Starch/Sta¨rke 2012, 64, 219–228
DOI 10.1002/star.201100087
219
RESEARCH ARTICLE
Starch–lipid composites containing cinnamaldehyde Cristina Bilbao-Sa´inz1, Bor-Sen Chiou1, Adriana de Campos2, Wen-Xian Du1, Delilah F. Wood1, Artur P. Klamczynski1, Greg M. Glenn1 and William J. Orts1 1 2
Western Regional Research Center, ARS, U.S. Department of Agriculture, CA, USA Embrapa Agricultural Instrumentation, CNPDIA/EMBRAPA, Sa˜o Carlos, SP, Brazil
The formulation of a starch–lipid composite containing cinnamaldehyde as antimicrobial agent has been studied. Cinnamaldehyde was incorporated as an emulsion using acetem 90–50 K as a carrier and Tween 60 as the emulsifier. Oil in water emulsions were prepared by direct emulsification using a high shear mixer or a high pressure homogenizer (Microfluidizer). Oil in water emulsions containing cinnamaldehyde were further used to prepare starch–oil composites by mixing the emulsions with a completely gelatinized starch solution (wx starch, native corn dent starch, and high AM corn starch). Results demonstrated that in the presence of the emulsifier Tween 60, stable composites could be obtained when sufficient amount of AM was present in the sample. Finally, stable composites were tested for their biocidal activity against Listeria monocitogenes; no survivors remained after 1 day of incubation with 0.25% cinnamaldehyde or after 7 days with 0.025% cinnamaldehyde.
Received: June 6, 2011 Revised: September 12, 2011 Accepted: September 13, 2011
Keywords: Antimicrobial activity / Corn starch / Nano-emulsion / Stability / Viscoelastic properties
1
Introduction
In the food, pharmaceutical, agricultural, and cosmetic industries there is strong interest in creating new products based on starch and oil systems in which emulsions and nano-emulsions can be used as delivery systems for nonpolar functional components. These systems require microbial, chemical, and physical stability as an important quality criterion. Essential oils can be considered as an alternative to synthetic antimicrobial due to their reduced health risk and their biodegradability. Cinnamaldehyde is the main component of cinnamon oil. It is generally regarded as safe (GRAS) and is used as flavoring agent in the food industry [1] and as a fragrance ingredient in fine fragrances, shampoos, toilet soaps, and other toiletries as well as in non-cosmetic products such as household cleaners and Correspondence: Dr. Cristina Bilbao-Sa´inz, Western Regional Research Center, ARS, U.S. Department of Agriculture, 800 Buchanan St, Albany, CA 94710, USA E-mail:
[email protected] Fax: þ1-510-559-5818 Abbreviations: HACS, high AM corn starch; PdI, polydispersity Index; TSA, Trypticase Soy Agar
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
detergents [2]. In recent years, there has been an increased interest in the use of cinnamaldehyde as a natural antimicrobial agent due to its antiviral, antibacterial, and antifungal activity [3–6]. On the other hand, stable and non-separable starch– lipid solutions are difficult to obtain. The physical stability of the starch–lipid systems has been achieved by processing the mixtures in dry forms as films [7], spray-dried capsules [8, 9], or extruded composites [10] or by using modified starch [8, 11–13]. Native starch granules are not particularly good stabilizers or emulsifying agents; however, by adding hydrophobic side chains to the starch molecules, the starch molecules become amphiphilic with good emulsifying properties. The most common hydrophobically modified starches are the so called OSA-starches obtained by reaction with octenyl succinic anhydride (OSA). Fanta and Eskins [14, 15] also obtained an aqueous stable starch–lipid composition by steam jet cooking native starch with a variety of different oils and lipophilic materials. Authors explained the unusual stability of jet-cooked starch–oil dispersions by the spontaneous formation of starch films at the oil water interface during the preparative process [16, 17]. Therefore, in the present work, stable antimicrobial starch-oil composites were obtained by mixing oil in water www.starch-journal.com
220
C. Bilbao-Sa´inz et al.
Starch/Sta¨rke 2012, 64, 219–228
emulsions containing cinnamaldehyde with starch solutions gelatinized in a pressure cooker. It was evaluated the effect of droplet size and type of starch ((wx starch, native corn dent, and high AM corn starch (HACS)) on the stability and rheological properties of the final starch–oil composites as well as the antimicrobial efficacy of the composites against Listeria monocytogenes.
