Effects of high intensity ultrasound and emulsifiers on crystallization behavior of coconut oil and palm olein

Food Research International 86 (2016) 54–63

Contents lists available at ScienceDirect

Food Research International journal homepage: www.elsevier.com/locate/foodres

Effects of high intensity ultrasound and emulsifiers on crystallization behavior of coconut oil and palm olein Jessica Mayumi Maruyama a, Ashwini Wagh b, Luiz Antonio Gioielli a, Roberta Claro da Silva a,⁎, Silvana Martini c a b c

Department of Biochemical and Pharmaceutical Technology, Faculty of Pharmaceutical Sciences, University of São Paulo, Av. Prof. Lineu Prestes n. 580, B16, CEP 05508-900, São Paulo, SP, Brazil The Clorox Service Company, 4900 Johnson Drive, Pleasanton, CA 94588, United States Department of Nutrition, Dietetics and Food Science, Utah State University, 8700 Old Main Hill, Logan, UT 84322-8700, United States

a r t i c l e

i n f o

Article history: Received 20 January 2016 Received in revised form 28 April 2016 Accepted 8 May 2016 Available online 14 May 2016 Keywords: Emulsifiers High intensity ultrasound Crystallization Coconut oil Palm olein

a b s t r a c t The objective of this research is to evaluate the crystallization behavior of coconut oil (CO) and palm olein (PO) as affected by the addition of two monoacylglycerols (MAG) emulsifiers and by the use of high intensity ultrasound (HIU). MAG with high content of palmitic, oleic, and linoleic acid (EM1) and MAG with high content of stearic acid and no unsaturated fatty acids (EM2) were used. Results show that the addition of emulsifiers did not affect crystallization kinetics of CO and similar solid fat contents (1.32 ± 0.94) (SFC) and melting enthalpies (5.90 ± 4.56) were obtained. The addition of EM1, however, significantly delayed the crystallization of PO as evidenced by a significantly lower SFC and melting enthalpy. SFC for PO was 8.56 ± 0.913 while SFC for PO + EM1 was 3.63 ± 1.38. Sonication induced the crystallization of CO samples crystallized with and without EM1 and EM2 while only induced the crystallization of PO + EM1 as measured with SFC. The induction in crystallization by HIU was also evidenced by higher enthalpy with values up to a range of 8 J/g to 11 J/g. A decrease in elasticity from 3.17 × 106 to 2.52 × 105 was observed in CO crystallized with emulsifiers which could be reverted by the application of HIU. Contrarily, the addition of emulsifiers increased elasticity of PO from 3.35 × 102 to 4.83 × 104 and sonication did not affect these values significantly. Differences observed in elasticity values are attributed not only to the amount of solid material obtained but also to the type of microstructure of the crystalline network formed during crystallization. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Emulsifiers, such as monoacylglycerols (MAG) have been used to change the crystallization properties of palm oil and dairy fat by food industry as shown by Basso et al. (2010) and Foubert, Dewettinck, Van De Walk, Dijkstra, and Quinn (2007). These studies showed that addition of MAG led to formation of greater number of seed crystals which are more stable to temperature fluctuations (Basso et al., 2010). Also, Verstringe, Danthine, Blecker, and Dewettinck (2004) reported that the addition of saturated MAG accelerates the crystallization of palm oil, enabling the use of palm olein in many industrial processes. The study of fat-emulsifier systems is of great interest since emulsifiers can be used as minor components in the production of edible fat products to provide desirable melting and crystallization properties (Marangoni & Wesdorp, 2013, Smith, Bhaggan, Talbot, & van Malssen, 2011). However, the role of emulsifiers as crystallization modifiers on natural and commercial fats is underexplored (Hasenhuettl, 2008). The main effects of these additives in the crystallization of fats occur during the stages of nucleation and growth of the crystalline material changing physical properties such as crystal size, solid fat content, and microstructure (Smith et al., ⁎ Corresponding author. E-mail address: [email protected] (R.C. Silva).

http://dx.doi.org/10.1016/j.foodres.2016.05.009 0963-9969/© 2016 Elsevier Ltd. All rights reserved.

2011). Garti (1988) and other authors (Smith et al., 2011; Martini et al. Rincón-Cardona et al., 2015; Cerdeira et al., 2003, 2005, 2006; Martini et al., 2002, 2004; Puppo et al., 2002) have described that when emulsifiers have similar chemical composition to the fat, they can be incorporated in the crystalline lattice and delay crystal growth. If the chemical composition of the emulsifier is very different to that on the fat the emulsifier can induce crystallization. However, the issue of promoting or inhibiting crystallization, however, is still debatable. High intensity ultrasound (HIU) has been used in the food industry as an efficient tool for large scale commercial applications, such as emulsification, homogenization, extraction, and viscosity alteration (Patist & Bates, 2008). In addition, previous research has shown that HIU can affect crystallization processes by affecting crystal nucleation, controlling the rate of crystal growth, promoting the formation of small and even-sized crystals, and preventing fouling of surfaces by the newly formed crystals (Kallies, Ulrich, & König, 1997; Luque de Castro & Priego-Capote, 2007; Virone, Kramer, van Rosmalen, Stoop, & Bakker, 2006). In particular, HIU has been used to change the crystallization behavior of several lipid systems. Studies have suggested that HIU affects primary or the secondary nucleation of lipids depending on different sonication parameters used such as processing time, duration, and acoustic pulse (Higaki, Ueno, Koyano & Sato, 2001; Ueno, Ristic, Higaki & Sato, 2003; Ueno, Sakata, Takeuchi & Sato, 2003; Patrick, Blindt &

