were liquid at roomtemperature. It was therefore not possible to analyze these samplesusing X-ray diffraction.

t microscopy of pure tripalmitin and with addition of 5% diacylglycerols.

Fig. 10. Images of crystalline structures during melting of pure tripalmitin and with addition of 5% DAGs.

442 R.C. Silva et al. / Food Research International 55 (2014) 436–444

SSS, SSS + SS and SSS + PP exhibited a single peak at 4.14 Å,characteristic of the α form (Table 2). When 5% OO was added to SSS,peaks were observed at 5.3 Å, 4.6 Å and 3.7 Å for the β form and at3.8 and 3.6 for the β′ form. This behavior is likely due to the fact thatOO remains in liquid state at room temperature, hampering the crystal-lization process under these conditions. Therefore, crystallizationoccurred more slowly allowing crystallization into more stable forms(β′ and β).

Unlike SSS, PPP exhibited more stable polymorphic forms evenwhen in its pure form. The addition of DAGs did not change the mor-phology of these polymorphic forms of PPP at room temperature. All

Table 1Initial and final crystallization temperatures of pure tristearin and tripalmitin withaddition of 5% diacylglycerols and induction and crystallization times.

Initialtemperature (°C)

Finaltemperature (°C)

Inductiontime (s)

Crystallizationtime (s)

SSS 53.92 50.95 36 18SSS + SS 53.84 51.53 37 14SSS + PP 53.07 48.08 41 30SSS + OO 51.92 41.15 48 60PPP 43.4 38.6 100 28PPP + SS 42.4 35.2 106 42PPP + PP 43.0 40.4 101 15PPP + OO 41.8 32.5 109 50

samples containing PPP exhibited peaks at 5.3 Å, 4.6 Å and 3.7 Å forthe β form, and at 3.8 and 3.6 for the β′ form.

4. Conclusions

The addition of 5% saturated diacylglycerols (SS, PP) delayedcrystallization (onset) of the saturated triacylglycerols (PPP and SSS).By contrast, the addition of diacylglycerols to unsaturated triacylglycerol

Fig. 11. XRD patterns at 25 °C for pure tristearin and tripalmitin with 5% DAGs.

Fig. 12. XRD patterns at 25 °C for pure tripalmitin and tripalmitin with 5% DAGs.

Table 2Polymorphic forms and short spacings at 25 °C for pure TAGS with addition of 5% DAGs.

Short spacings (Å)

Sample 5.3 4.6 4.2 4.15 3.8 3.7 Polymorphic forms

SSS – – – 4.14 (s) – – αSSS + SS – – – 4.14 (s) – – αSSS + PP – – – 4.14 (s) – – αSSS + OO 5.33 (w) 4.60 (s) 3.99 (w) – 3.85 (m) 3.68 (m) β + β′PPP 5.30 (w) 4.60 (s) 4.18 (w) – 3.85 (m) 3.71 (m) β + β′PPP + SS 5.24 (w) 4.60 (m) 4.19 (w) – 3.85 (w) 3.70 (w) β + β′PPP + PP 5.37 (w) 4.60 (m) 4.19 (w) – 3.86 (w) 3.70 (w) β + β′PPP + OO 5.42 (w) 4.60 (s) 4.18 (s) – 3.85 (m) 3.70 (m) β + β′

443R.C. Silva et al. / Food Research International 55 (2014) 436–444

(OOO) accelerated crystallization, where an extra crystallization peakappeared upon the addition of 5% SS and PP. Based on the microscopyresults, it can be concluded that the addition of unsaturated diacylglyc-erol (OO) affected the polymorphic transition of the triacylglycerolsstudied, where the results foundwere consistentwith those determinedby XRD and DSC.

Acknowledgments

The authors gratefully acknowledge the generous support of theBrazilian research funding agencies, the Conselho Nacional deDesenvolvimento Científico e Tecnológico (CNPq), the Coordenaçãode Aperfeiçoamento de Pessoal de Nível Superior (CAPES), and theFundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP).

References

AOCS (2004). American oil chemists' society Official methods and recommended practicesof the American Oil Chemists' Society (5th ed.). Champaign.

Berry, S. E. E. (2009). Triacylglycerol structure and interesterification of palmitic andstearic acid-rich fats: An overview and implications for cardiovascular disease.Nutrition Research Reviews, 22, 3–17.

Cebula, D. J., McClements, D. J., & Povey, M. J. W. (1990). Small angle neutron scatteringfrom voids in crystalline trilaurin. Journal of the American Oil Chemists' Society, 67,76–78.

Cheong, L., Zhang, H., Xu, Y., & Xu, X. (2009). Physical characterization of lard partialacylglycerols and their effects on melting and crystallization propertiesof blends with rapeseed oil. Journal of Agriculture and Food Chemistry, 57,5020–5027.

Craven, R. J., & Lencki, R. W. (2011a). Binary phase behavior of diacid 1,3-diacylglycerols.Journal of the American Oil Chemists' Society, 88, 1125–1134.

Craven, R. J., & Lencki, R. W. (2011b). Crystallization and polymorphism of1,3-acyl-palmitoyl-rac-glycerols. Journal of the American Oil Chemists' Society, 88,1113–1123.