2
Materials and methods
2.1 Preparation of oil in water emulsions Grindsted acetem 90–50 K was kindly provided by Danisco USA Inc. (New Century, KS, USA). This product is an acetic acid ester of monoglycerides made from edible, partially hydrogenated soybean oil. Polyoxyethylene sorbitan monostearate (Tween 60) and cinnamaldehyde were purchased from Sigma–Aldrich Co. (Milwaukee, WI, USA). In the experiments described, the non-aqueous phase was 6 g of Grindsted acetem 90–50 K and/or 4 g of cinnamaldehyde. The aqueous phase was 50 g of distilled water and the surfactant 3 g of Tween 60. Two techniques were used to prepare the emulsions: The first method employed a high shear laboratory mixer (Silverston model L5M-A, East Longmeadow, MA, USA). A surfactant suspension was initially prepared by dispersing Tween 60 in distilled water with the aid of a magnetic stirrer. Afterwards the oil phase was added to the suspension. Emulsions were produced using the high shear mixer at 5000 rpm for 5 min followed by 10 more minutes of mixing at 10 000 rpm. Samples were placed in an ice bath to minimize temperature rise in the sample. The second method employed a high pressure homogenizer (M-110Y Microfluidizer1 Processor) from Microfluidics Corp. (Newton, MA, USA). Mixtures containing the aqueous and non-aqueous phase were passed five times through the microfluidizer homogenization pressure was set at 18 000 psi. The samples were kept cooled using a cooling jacket containing ice.
2.2 Preparation of starch–oil composites wx corn starch (Amioca), native dent corn starch (Melojel), and HACS (Hylon VII) were purchased from National
Starch Inc. (Bridgewater, NJ, USA). The AM content of the starches was approximately 2, 28, and 70% for wx corn, dent corn, and HACS, respectively, as per vendor specifications. Different water phase starch based preparations were made by mixing wx corn (10%), dent corn (10%), or HACS (6, 8, and 10%) in distilled water. Aqueous starch suspensions were processed using a 1 L pressure reactor (Paar Instrument Co., Moline, IL, USA) equipped with a mixer and heat controller (Model 4843). Starch samples were heated up to 1408C/min and held for 10 min before the paste was cooled to 878C. Samples were stirred continuously at 300 rpm. The paste was poured and mixed with the different oil–water emulsions and nano-emulsions using a Silverton homogenizer at 5000 rpm for 5 min. Table 1 shows the final composition in the starch–lipid composites.
2.3 Particle size distribution The volume distributions of oil droplets in the oil in water emulsions were measured right after preparation using a Malvern Zetasizer Nano ZS dynamic light scattering particle size analyzer (Malvern Instruments Ltd., Westborough, MA, USA) at 258C. The emulsions were first diluted with purified water to 1/50 of their original concentrations. The instrument used the method of photon correlation spectroscopy (PCS) to measure particle size in constant random thermal, or Brownian, motion. This motion causes the intensity of light scattered from the particles to vary with time. Large particles move slowly than small ones, so that the rate of fluctuation of the light scattered from them is also slower. PCS uses the rate of change of these light fluctuations to determine the size distribution of the particles scattering light. Mean particle diameters as ‘‘z-average’’ diameters and polydispersity index (PdI) were reported.
2.4 Flow and dynamic rheological measurements Viscosity of the pure compounds, acetem, cinnamaldehyde, and the mixture of them (6:4 acetem/cinnamaldehyde) was measured using a Peltier plate in an AR2000 rheometer (TA Instruments). The geometry used was a
Table 1. Composition of the starch-lipid composites Solution Starch (%)
Starch (%)
Starch–lipid composites Acetem (%) Cinnamaldehyde (%)
Water (%)
6 8 10
3.68 4.91 6.13
3.68 3.68 3.68
88.34 87.12 85.89
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.45 2.45 2.45
www.starch-journal.com
Starch/Sta¨rke 2012, 64, 219–228
2.5 Cryo-SEM observations Starch–oil composite samples were placed in the SEM sample holder and plunged into subcooled nitrogen (nitrogen slush) close to the freezing point of nitrogen (2108C). The frozen sample was transferred to the cryostage and then fractured, etched, and coated. All samples were viewed and photographed in a Hitachi S-4700 field emission scanning electron microscope (Hitachi, Japan) at 2 kV.
2.6 Bacterial strain and growth conditions L. monocytogenes was obtained from University of California, Berkeley (our strain designation RM2199; original designation strain F2379) isolated from cheese associated with an outbreak. Frozen culture of L. monocytogenes was streaked on Trypticase Soy Agar (TSA) and then incubated at 378C for 24 h. One isolated colony was re-streaked on TSA and then incubated at 378C for 24 h. This was followed by inoculating one isolated colony into a tube with 5 mL Trypticase Soy Broth (TSB) and incubating at 378C overnight with agitation.
prepared using either the high shear mixer at 5000 rpm for 5 min or the microfluidizer (five passes at 18 000 psi). A control starch–lipid composite was also prepared without the addition of cinnamaldehyde. Overnight bacterial culture was added at 1% v/v (0.1 mL bacterial culture þ 9.9 mL solution) to undiluted and to 10- and 100-fold dilutions of the starch–lipid composites. After the addition of bacterial culture and along 7 days of storage at room temperature, 1 mL samples were taken from those inoculated emulsions and immediately diluted in 0.1% peptone water at room temperature. For viable counts, samples were either spread plated 0.1 mL onto duplicate TSA plates or spot plated in duplicate (20 mL) onto TSA plates, if CFU decreased to 1 log CFU/g or less, plated 0.25 mL on each of four plates. Plates were incubated at 378C for 24 h.