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

Janssen, 2014; Martini et al., 2005, Martini, Suzuki & Hartel, 2008; Suzuki, Lee, Padilla & Martini, 2010; Ye, Wagh & Martini, 2011; Chen, Zhang, Sun, Wang & Xu, 2013; Ye, Tan, Kim & Martini, 2014; Rincón-Cardona, Agudelo-Laverde, Herrera & Martini, 2015; Ye & Martini, 2015). Previous studies have shown that HIU can be used in combination with a novel emulsifier to change the crystallization behavior of milk fat and broaden the range of physical properties achieved by this fat (Wagh, Walsh, & Martini, 2013). There is a need to explore the use of sonication with other emulsifier and fat systems to generate a wider range of physical properties in the fat and broaden their use in food products. Previous research by our group showed that the addition of MAG in different proportions to coconut oil (CO) and palm olein (PO) not only affected the type of crystals formed but also the number of crystals formed which was due to either an induction or a delay in the crystallization. The type of emulsifier used and the chemical composition of the emulsifier mainly affected the crystallization behavior under static conditions of the final fat blend with the emulsifier produced. Based on our previous research the objective of this work was to study the effects of high intensity ultrasound (HIU) on the crystallization behavior of CO and PO with added MAG to change physical properties of these systems when crystallized with agitation. 2. Materials and methods 2.1. Materials Palm olein (PO) was donated by Agropalma S/A (Pará, Brazil) and coconut oil (CO) was donated by Copra Alimentos Ltda. (Alagoas, Brazil). Emulsifiers were provided by Danisco Brasil Ltda (São Paulo, Brazil). PO and CO were stored at 0 °C prior to use. All other reagents and solvents were of analytical grade. Two emulsifiers were used in this study: EM1 — distilled monoglyceride (N90% MAGs) produced from edible vegetable oil; and EM2 — mono-and di-glyceride (N52% MAGs) produced from refined fully hydrogenated vegetable fats. Detailed chemical composition of coconut oil, palm olein, EM1 and EM2 are reported in Maruyama, Soares, D'Agostinho, Gioielli, and Silva (2014). Briefly, CO shows predominance of saturated fatty acids (83.3%), mainly lauric acid (42.6%), myristic acid (21.0%) and palmitic acid (12.2%). Unsaturated fatty acids not only are predominant in PO (57.1%), mainly oleic acid in 47.6%, but also have high amounts of palmitic acid (36.9%). EM1 consisted in approx. 60% of linoleic acid, 16% of oleic acid, and 23% of palmitic acid; while EM2 consisted of 89% of stearic acid and 10% of palmitic acid with b 1% of unsaturated fatty acids. 2.2. Mixture preparation Oils and emulsifiers were melted in the microwave (±70 °C) and mixed to obtain a 3% (wt/wt) of the two different commercial emulsifiers (MAG) in each oil. Therefore a total of 4 samples were analyzed: (a) coconut oil + 3% of EM1, (b) coconut oil + 3% of EM2, (c) palm olein + 3% of EM1, and (d) palm olein + 3% of EM2. CO and PO sample without the addition of emulsifier were used as control. Based on the previous study by Maruyama et al. (2014) a 3% concentration of emulsifier was used to obtain the greater effect in CO and PO. 2.3. Crystallization experiments Samples were heated to 80 °C and kept at this temperature for 30 min to allow complete melting of the fat. The melted lipid samples were then placed in a thermostatized crystallization cell as described elsewhere (Martini et al., 2008) which was set at different crystallization temperatures (PO, PO + EM1 and PO + EM2: 19 °C; CO, CO + EM1 and CO + EM2: 21 °C). Different crystallization experiments

55

Table 1 Sonication conditions and nomenclature used to evaluate the effect of high intensity ultrasound (HIU) on the crystallization behavior and functional properties of samples.