Elisabettini, P., Desmedt, A., Gibon, V., & Durant, F. (1995). Effect of sorbitan tristearate onthe thermal and structural properties of monoacid triglycerides: Influence of a cis ortrans double bond. Fat Science Technology, 97, 65–69.

Foubert, I., Vanhoutte, B., & Dewettinck, K. (2004). Temperature and concentrationdependent effect of partial glycerides on milk fat crystallization. European Journal ofLipid Science and Technology, 106, 531–539.

Garbolino, C., Bartoccini, M., & Floter, E. (2005). The influence of emulsifiers on thecrystallization behaviour of a palm oil-based blend. European Journal of Lipid Scienceand Technology, 107, 616–625.

Garti, N., Wellner, E., & Sarig, S. (1982). Crystal structure modification of tristearin by foodemulsifiers. Journal of the American Oil Chemists' Society, 59, 181–185.

Herrera, M. L., & Marquez Rocha, F. J. (1996). Effects of sucrose ester on the kinetics ofpolymorphic transition in hydrogenated sunflower oil. Journal of the American OilChemists' Society, 73, 321–326.

Kellens, M., Meeusen, W., & Reynaers, H. (1992). Study of the polymorphism and thecrystallization kinetics of tripalmitin: A microscopic approach. Journal of theAmerican Oil Chemists' Society, 69, 906–911.

Kodali, D. R., Atkinson, D., Redgrave, E. T. G., & SMALL, D.M. (1987). Structure andpolymorphism of 18-carbon fatty acyl triacylglycerols: Effect of unsaturation andsubstitution in the 2-position. Journal of Lipid Research, 28, 403–413.

Metin, S., & Hartel, R. W. (2005). Crystallization of fats and oils. In F. Shahidi (Ed.),Bailey's industrial oil and fat products (pp. 45–76). New York: Wiley Interscience.

Oh, J. H., McCurdy, A.R., Clark, S., & Swanson, B. G. (2002). Characterization and thermalstability of polymorphic forms of synthesized tristearin. Journal of Food Science, 67,2911–2917.

Oh, J. H., McCurdy, A.R., Clark, S., & Swason, B. G. (2005). Stabilizing polymorphictransitions of tristearin using diacylglycerols and sucrose polyesters. Journal of theAmerican Oil Chemists' Society, 82, 13–19.

Ribeiro, A. P. B., Basso, R. C., Grimaldi, R., Gioielli, L. A., & Gonçalves, L. A. G. (2009).Instrumental methods for the evaluation of interesterified fats. Food AnalyticalMethods, 2, 282–302.

Rousseau, D., Marangoni, A. G., & Jeffrey, K. R. (1998). The influence of chemical in-teresterification on physicochemical properties of complex fat systems: 2. Mor-phology and polymorphism. Journal of the American Oil Chemists' Society, 75,1833–1839.

444 R.C. Silva et al. / Food Research International 55 (2014) 436–444

Sato, K. (2001). Crystallization behaviour of fats and lipids: A review. Chemical EngineeringScience, 56, 2255–2265.

Siew, W. L., & Ng, W. L. (1996). Effect of diacylglycerols on the crystallization of palmoleins. Journal of the Science of Food and Agriculture, 71, 496–500.

Siew, W. L., & Ng, W. L. (1999). Influence of diglycerides on the crystallization of palm oil.Journal of the Science of Food and Agriculture, 79, 722–726.

Smith, K. W., Bhaggan, K., Talbot, G., & van Malssen, K. F. (2011). Crystallization of fats:Influence of minor components and additives. Journal of the American Oil Chemists'Society, 88, 1085–1101.

Smith, P. R., & Povey, M. J. (1997). The effect of partial glycerides on trilaurin crystalliza-tion. Journal of the American Oil Chemists' Society, 74, 169–171.

Timms, R. E. (1995). Crystallization of fats. In R. J. Hamilton (Ed.), Developments inoils and fats (pp. 204–223). London: Chapman & Hall.

Toro-Vazquez, J. F., Rangel-Vargas, E., Dibildox-Alvarado, E., & Charó-Alonso, M.A. (2005).Crystallization of cocoa butter with and without polar lipids evaluated by rheometry,calorimetry and polarized light microscopy. European Journal of Lipid Science andTechnology, 107, 641–655.

Verstringe, A., Danthine, S., Blecker, C., Depypere, F., & Dewettinck, K. (2013). Influence ofmonopalmitin on the isothermal crystallizationmechanism of palm oil. Food ResearchInternational, 51, 344–353.

Wright, A. J., Hartel, R.W., Narine, S. S., &Marangoni, A. G. (2000). The effect of minor compo-nents onmilk fat crystallization. Journal of the American Oil Chemists' Society, 77, 463–475.

Wright, A. J., Hartel, R. W., Narine, S. S., & Marangoni, A. G. (2002). The effect of minorcomponents on milk crystallization, microstructures and rheological properties. InA. G. Marangoni, & S. S. Narine (Eds.), Physical properties of lipids (pp. 125–161).New York (USA): Marcel Dekker.