3
Results and discussion
3.1 Droplet size distribution of the emulsions Droplet size distribution measurements were obtained for emulsions prepared at different oil phase compositions and using different emulsification techniques. Figure 1 shows the droplet size distribution profiles (by volume distribution) for emulsions prepared with 6 g acetem, 4 g cinnamaldehyde and the mixture of both (6 g acetem þ 4 g cinnamaldehyde) in 50 mL water using a high shear mixer (5 min at 5000 rpm followed by 10 min at 10 000 rpm). Table 2 summarizes the z-avg and PdI values of the different emulsions. When cinnamaldehyde was used as the non-aqueous phase, volume distributions showed a bimodal distribution with average droplet size around 150 nm. On the other hand, when instead of cinnamaldehyde, acetem was used as the oil phase, profiles exhibited a monomodal distribution with droplet sizes
18 16 Size distribution by volume (%)
plate and plate geometry (60 mm). The sample was placed on the Peltier plate. The top plate was then lowered to a gap of 0.5 mm from the Peltier plate. The temperature was programmed to increase from 4 to 308C at 18C/min. The shear rate was set at 10 s1. Rheological properties of the starch–oil composites stored at 48C were also measured using a Peltier plate in an AR2000 rheometer (TA Instruments). The geometry used was a plate and plate geometry (20 mm). The sample was placed on the Peltier plate. The top plate was then lowered to a gap of 1.0 mm from the Peltier plate. The edge of the sample was covered with a thin layer of silicone oil (Aldrich) to prevent water evaporation during measurements. The steady shear tests were performed at 48C. The shear rate was programmed to increase from 0.01 to 1000 s1. The frequency sweep tests were performed at 48C over the angular frequency range of 0.04–100 rad/s. The strain amplitude for the frequency sweep measurements was selected as 1%, which was in the linear viscoelastic region for all samples. The mechanical spectra were obtained recording elastic modulus (G0 ) and viscous modulus (G00 ) as a function of angular frequency.
221
14 12 10 8 6 4 2 0 1
2.7 Kinetics of killing To compare the effect of droplet size on the biocidal activity of cinnamaldehyde, starch–lipid composites were prepared by mixing 6% HACS solution with emulsions ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
10
100
1000
10000
size (nm)
Acetem
Cinnamaldehyde
Acet. + Cinn.
Figure 1. Size distribution by volume of emulsion droplets prepared with different non-aqueous phase composition and Tween 60 as surfactant. www.starch-journal.com
C. Bilbao-Sa´inz et al.
222
Starch/Sta¨rke 2012, 64, 219–228
Table 2. z-avg and PdI values for emulsions prepared with different non-aqueous phase compositions PdI
80.6 10.7a 152.5 24.3b 141.4 3.0b
0.249 0.005 0.162 0.070 0.373 0.044
Means with the same superscript letter are not significantly different from each other.
Size distribution by volume (%)
Acetem Cinnamaldehyde Acet. þ cinn.
z-avg (nm)
25
20
15
10
5
0 1
10
Viscosity (Pa.s)
100
1000
10000
size (nm)
1
0.1
HS mixer: 5 min at 5000 rpm
HS mixer: 20 min at 5000 rpm
HS mixer: 60 min at 5000 rpm
HP homog: 5 passes, 18000psi
Figure 3. Effect of homogenization technique on droplet size distribution of acet. þ cinn. in water emulsions.
0.01
0.001 0
5
10
15
20
25
30
35
Temperature (ºC) Acetem
Cinnamaldehyde
Acet. + Cinn.
Figure 2. Viscosity of acetem, cinnamaldehyde, and acet./cinn. (3:2) during heating. around 80 nm. The combination of both compounds showed distribution profiles with three distinctive peaks and droplet size average around 150 nm. In an attempt to explain these results the oil phase viscosity was obtained at 10 s1 shear rate (Fig. 2). Previous authors have reported that by raising the viscosity of the dispersed phase, droplet disruption, and break-up would be more difficult and hence, emulsion droplet size will increase [18, 19]. Therefore, the plots observed in Fig. 1 seem to suggest that the interfacial adsorption rate of the Tween 60 should be higher in acetem/water systems than in cinnamaldehyde/water systems resulting in smaller droplet sizes in spite of the higher viscosity and higher oil concentration of the emulsions. The addition of acetem to cinnamaldehyde increased the viscosity values of the cinnamaldehyde as well as the oil phase concentration resulting in distribution profiles with similar droplet sizes but higher polydispersity values.