Palm olein (PO) PO + EM1 PO + EM2 Coconut oil (CO) CO+ EM1 CO + EM2

Stop agitation

Crystallization temperature (°C)

HIU applicationa

15 min

19 °C

15 min

21 °C

5 min 15 min 0 min 30 min 35 min 45 min

a For palm olein and the mixtures, HIU was applied to the sample when the first crystals were observed by the naked eye. For coconut oil and mixtures, HIU was applied 30 min after the first crystals were observed by the naked eye.

were performed during this research to define the temperatures that were more appropriate for each sample and that would result in more relevant changes after HIU application. Samples were crystallized at a cooling rate of 5 °C/min with agitation using a magnetic stirrer (200 rpm) to increase the heat transfer between the sample and the external circulating water. After 15 min, sample reached the set temperature (Tc) and the agitation was stopped to avoid dissolution of bubbles generated during the sonication process. Crystallization time was recorded from the moment agitation was stopped (t = 0) and were kept at Tc for 90 min. For palm olein and the mixtures, HIU was applied to the sample when the first crystals were observed by the naked eye. In the case of coconut oil and mixtures sonication did not affect crystallization under these conditions (data not shown) and therefore HIU was applied 30 min after the first crystals were observed by the naked eye. Table 1 shows the time and conditions used in this experimental design. After 90 min at Tc physical properties were measured using the techniques described below. Samples crystallized without HIU application were used as control groups. 2.4. HIU application HIU was applied using a Misonix S-3000 sonicator (Misonix Inc., Farmingdale, N.Y., U.S.A.) operating at an acoustic frequency of 20 kHz for 10 s using 50 W of electrical power. A tip of 12.7 mm diameter and amplitude of vibration of 108 μm was used. A diagram of the experimental setup including the crystallization cell, thermocouple, agitation, and sonicator configuration can be found in Martini et al. (2008). 2.5. Polarized light microscopy measurements (PLM) Crystal morphology was recorded during crystallization. A drop of lipid sample was taken from the crystallization cell at different times and placed between a slide and a cover-slide to evaluate crystals' microstructure during crystallization using a polarized light microscope (PLM, Olympus BX 41, Tokyo, Japan) with a digital camera attached. 2.6. Thermal behavior measurements A differential scanning calorimeter (DSC, DSC Q20, TA Instrument, DE) was used to evaluate the melting behavior of the crystallized material. After the crystallization experiment, in which the sample was held at Tc for 90 min, 5–15 mg of material was placed in a hermetic aluminum pan for DSC use. The sample was heated to 60 °C at 5 °C/min to evaluate the melting behavior of crystallized material. Through this procedure, the onset (Ton) and peak temperature (Tp) and enthalpy (ΔH) of the melting process was determined. 2.7. Melting point The melting point of samples was determined by AOCS official Method Cc 1-25 (2009).

56

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

2.8. Rheology measurements A TA Instruments AR-G2 Magnetic Bearing Rheometer (TA Instruments, AR-G2) was used to evaluate the viscoelastic properties of the samples. Oscillatory tests were performed by strain sweep step to obtain viscoelastic parameters such as the storage (G′) and loss (G″) modulus. The experiments were carried out using a parallel plate geometry (40 mm diameter). A constant frequency of 1 Hz (6.28 rad/s) was used, and strain values were set from 0.0008 to 10%. The viscoelastic behavior of the samples was measured after 90 min at Tc. 2.9. Solid fat content (SFC) A NMS 120 minispec NMR Analyzer (Bruker, MA, USA) was used to evaluate the solid fat content of samples. The SFC was determined by AOCS official Method CD 16b-93 (2009). 2.10. Statistical analysis Crystallization experiments were performed in duplicate. Crystal microstructure, solid fat content, viscoelasticity, and melting behavior were measured from each replicate and mean values and standard errors are reported when appropriate. Significant difference between treatments was evaluated using a two-way ANOVA and a Tukey's multiple comparison post-hoc test (p b 0.05). 3. Results and discussion 3.1. Crystallization behavior The CO used in this research had a melting point of 23.5 ± 0.4 °C. The addition of 3% EM1 slightly reduced the melting point to 22.7 ± 0.5 °C while the addition of 3% EM2 resulted in a melting point of 22.6 ± 0.3 °C. PO showed a melting point of 21.4 ± 0.8 °C. The addition of emulsifiers slightly increased the melting point with values of 22.9 ± 0.2 °C for PO + EM1 and 22.0 ± 0.5 °C for PO + EM2. Based on these results the crystallization temperatures (Tc) chosen in this research were 21 °C and 19 °C for the CO and PO samples, respectively to obtain a supercooling of approximately 2 °C. Figs. 1 and 2 show the microstructure of CO and PO, respectively and their mixtures with emulsifiers crystallized with and without the use of HIU at different time points. Pictures were captured at 10, 30, 50 and 70 min at Tc. To understand the crystallization of CO, the first vertical panel can be considered as a baseline for crystallization of CO crystallized at 21 °C (Fig. 1). On adding EM1 to CO, a delay in the crystallization of CO (third vertical panel in Fig. 1) is observed as evidenced by the presence of fewer crystals after 70 min. A similar behavior was observed for CO crystallized with EM2. The presence of fewer crystals for the same time point suggests a delay in the crystallization of CO. According to Smith et al. (2011), since the EM1 and EM2 are present at 3% of the formula, they can be considered as minor components or additives which influence crystallization. As also observed by Cerdeira and collaborators (2005) the use of emulsifiers alter properties of the fat surface and the fat crystallization process, resulting in an altered solid fat content and crystal size. Such components are deemed to influence nucleation, growth or both phenomena. The effect of additives on both nucleation and growth need not be identical. Previous work by Maruyama et al. (2014) has shown that the addition of 3% EM1 in coconut oil promoted crystal growth but did not affect the onset of crystallization. These effects could be explained by structural similarities between triacylglycerols (TAG) and MAG that allows them to partially co-crystallize (Smith et al., 2011). Since CO has a high content of saturated fatty acids (approximately 83%) (Maruyama et al., 2014) the crystallization behavior of the system is driven by the CO and no delay in crystallization is observed. Maruyama et al. (2014) showed, however that the addition

of EM2 induced the onset of crystallization in CO but did not promote crystal growth. The induction in crystallization might be due to the high percentage of saturated fatty acids present in EM2 allowing this emulsifier to act as a heteronuclei in the system. These results are somehow different from the ones observed in Fig. 1 where a delay in the onset of crystallization is observed. This difference may be a consequence of the different experimental set-ups and different analytical techniques used. Maruyama et al.'s (2014) work was performed under static conditions where no agitation was applied, while our experiments were performed with agitation. During crystallization process, agitation is also essential since it allows sufficient mobility of the nuclei to encourage crystal growth. The use of HIU in these samples slightly reduced crystal sizes in CO without emulsifier and induced crystallization as in the case of CO + EM2. Fig. 2 shows the effect of EM1 and EM2 on PO crystallization and the effect of HIU. The first vertical panel shows crystallization of PO only. The third and fifth panels indicate delay in crystallization of PO on addition of EM1 and EM2 respectively. A significant delay in the crystallization of PO is observed when both emulsifiers are used and this delay is more pronounced for EM1. For example, at 70 min a decreasing number of crystals can be observed in the following samples PO N PO + EM2 N PO + EM1. Maruyama et al. (2014) showed that EM1 and EM2 delayed the onset of crystallization of PO but promoted crystal growth. HIU slightly induced crystallization in PO (vertical panel 2) where more and smaller crystals are observed. The induction is also seen in PO + EM1 or PO + EM2, however, the effect is more noticeable in the later timepoints (70 min) than earlier timepoints when the picture was captured (10 min). In addition, HIU was more effective on PO + EM2 (vertical panel 6) compared to PO + EM1 (vertical panel 4). For the conditions used in this study the crystallization behavior of CO and PO observed by PLM was affected by EM1 and EM2 addition, both lipids showed a delay in the crystallization in presence of these emulsifiers, due to the interaction between lipid and emulsifier in the heteregeneous nucleation was harmed, thus the lipids require more time to crystallize. On the other hand, HIU appears to induce nucleation producing smaller crystals and demonstrated the ability to restore the crystalline structure lost with the emulsifier addition. This occurs because to the application of HIU leads to an increase in energy of the system. The generation of bubbles in the system induces primary nucleation and hence overall crystallization (Wagh et al., 2013). 3.2. Solid fat content (SFC) Even though PLM images can be used to monitor crystal formation, crystallization behavior can be better quantified by measuring solid fat content (SFC) values as a function of time. Figs. 3 and 4 show the solid fat content determined by NMR of sonicated and non-sonicated samples. Mean values and standard errors of triplicate crystallization runs are reported. Sonicated CO samples showed very similar SFC to non-sonicated samples for the first 50 min at Tc (Fig. 3a). However, at 60 min, a significant increase in SFC was observed in sonicated samples suggesting that HIU significantly increased the crystallization rate of CO. In addition, a significantly (p b 0.001) higher SFC was observed for the sonicated samples (14.2% ± 2.3%) compared to the non-sonicated samples (1.3% ± 0.9%) after 90 min at Tc. The addition of emulsifiers did not affect (p N 0.05) the final SFC of CO (Fig. 3b and c). Only a slight increase in SFC was observed for CO + EM2 with a value of 2.8% ± 0.9% after 90 min at Tc. Even though a slight delay in crystallization was reported in Fig. 1 with emulsifier addition this delay is not observed in the SFC values. It is important to note at this point that PLM is more sensitive than SFC measurements which have a sensitivity of approximately 1%. PLM can be used to monitor the onset of crystallization, SFC is very useful to quantify crystallization kinetics. The lack of effect of EM1 on

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

Fig. 1. Microstructure of raw material (coconut oil [CO]) and emulsifier mixutres (EM1 and EM2) with and without HIU application at different times. Samples were crystallized at 21 °C. The bar represents 100 μm.

57

58 J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

Fig. 2. Microstructure of raw material (palm olein [PO]) and emulsifier mixutres (EM1 and EM2) with and without HIU application at different times. Samples were crystallized at 19 °C. The bar represents 100 μm.

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

59

Fig. 3. Solid fat content (SFC, %) of sonicated and non-sonicated CO (A) CO + EM1 (B) and CO + EM2 (C).