Figure 3 shows the droplet size distribution of emulsions prepared with acetem and cinnamaldehyde as the non-aqueous phase using a high pressure homogenizer. The results obtained using a high shear mixer have been included for comparison. Table 3 summarizes the z-avg and PdI values obtained for the different homogenization conditions. It is observed that distribution profiles of emulsions prepared using the high pressure homogenizer showed a single peak with smaller droplet sizes (z-avg ¼ 83.4 0.5 nm) as well as lower PdI (PdI ¼ 0.126 0.015) in comparison with those found for emulsions prepared with the high shear mixer; even after 60 min of homogenization the droplet size was 144.6 3.5 nm and PdI 0.436 0.024. The two different homogenization devices used in this work operate with different emulsification systems. The high shear mixer consists of a rotor with four blades housed concentrically inside an emulsor screen with fine perforations. In this kind of rotor–stator systems, as the rotor rotates, it generates a lower pressure to draw the liquid into an out of the assembly, thereby resulting in circulation and emulsification [20]. On the other hand, in high pressure homogenizers, in the nozzle of the Microfluidizer two jets of emulsion collide with one another, the system forces the flow stream by high pressure through microchannels toward an impingement area creating a tremendous shearing action. Many authors have found
Table 3. Effect of homogenization conditions on droplet size and PdI z-avg (nm) HS mixer: 5 min at 5000 rpm HS mixer: 20 min at 5000 rpm HS mixer: 60 min at 5000 rpm HP homogenizer: five passes, 18 000 psi
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
3591 159.4 144.6 83.4
345 1.1 3.5 0.5
PdI 0.282 0.675 0.436 0.126
0.046 0.021 0.024 0.015
www.starch-journal.com
Starch/Sta¨rke 2012, 64, 219–228 that droplet disruption in high shear mixers is generally less efficient than in high pressure devices [18, 19] because, according to Stang et al. [21], the dispersing zones of rotor–stator systems usually have larger volumes than those of homogenizing nozzles. Consequently, at constant energy density and volume flow rate, the mean power density in rotor–stator systems is lower than in the nozzles of high-pressure devices.
3.2 Rheology results Starch–lipid composites were obtained by homogenizing the emulsions containing cinnamaldehyde with different gelatinized starch solutions. Consistency/viscosity of the starch–lipid composites is an important attribute from the engineering and consumer viewpoints. Therefore, the composites were rheologically characterized for designing unit operations (pumping and mixing), and ensuring product acceptability. Dent corn and HACS–lipid composites behaved as gels, whereas wx corn–lipid composite behaved as viscoelastic liquids. This is shown in Fig. 4a. Both the dent corn and HACS samples had G0 values approximately one order of magnitude higher than their G00 values. Also, these samples had G0 values that showed little frequency dependence over the experimental frequency range. These results indicated that both samples behaved as gels. In contrast, the wx corn sample had comparable G0 and G00 values that showed frequency dependence, indicating that it remained as a viscoelastic liquid. These results can be explained by the difference in AM contents for each sample. During gelatinization, AM chains leach out of the starch granules. As the sample cooled, the AM chains (and to some extent, amylopectin chains) can retrograde to form more crystalline structures. Samples with higher AM contents can then form stronger gels through this process. The HACS sample had higher AM content than the dent corn sample and consequently, had greater G0 values. In addition, the wx sample had a very low AM content, which prevented the sample from forming a gel. Previous studies also showed that potato [22], wheat [23], and sorghum starches [24] containing varying AM contents exhibited different rheological behavior. In those studies, samples containing higher AM contents had greater G0 values and showed less frequency dependence than samples containing lower AM contents, in agreement with the results in this study. The viscosity results for the different starch–lipid composites exhibited trends similar to the dynamic moduli data (Fig. 5). All samples also showed shear-thinning behavior as their viscosity values became much lower at high shear rate. This shear-thinning can be attributed to the disruption of interactions involving AM and amylopectin chains during shearing. Figure 5a shows that the HACS–lipid composite had the highest viscosity, whereas the wx corn–lipid ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
223 composite had the lowest viscosity. At low shear rate, the dent corn and HACS samples had viscosity values approximately three times those of the wx corn sample. Since the HACS-lipid composite had the largest moduli values, we examined the effects of HACS concentration on rheological properties of HACS-lipid samples. All the samples behaved as gels since their G0 values had little dependence on frequency and were greater than their G00 values. This is shown in Fig. 4b. The samples with higher starch concentrations were stronger gels because they had greater G0 values than those with lower starch concentrations. In fact, an increase in starch concentration from 6 to 10% resulted in samples having an order of magnitude increase in their G0 values. The viscosity data (Fig. 5b) showed similar trends as the 10% sample had higher viscosity values than the 6% sample. These results indicated that the rheological properties of HACS–lipid composites can be widely varied by changing only a small amount of starch. The addition of Tween 60 to HACS–lipid composites reduced their moduli and viscosity values by more than one order of magnitude, as shown in Figs. 4c and 5c. This might be due to an increased formation of lipid–AM complexes in the presence of Tween. These complexes inhibited retrogradation of AM, leading to weaker materials. In a previous study [25] the authors compared the dynamic moduli values of gelatinized corn and gelatinized defatted corn gels. They found that the defatted sample had G0 values that were more than one order of magnitude greater than the non-defatted sample. The authors attributed this behavior to AM–lipid complexes formed in the non-defatted sample that inhibited AM retrogradation. The results in Figs. 4c and 5c are also consistent with SEM micrographs shown in Fig. 6, which indicated that in samples without Tween, the lipid phase did not seem to interact much with the starch phase. In contrast, samples containing Tween showed more interactions between the lipid and starch phases (Fig. 6d). The use of different homogenization techniques for producing lipid–water emulsions had little effect on rheological properties of the final HACS–lipid composites. The high shear and high pressure homogenized samples had comparable G0 and viscosity values, as shown in Figs. 4d and 5d. The two samples had very similar droplet size (141 and 83 nm for high shear and high pressure homogenized samples, respectively), besides both HACS–lipid samples had the same concentration of lipids, which might have a more significant effect on rheological properties than droplet size distribution.
3.3 Stability of starch–lipid composites For stability studies, starch-lipid composites were stored for three months at 4 and 228C. Creaming of composite www.starch-journal.com
C. Bilbao-Sa´inz et al.
224
a)
Starch/Sta¨rke 2012, 64, 219–228
10000
10000
1000
1000
100
100
G'
G'' 10
10
1
1
0.1 0.01
0.1
1
10
0.1 0.01
100
0.1
1
ω (rad/s) Waxy corn starch
b)
G'
Native corn starch
HACS
Waxy corn starch
10000
10000
1000
1000
G''
100
10
100
Native corn starch
HACS
100
10
1 0.01
0.1
1
10
1 0.01
100
0.1
1
ω (rad/s) 6% Starch
c)
10
ω (rad/s)
8% Starch
10
100
ω (rad/s) 10% Starch
6% Starch
100000
10000
10000
1000
8% Starch
10% Starch
1
10
1000 100 G'
G''
100
10 10 1
1 0.1 0.01
0.1
1
10
0.1 0.01
100
0.1
ω (rad/s) w/o Tween 60
d)
100
ω (rad/s)
with Tween 60
w/o Tween 60
10000
10000
1000
1000
100
with Tween 60
100
G'
G'' 10
10
1
1
0.1 0.01
0.1
1
10
100
0.1 0.01
ω (rad/s) HS mixer
HP Homog.
0.1
1
10
100
ω (rad/s) HS mixer
HP Homog.
Figure 4. Effect on the moduli values of starch–lipid composites of (a) type of starch, (b) HACS concentration, (c) presence or absence of emulsifier, (d) emulsification technique. samples was monitored visually. Stable starch–lipid composites were obtained when 10% gelatinized HACS solutions were mixed with acetem and/or cinnamaldehyde. At this high AM concentration the lipophilic components did not separate or coalesce. The stability was explained by ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
the fact that gelatinized HACS increased viscosity of the continuous water phase which minimizes droplets mobility and decreases collision frequency allowing the lipophilic droplets to be embedded in the gelatinized starch matrix. At the lower starch concentration of 6%, macroscopic www.starch-journal.com
Starch/Sta¨rke 2012, 64, 219–228
a)
225
b)
10000
10000 1000 Viscosity (Pa.s)
Viscosity (Pa.s)
1000 100 10
100 10 1
1 0.1 0.01
0.10
1.00
10.00
100.00
0.1 0.01
1000.00
0.10
1.00
Waxy corn
c)
Native dent corn
10.00
100.00
1000.00
Shear rate (1/s)
Shear rate (1/s) 6% Starch
HACS
d)
1000000
8% Starch
10% Starch
10000
100000 Viscosity (Pa.s)
Viscosity (Pa.s)
1000 10000 1000 100 10
10 1
1 0.1 0.01
100
0.10
1.00
10.00
100.00
1000.00
Shear rate (1/s) with Tween 60
w/o Tween 60
0.1 0.01
0.10
1.00
10.00
100.00
1000.00
Shear rate (1/s) HP homog.