Fig. 4. Solid fat content (SFC %) of sonicated and non-sonicated PO (A) PO + EM1 (B) and PO + EM2 (C).

crystallization kinetics can be explained by the chemical structure of EM1 and CO. Due to their structural similarities MAG in EM1 can cocrystallize with CO but since CO has significantly higher content of saturated fatty acids (approximately 83% of saturated fatty acids) (Maruyama et al., 2014), the crystallization behavior of the system will be driven by CO crystallization kinetics. The slight increase in SFC observed in CO + EM2 can be attributed to the high content of saturated fatty acids in EM2 (approximately 99% saturated fatty acids) which might slightly induce the crystallization of the system. Sonication significantly increased SFC in CO + EM1 (p b 0.01) and CO + EM2 (p b 0.05) with values of 11.4 ± 1.2% and 9.7 ± 1.6% for CO + EM1 and CO + EM2, respectively for the sonicated samples and 1.5 ± 0.9% and 2.8 ± 0.9% for CO + EM1 and CO + EM2, respectively for the non-sonicated samples. It is interesting to note that even though HIU generated very small changes in crystal microstructure as discussed in Fig. 1 it is clear from Fig. 3 that HIU induces the crystallization of the system evidenced by a faster crystallization rate and a higher final SFC. Fig. 4 shows the SFC of PO and mixtures with EM1 and EM2 followed by the effect of HIU on all these combinations. The SFC profile of PO is slightly different from the one observed for CO samples and described

in Fig. 3. The first difference is observed on the effect of emulsifier addition. A significant reduction (p b 0.01) in SFC is observed in PO when EM1 is used. SFC of PO without emulsifier was 8.6% ± 0.9% while SFC of PO + EM1 was 3.6% ± 1.4% after 90 min at Tc. In addition, a small increase in SFC was obtained when PO was crystallized with EM2 with a SFC of 10.8% ± 1.3%. These results support the microstructures reported in Fig. 2 where a significant delay in the crystallization was observed for PO + EM1 samples. Similar to the discussion above for CO, changes in crystallization behavior of PO can be explained by evaluating the chemical composition of the emulsifiers and PO. Contrary to CO, PO has a lower content of saturated fatty acids (approximately 43%) (Maruyama et al., 2014) and it will therefore be more affected by the addition of emulsifiers. Even though EM1 and PO can co-crystallize the dissimilarities in their fatty acid chemical composition results in a delay in the crystallization of the system (Aronhime, Sarig, & Garti, 1987; Garti, 1998; Garti, Aronhime, & Sarig, 1989). When EM2 is used, its high content of saturated fatty acids slightly contributes to the overall crystallization process resulting in a slight increase in SFC. Interestingly, the use of HIU generated significantly higher SFC (p b 0.01) (13.2% ± 1.5%) only for PO crystallized with EM1 and only a slight increase in

60

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

SFC was observed for PO without emulsifier and PO + EM2 after sonication (p N 0.05). Previous research has shown that sonication does not change the crystallization behavior in a significant manner when crystallization occurs rapidly or at high supercooling (Chen et al., 2013; Martini et al., 2008; Rincón-Cardona et al., 2015; Ye & Martini, 2015; Ye et al., 2011). It is possible that sonication is not effective at inducing crystallization in PO and PO + EM2 due to their fast crystallization and high final SFC.

3.3. Melting behavior by DSC The onset melting temperature (Ton), peak melting temperature (Tp) and melting enthalpy (ΔH) of coconut oil and palm olein, with (w/) or without (wo) emulsifier addition and with (w/) or without (wo) HIU application are shown in Table 2. The melting curves of CO and PO and their mixtures with MAG EM1 and EM2 crystallized without and with the use of HIU are shown in Figs. 5 and 6. Melting behavior of CO was characterized by 2 endothermic peaks, the peak at 28.6 °C is defined as peak 1, while a second and smaller peak is denoted as peak 2. The oils are mixtures of several species of triacylglycerols with different melting points. The component with the lowest melting point tend to melt first and represent the most unsaturated triglycerides, while components of higher melting point which represent the most saturated triglycerides will melt later. The addition of EM1 did not change the melting behavior of CO but the addition of EM2 resulted in a single melting peak. Similar results were observed by Maruyama et al. (2014). The use of HIU slightly affected the shape of the melting curves where more fractionation was observed in sonicated CO and CO + EM1 samples (note the more pronounced two melting peaks obtained in sonicated samples) and a sharper melting profile was observed in sonicated CO + EM2 samples. Endothermic peaks of CO were sharper compared to those of PO; this was due to the fact that the TAG distribution of CO was not as broad as that of PO. The melting behavior of PO samples was characterized by a single peak which shape did not change with the addition of emulsifiers. The use of HIU had very little effect on the shape of the melting peak of PO samples with and without emulsifiers. No significant differences (p N 0.05) were observed in Ton or Tp values of CO with and without the addition of emulsifier and processed with and without HIU (Table 2). Ton values ranged from 24.0 °C to 24.6 °C while Tp values ranged from 27.0 °C and 28.7 °C. No significant differences (p N 0.05) were observed for melting enthalpies of CO with and without the addition of emulsifiers with enthalpy values between 1.8 J/g and 2.4 J/g (Table 2). HIU significantly (p b 0.05) increased enthalpy values up to a range of 8 J/g to 11 J/g (Table 2). These enthalpy values correlate well with the SFC values previously discussed (Fig. 3). The effect of emulsifier addition on the melting properties of CO is somehow different from Maruyama et al. study where the addition of