HS mixer
Figure 5. Effect on the viscosity values of starch–lipid composites of (a) type of starch, (b) HACS concentration, (c) presence or absence of emulsifier, (d) emulsification technique.
phase separation was observed immediately after preparation as a result of rapid coalescence. Thus, since starch by itself proved to be a poor emulsifier agent, Tween 60 was added as emulsifier in the next experiments. No phase separation was observed for any of the starch–lipid composites when emulsions containing Tween 60 were mixed with gelatinized HACS or native corn starch. This could be attributed to the formation of complexes between the AM and acetem [26]. Fanta et al. [27, 28] observed the formation of helical inclusion complexes when aqueous mixtures of high-AM starch and FFA are processed by steam jet cooking at 1408C. These AM– lipid complexes are less soluble, more hydrophobic, and possibly more surface active than AM itself, allowing them to accumulate rapidly at the oil–water interface [17]. In contrast, composites containing wx corn starch showed phase separation immediately after preparation. Ye et al. [29] also observed that addition of AP to corn oil emulsions made with hydrolyzed whey protein and lecithin resulted in flocculation of oil droplets and subsequent coalescence during storage. In the same way, Chanamai and McClemments [13] observed creaming in oil in water emulsions stabilized with Tween 20 when AP content ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
exceeded a critical concentration. The instability of the emulsions was attributed to depletion interactions. The origin of these interactions is the exclusion of colloidal particles from a narrow region surrounding the oil droplets, the depletion zone. The difference in colloidal concentration in the depletion zone and surrounding continuous phase cause an osmotic potential difference that favors the diffusion of solvent molecules from the depletion zone into the surrounding continuous phase, till droplets aggregate and thereby eliminate the osmotic potential difference [30]. Finally, it is worth mentioning that stable starch–lipid composites were obtained when starch solutions were totally gelatinized using the pressure reactor in a separate step and then blended with the oil in water emulsion under high-shear conditions but not when aqueous mixtures of starch and oil in water emulsions were processed together in the pressure reactor.
3.4 Microscopic observations Figure 6 shows cryo-SEM micrographs of the different starch–lipid composites. An SEM picture of the starting material, that is gelatinized HACS is shown in Fig. 6a. www.starch-journal.com
226
C. Bilbao-Sa´inz et al.
Starch/Sta¨rke 2012, 64, 219–228
Figure 6. Cryo-SEM micrographs of (A) gelatinized HACS, (B) gelatinized HACS and Tween 60, (C) gelatinized HACS mixed with acetem and cinnamaldehyde, (D) gelatinized HACS mixed with oil/water emulsion containing cinnamaldehyde and Tween 60 as surfactant.
When these granules were heated under mixing at 1408C in the pressure reactor, it was obtained an initial product consisted of an open network structure of mainly strands. The starch granules appeared totally disintegrated and no granular fragments could be identified. Previous works [31, 32] also reported that no residual starch granule fragments were present in starch dispersions cooked under these high temperature and shear conditions. When Tween 60 was added to the starch solution, the interwoven strands network was not observed (Fig. 6b). Richardson and coworkers [33] observed that the AM aggregated in thick strands or random aggregations when AM dispersions were heated with emulsifiers. When gelatinized HACS solutions were mixed with acetem and cinnamaldehyde, in the absence of Tween 60, the AM formed a continuous bulk phase where the oil droplets were embedded in the AM matrix due to the dense packing (Fig. 6c). The AM network did not seem to be completely attached to the oil globules suggesting that there were not enough AM–lipid complexes to form a connected network. On the contrary, in the presence of emulsifier, AM chains form a network which connects the ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
oil droplets (Fig. 6d). In this case, the homogenizer was capable of forming small droplets due to the ability of the emulsifiers to facilitate droplet formation during homogenization as well as to prevent droplet coalescence both during and after homogenization [34]. As the size of the oil droplet decreases, the starch–lipid interface increases facilitating the formation of AM–lipid inclusion complexes.
3.5 Bactericidal activity The bactericidal effect of starch–lipid composites containing cinnamaldehyde was tested against Listeria monocytogenes. No survivor was found after 1 day of incubation for undiluted and 10-fold diluted samples(0.24%). At the lowest concentration of 0.024%, a reduction in the bacterial count was observed along 7 days of storage (Fig. 7). The control starch–lipid composite without added cinnamaldehyde did not show any reduction in the number of CFU during this time. Activity of cinnamaldehyde has been previously reported against L. monocytogenes [35–37]. In previous experiments investigating the mechanism of bactericidal action of cinnamaldehyde it was observed that www.starch-journal.com
Listeria monocytogenes (Log CFU/mL)
Starch/Sta¨rke 2012, 64, 219–228
227
8 7 6 5 z-avg=83 nm
4
z-avg=3591 nm
3 2
The authors have declared no conflict of interest.