Table 2 Onset melting temperature (Ton), peak melting temperature (Tp), melting enthalpy (ΔH) of coconut oil and palm olein, with (w/) or without (wo) emulsifier addition and with (w/) or without (wo) HIU application. Samples

Ton (°C)

Tp (°C)

ΔH (J/g)

CO woHIU CO w/HIU CO + EM1 woHIU CO + EM1 w/HIU CO + EM2 woHIU CO + EM2 w/HIU PO woHIU PO w/HIU PO + EM1 woHIU PO + EM1 w/HIU PO + EM2 woHIU PO + EM2 w/HIU

23.56 ± 0.47 23.97 ± 0.11 24.16 ± 0.81 23.99 ± 0.40 24.27 ± 1.04 24.62 ± 0.91 26.64 ± 0.67 27.40 ± 1.25 27.85 ± 0.45 27.86 ± 1.58 27.11 ± 0.61 26.62 ± 0.48

28.45 ± 0.17 26.87 ± 1.02 26.95 ± 1.10 27.99 ± 0.37 27.31 ± 1.38 27.37 ± 0.80 32.81b ± 0.30 32.79b ± 0.75 35.92a ± 1.29 31.94b ± 0.52 32.60b ± 0.68 31.56b ± 0.31

5.90 b ± 4.56 8.20 ª ,b ± 3.05 5.26 b ± 6.35 12.07a ± 4.49 2.68b ± 1.30 11.16a ± 2.64 7.77a ± 0.65 6.07a ± 1.64 2.33b ± 1.19 4.69ab ± 1.05 5.92a ± 1.25 6.30a ± 1.44

Fig. 5. DSC Melting profiles of CO samples crystallized at 21 °C with and without HIU application. (A) CO, (B) CO + EM1, and (C) CO + EM2.

EM1 significantly increased the melting enthalpy and the addition of EM2 significantly increased the Ton. However, it is important to note at this point that Maruyama et al. study was performed in a DSC without agitation with a small amount of sample and that the sample was crystallized at −60 °C to achieve full crystallization of sample. On the other hand, results reported in the present study crystallization were performed at higher temperatures and with agitation. The addition of emulsifiers and the use of HIU did not affect Ton values of PO (p N 0.05) with values ranging from 26.6 to 28.6 °C (Table 2). However, the addition of EM1 significantly increased (p b 0.05) Tp values from approximately 32.8 °C to 36.0 °C. However, this value decreased again to approximately 32.0 °C when the sample was crystallized with sonication (Table 2). Enthalpy values of PO were significantly reduced (p b 0.05) by the addition of EM1 but only slightly reduced by the addition of EM2 (Table 2). HIU slightly decreased enthalpy values for the PO sample (p N 0.05), slightly increase the enthalpy for PO + EM1 (p N 0.05) and did not change enthalpy values for PO + EM2 (p N 0.05). This data corroborates the SFC data shown in Fig. 4 where SFC is not affected by sonication for PO and PO + EM2 while a significant higher SFC was observed for PO + EM1. The effect of emulsifier addition

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

61

Fig. 7. Viscoelastic parameters (A) storage modulus G’; (B) loss modulus, G”; and (C) phase angle, delta of CO and its mixtures with emulsifiers (EM1 and EM2) crystallized at 21 °C with and without HIU. Mean values and standard errors are reported. Bars identified with the same letter are not significant different (p b 0.05).

Fig. 6. DSC Melting profiles of PO samples crystallized at 19 °C with and without HIU application. (A) PO, (B) PO + EM1, and (C) PO + EM2.

on melting behavior of PO is somehow different from the ones reported by Maruyama et al. (2014) since these authors reported higher Ton, Tp and enthalpy of melting when the emulsifiers were used. As previously mentioned, these differences might be due to the different experimental design and conditions used.

3.4. Rheology measurements Fig. 7 shows the viscoelastic properties of CO samples crystallized with and without HIU application after 90 min at Tc. The addition of emulsifiers to CO decreased G’ values, and this decrease becomes significant (p b 0.05) for CO + EM2 (Fig. 7A). This decrease in CO elasticity with the added emulsifier is somehow surprising since SFC for these samples are not different. Therefore, it is likely that the lower elasticity is due to differences in the crystal lattice and crystal size. As a consequence of changes in the microstructure and crystallization behavior, the HIU also affected the elasticity of the samples. The use of HIU increased G’ values and this increase is statistically significant