1 0 0
1
2
3
6
Time (day)
Figure 7. Effect of starch–lipid composites containing 0.024% cinnamaldehyde on the survival of L. monocytogenes.
treatment of Listeria monocytogenes with cinnamaldehyde inhibited generation of adenosine triphosphate from glucose [36]. To compare the effect of the emulsion droplet size on antimicrobial properties, starch-lipid composites with different droplet sizes (83 and 3500 nm) were tested against L. monocytogenes. The composite with smaller droplet size was more efficient in killing the bacterial cells. At 0.024% concentration, cinnamaldehyde reduced the number of CFU by more than 2 logs and 4 logs after 3 days and 6 days incubation, respectively. In comparison, the composites with larger droplet size reduced the number of CFU by 1 log and 3 logs after 3 and 6 days incubation, respectively. The effectiveness of the nano-emulsion was attributed to the higher surface area of the lipid droplets containing cinnamaldehyde in contact with the microorganisms.
4
lipid complexes that inhibited AM retrogradation. Finally, the antibacterial activity of the starch–lipid composites containing cinnamaldehyde was proved against L. monocytogenes. The effectiveness of the composites was affected by the emulsion droplet size; the composite with the smallest lipid droplet size was more efficient in killing the bacterial cells.
Conclusions
Cinnamaldehyde was used as antimicrobial agent in starch–lipid composites. An initial product was obtained consisting of oil in water emulsion containing cinnamaldehyde and acetem 90–50 K as a carrier. Emulsions were further mixed with starch solutions to prepare starch–lipid composites. It has been found that stable emulsions could be obtained when totally gelatinized native corn starch or HACS were mixed with emulsions containing Tween 60 as emulsifier. Cryo-SEM micrographs of stable composites showed a network where the lipid phase is connected to the starch phase. The type of starch affected the rheological properties of the samples since higher AM content resulted in stronger gels. On the other hand the presence of the emulsifier Tween 60 in the composition decreased the moduli values probably due to the formation of AM– ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
5 References [1] Fenaroli, G., in: George A. Burdock Group (Eds.), Fenaroli’s Handbook of Flavor Ingredients, CRC Press, Boca Raton, Orlando, Florida 2009. [2] Cocchiara, J., Letizia, C. S., Lapczynski, A., Api, A. M., Fragrance material review on cinnamaldehyde. Food Chem. Toxicol. 2005, 43, 867–923. [3] Singh, G., Maurya, S., de Lampasona, M. P., Catalan, C. A. N., A comparison of chemical, antioxidant and antimicrobial studies of cinnamon leaf and bark volatile oils, oleoresins and their constituents. Food Chem. Toxicol. 2007, 45, 1650– 1661. [4] Burt, S., Essential oils: their antibacterial properties and potential applications in foods – a review. Int. J. Food Microbiol. 2004, 94, 223–253. [5] Roller, S., Seedhar, P., Carvacrol and cinnamic acid inhibit microbial growth in fresh-cut melon and kiwifruit at 48C and 88C. Lett. Appl. Microbiol. 2002, 35, 390–394. [6] Weissinger, W. R., McWatters, K. H., Beuchat, L. R., Evaluation of volatile chemical treatments for lethality to Salmonella on alfalfa seeds and sprouts. J. Food Protect. 2001, 64, 442–450. [7] Yilmaz, G., Jongboom, R. O. J., van Soest, J. J. G., Feil, H., Effect of glycerol on the morphology of starch-sunflower oil composites. Carbohydr. Polym. 1998, 38, 33–39. [8] Varona, S., Martı´n, A., Cocero, M. J., Formulation of a natural biocide based on lavandin essential oil by emulsification using modified starches. Chem. Eng. Proc. 2009, 48, 1121–1128. [9] Baranauskiene, R., Bylaite, E., Zukauskaite, J., Venskutonis, R. P., Flavor retention of peppermint [Mentha piperita] essential oil spray-dried in modified starches during encapsulation and storage. J. Agric. Food Chem. 2007, 55, 3027–3036. [10] Bhatnagar, S., Hanna, M. A., Amylose-lipid complex formation during single-screw extrusion of various corn starches. Cereal Chem. 1994, 71, 582–587. [11] Yusoff, A., Murray, B. S., Modified starch granules as particlestabilizers of oil-in-water emulsions. Food Hydrocolloids 2011, 25, 42–55. [12] Tesch, S., Gerhards, C., Schubert, H., Stabilization of emulsions by OSA starches. J. Food Eng. 2002, 54, 167–174. [13] Chanamai, R., McClements, D. J., Depletion flocculation of beverage emulsions by gum arabic and modified starch. J. Food Sci. 2001, 66, 457–463. [14] Fanta, G. F., Eskins, K., Stable starch-lipid compositions prepared by steam jet cooking. Carbohydr. Polym. 1995, 28, 171–175. [15] Eskins, K., Fanta, G. F., ‘‘Non-Separable Starch-Oil Compositions’’, U.S. Patent 5676994, 1997.
www.starch-journal.com
228
C. Bilbao-Sa´inz et al.
[16] Fanta, G. F., Felker, P. C., Eskins, K., Baker, F. L., Aqueous starch -oil dispersions prepared by steam jet cooking. Starch films at the oil-water interface. Carbohydr. Polym. 1999, 39, 25–35. [17] Fanta, G. F., Felker, F. C., Shogren, R. L., Knutson, C. A., Starch-parffin wax compositions prepared by steam jet cooking. Examination of starch adsorbed at the paraffin-water interface. Carbohydr. Polym. 2001, 46, 29–38. [18] Qian, C., McClemments, D. J., Formation of nanoemulsion stabilized by model food-grade emulsifiers using high pressure homogenization: Factors affecting particle size. Food Hydrocolloids 2010, 25, 1000–1008. [19] Jafari, S. M., Assadpoor, E., He, Y., Bhandari, B., Recoalescence of emulsion droplets during high-energy emulsification. Food Hydrocolloids 2008, 22, 1191–1202. [20] Maa, Y. F., Hsu, C., Liquid-liquid emulsification by rotor-stator homogenization. J. Controlled Release 1996, 38, 219–228. [21] Stang, M., Schuchmann, H., Schubert, H., Emulsification in high pressure homogenizers. Eng. Life Sci. 2001, 1, 151–157. [22] Ortega-Ojeda, F. E., Larsson, H., Eliasson, A.-C., Gel formation in mixtures of high amylopectin potato starch and potato starch. Carbohydr. Polym. 2004, 56, 505–514. [23] Sasaki, T., Yasui, T., Matsuki, J., Satake, T., Rheological properties of mixed gels using waxy and non-waxy wheat starch. Starch/Sta¨rke 2002, 54, 410–414. [24] Sang, Y., Bean, S., Seib, P. A., Pedersen, J., Shi, Y.-C., Structure and functional properties of sorghum starches differing in amylose content. J. Agric. Food Chem. 2008, 56, 6680–6685. [25] Byars, J. A., Fanta, G. F., Felker, F. C., The influence of oil on the properties of slowly-cooled jet-cooked normal corn starch dispersions. Carbohydr. Polym. 2006, 63, 316–322. [26] Eliasson, A.-C., Krog, N., Physical properties of amylosemonoglyceride complexes. J. Cereal Sci. 1985, 3, 239–248. [27] Fanta, G. F., Felker, F. C., Byars, J. A., Kenar, J. A., Shogren, R. L., Starch-soybean oil composites with high oil:starch
ß 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Starch/Sta¨rke 2012, 64, 219–228 ratios prepared by steam jet cooking. Starch/Sta¨rke 2009, 61, 590–600. [28] Fanta, G. P., Shogren, R. L., Salch, J. H., Steam jet cooking of high-amylose starch-fatty acid mixtures. An investigation of complex formation. Carbohydr. Polym. 1999, 38, 1–6. [29] Ye, A., Hemar, Y., Singh, H., Flocculation and coalescence of droplets in oil-in-water emulsions formed with highly hydrolysed whey proteins as influenced by starch. Colloid Surf., B 2004, 38, 1–9. [30] McClements, D. J., Food Emulsions, Principles, Practice, and Techniques, CRC Press Publisher, Boca Raton, FL 1999. [31] Glenn, G. M., Klamczynski, A., Chiou, B.-S., Orts, W. J. et al., Temperature related structural changes in wheat and corn starch granules and their effects on gels and dry foam. Starch/Sta¨rke 2008, 60, 476–484. [32] Glenn, G. M., Irving, D. W., Starch-based microcellular foams. Cereal Chem. 1995, 72, 155–161. [33] Richardson, G., Kidman, S., Langton, M., Hermansson, A. M., Differences in amylose aggregation and starch gel formation with emulsifiers. Carbohydr. Polym. 2004, 58, 7–13. [34] McClements, D. J., Food Emulsions: Principles, Practice, and Techniques, 2nd Edn., CRC Press, Boca Raton, FL 2005. [35] Gill, A. O., Holley, R. A., Mechanisms of bactericidal action of cinnamaldehyde against Listeria monocytogenes and of eugenol against L. monocytogenes and Lactobacillus sakei. Appl. Environ. Microbiol. 2004, 70, 5750–5755. [36] Gill, A. O., Holley, R. A., Disruption of Escherichia coli, Listeria monocytogenes and Lactobacillus sakei cellular membranes by plant oil aromatics. Int. J. Food Microbiol. 2006, 108, 1 –9. [37] Du, W.-X., Olsen, C. W., Avena-Bustillos, R. J., McHugh, T. H. et al., Effects of allspice, cinnamon, and clove bud essential oils in edible apple films on physical properties and antimicrobial activities. J. Food Sci. 2009, 74, M372– M378.
www.starch-journal.com