(p b 0.05) for CO + EM2 (Fig. 7A). Similar behavior is observed for G” and delta values (Fig. 7B and C). The increase in G’ indicates the formation of a more elastic crystalline network. Indeed, these results show that if the elasticity of the material is lost due to the addition of an emulsifier, this physical property can be recovered by using HIU, similar results effects in a elasticity was observed by Martini et al. (2008) and these authors suggest that HIU can be used as an additional processing variable to improve the crystal network of lipids to obtain a more viscous material has the potential application in lipid-based foods. It is interesting to note that even if SFC (Fig. 3) and melting enthalpy (Table 2) values obtained for CO were not affected by the addition of emulsifier, a less elastic material is obtained. This change is elasticity can be attributed to the different crystal microstructures reported in Fig. 1. Fig. 8 shows the viscoelastic properties of PO samples crystallized with and without HIU application after 90 min at Tc. Contrary to the results observed for CO, G’ values increase with the addition of emulsifiers and this effect becomes statistically significant (p b 0.05) for PO + EM2 (Fig. 8A). Surprisingly sonication did not affect elasticity values, with the exception of PO + EM2 where a significant (p b 0.05) decrease in G’ values was observed. Similar differences were observed for G” values (Fig. 8B) while no significant differences (p N 0.05) were found in delta values as a function of emulsifier addition and sonication (Fig. 8C).

62

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63

References

Fig. 8. Viscoelastic parameters (A) storage modulus, G’; (B) loss modulus, G”; and (C) phase angle, delta of PO and its mixtures with emulsifiers (EM1 and EM2) crystallized at 19 °C with and without HIU. Mean values and standard errors are reported. Bars identified with the same letter are not significant different (p b 0.05).

4. Conclusions This study showed that the use of emulsifiers can affect the crystallization properties of these lipids and the way that these alterations occur is directly dependent on the chemical composition of emulsifiers. Moreover HIU promoted crystallization increased in the nucleation in CO and PO and this effect depends on the type of emulsifier present in the system. Overall, this study shows that HIU can be used in combination with other processing tools such as the addition of emulsifiers to generate a wide range of physical properties needed for the food industry.

Acknowledgements The authors gratefully acknowledge the generous support from the Brazilian agencies, Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), Agropalma S/A, Danisco Brasil Ltda, Copra Indústria, Agência USP de Inovação (AUSPIN) and Utah State University (USU), for all support. We would like to thank Dr. Yubin Ye for his help. This paper was approved as paper # 8861.

Aronhime, J. S., Sarig, S., & Garti, N. (1987). Mechanistic considerations of polymorphic transformations of tristearin in the presence of emulsifiers. Journal of American Oil Chemistry Society, 64(4), 529–533. Basso, R. C., Ribeiro, A. P. B., Masuchi, M. H., Gioielli, L. A., Gonçalves, L. A. G., Santos, A. O., ... Grimaldi, R. (2010). Tripalmitin and monoacylglycerols as modifiers in the crystallisation of palm oil. Food Chemistry, 122, 1185–1192. Cerdeira, M., Martini, S., Candal, R. J., & Herrera, M. L. (2006). Polymorphism and growth behavior of low-trans fatty acids blends formulated with and without emulsifiers. J. Am. Oil Chem. Soc., 83, 489–496. Cerdeira, M., Martini, S., Hartel, R. W., & Herrera, M. L. (2003). Effect of sucrose ester addition on nucleation and growth behavior of milk fat–sunflower oil blends. J. Agric. Food Chem., 51, 6550–6557. Cerdeira, M., Pastore, V., Vera, L., Martini, S., Candal, R. J., & Herrera, M. L. (2005). Nucleation behavior of blended high-melting fractions of milk fat as affected by emulsifiers. Eur. J. Lipid Sci. Technol., 12, 877–885. Chen, F., Zhang, H., Sun, X., Wang, X., & Xu, X. (2013). Effects of ultrasonic parameters on the crystallization behavior of palm oil. Journal of the American Oil Chemists' Society, 90, 941–949. Foubert, I., Dewettinck, K., Van De Walk, D., Dijkstra, A. J., & Quinn, P. J. (2007). Physical properties: Structural and physical characteristics. In F. D. Gunstone, J. L. Harwood, & A. Dijkstra (Eds.), J (Eds), The Lipid Handbook (pp. 471–508) (Florida: Boca Raton). Garti, N. (1988). Effects of surfactants on crystallization and polymorphic transformation of fats and fatty acid. In N. Garti, & K. Sato (Eds.), Crystallization and Polymorphism of Fats and Fatty Acids (pp. 267–303). New York (USA): Marcel Dekker. Garti, N. (1998). Effects of surfactants on crystallization and polymorphic transformation of the fats and fatty acids. In K. Sato (Ed.), Crystallization and polymorphism of fats and fatty acids (pp. 267–303). New York: Dekker. Garti, N., Aronhime, J., & Sarig, S. (1989). The role of chain length and an emulsifier on the polymorphism of mixtures of triglycerides. Journal of the American Oil Chemists' Society, 66, 1085–1089. Hasenhuettl, G. L. (2008). Overview of Food Emulsifiers. In G. L. Hasenhuettl, & R. W. Hartel (Eds.), Food emulsifiers and their applications (pp. 1–7) (New York: Springer). Higaki, K., Ueno, A., Koyano, T., & Sato, K. (2001). Effects of ultrasonic irradiation on crystallization behavior of tripalmitoylglycerol and cocoa butter. Journal of the American Oil Chemists' Society, 78, 513–518. Kallies, B., Ulrich, J., & König, A. (1997). Production of crystal suspensions for granulation processes. Chemical Engineering Research and Design, 75, 142–146. Luque de Castro, M. D., & Priego-Capote, F. (2007). Ultrasound-assisted crystallization (sonocrystallization). Ultrasound Sonochemistry, 14, 717–724. Marangoni, A. G., & Wesdorp, L. H. (2013). Structure and properties of fat crystal network 2nd edn. Boca Raton: Taylor and Francis. Martini, M. C., Puppo, R. W., & Hartel, M. L. (2002). Herrera: Effect of sucrose esters and sunflower oil addition on crystalline microstructure of a high-melting milk fat fraction. J Food Sci., 67, 3412–3418. Martini, S., Cerdeira, M., & Herrera, M. L. (2004). Effect of sucrose esters on the crystallization behavior of bulk oil systems. J. Am. Oil Chem. Soc., 81, 209–211. Martini, Pastore, V., Vera, L. V., Martini, S., Candal, R. J., & Herrrera, M. L. (2005). Nucleation behavior of blended high-melting fractions of milk fat as affected by emulsifiers. Eur. J. Lipid Sci. Technol., 107, 877–885. Martini, S., Suzuki, A. H., & Hartel, R. W. (2008). Effect of high intensity ultrasound on crystallization behavior of anhydrous milk fat. Journal of the American Oil Chemists' Society, 85, 621–628. Maruyama, J. M., Soares, F. A. S. M., D'Agostinho, N. R., Gonçalves, M. I. A., Gioielli, L. A., & Silva, R. C. (2014). Effects of emulsifier addition on the crystallization and melting behavior of palm olein and coconut oil. Journal of Agricultural and Food Chemistry, 62, 2253–2263. Patist, A., & Bates, D. (2008). Ultrasonic innovations in the food industry: From the laboratory to commercial production. Innovative Food Science & Emerging Technologies, 9, 147–154. Patrick, M., Blindt, R., & Janssen, J. (2014). The effect of ultrasonic intensity on the crystal structure of palm oil. Ultrasonics Sonochemistry, 11, 251–255. Puppo, M. C., Martini, S., Hartel, R. W., & Herrera, M. L. (2002). Effects of sucrose esters on isothermal crystallization and rheological behavior of blends of milk-fat fraction sunflower oil. J Food Sci., 67, 3419–3426. Rincón-Cardona, J. A., Agudelo-Laverde, L. M., Herrera, M. L., & Martini, S. (2015). Effect of high intensity ultrasound on physical properties of soft stearins of high stearic high oleic sunflower oil. Journal of the American Oil Chemists' Society, 92, 473–482. 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. Suzuki, A. H., Lee, J., Padilla, S. G., & Martini, S. (2010). Altering functional properties of fats using power ultrasound. Journal of Food Science, 75, 208–214. Ueno, S., Ristic, R. I., Higaki, K., & Sato, K. (2003a). In situ studies of ultrasound-stimulated fat crystallization using synchrotron radiation. Journal of Physical Chemistry B, 107, 4927–4935. Ueno, S., Sakata, J., Takeuchi, M., & Sato, K. (2003b). Study for ultrasonic stimulation effect on promotion of crystallization of triacylglycerols. Photon factory report #20 part B (pp. 182). Verstringe, S., Danthine, S., Blecker, C., & Dewettinck, K. (2004). Influence of a commercial monoacylglycerol on the crystallization mechanism of palm oil as compared to its pure constituents. Food Research International, 62, 694–700.

J.M. Maruyama et al. / Food Research International 86 (2016) 54–63 Virone, C., Kramer, H. J. M., van Rosmalen, G. M., Stoop, A. H., & Bakker, T. W. (2006). Primary nucleation induced by ultrasonic cavitation. Journal of Crystal Growth, 294, 9–15. Wagh, A., Walsh, M. K., & Martini, S. (2013). Effect of lactose monolaurate and high intensity ultrasound on crystallization behavior of anhydrous milk fat. Journal of the American Oil Chemists' Society, 90, 977–987. Ye, Y., & Martini, S. (2015). Application of high intensity ultrasound to palm oil in a continuous system. Journal of Agricultural and Food Chemistry, 63, 319–327.

63

Ye, Y., Tan, C. Y., Kim, D., & Martini, S. (2014). Application of high intensity ultrasound to a zero-trans shortening during temperature cycling under different cooling rates. Journal of the American Oil Chemists' Society, 91, 1155–1169. Ye, Y., Wagh, A., & Martini, S. (2011). Using high intensity ultrasound as a tool to change the functional properties of interesterified soybean oil. Journal of Agricultural and Food Chemistry, 59, 10712–10722.