Method development and analysis of textural properties of quinoa starch stabilized dried emulsion

H. A. Rashid and M. Sjöö*

Speximo AB, Medicon Village, 223 81 Lund, Sweden
Department of Food Technology, Engineering and Nutrition, Lund University, PO Box 124, 221 00 Lund, Sweden
*Corresponding author:


Textural properties are important parameters for proper handling, transportation, and storage within the industry of emulsion-based powders. However, there is not a suitable method to determine that parameter. In this paper, a textural property analysis method has been developed. A TVT 6700 texture analyzer (Perten Instruments AB, Sweden) was used as the testing instrument for the method development and analysis. Reproducible data have been obtained with the method. The method was validated with two types of dried-emulsion powders (orange flavor and prickly pear oil) with very small standard deviation of 0.10 and 0.15 respectively. Native and OSA-modified quinoa starch stabilized emulsions were heat treated at either 60°C or 75°C and then freeze dried. Freeze-dried emulsion cakes were analyzed using the developed method. Samples were compressed under three different forces 4,500 g, 5,500 g, and 6,500 g. There was no significant (0.05% level) difference in their physical strength for 4,500 g force compression. At 5500 g force compression, the 60°C-treated dried emulsion stabilized with OSA-modified starch showed significant (0.05% level) weakness. In higher force (6,500 g) compression, 60°C heat-treated dried emulsions of both types released higher amount of oil than 75°C heat-treated dried emulsions. However, there was no significant difference in oil released. Later, powder made by diluting and drying the cake (1:2, 1:4, and 1:6 cake to water ratio) was analyzed with the texture analyzer. The textural property was similar for cake and 1:2 diluted powder. The other two diluted powders were considerably (0.05% level) weaker. A negative correlation (r= -.576) was found between the amount of oil released and the particle size diameter of rehydrated emulsions.


Fats and oils are important ingredients in human and animal diets for both nutritional and sensorial points of view (Saldaña & Martínez-Monteagudo, 2013). Not only are they important for nutritional and caloric values, fats and oils are also important for their deliciousness. Furthermore, they are used as carrier materials of fat-soluble vitamins (A, D, E, and K), color, flavor, and bioactive compounds. They are also essential as raw material for food, pharmaceuticals, and cosmetic products (Saldaña & Martínez-Monteagudo, 2013; Solís-Fuentes et al., 2010). Fats and oils are very prone to oxidation, which is accelerated by light, heat, and oxygen. An off flavor and toxic compound are the end products of lipid oxidation. Thus, attempts to supplement foods with lipid and lipid-soluble compounds may be useless because of their extreme vulnerability to oxidation (Aberkane et al., 2013). A system developed to overcome these problems can be as emulsion (Mao & Miao, 2015). Emulsions are of two types: oil in water (O/W) or water in oil (W/O) (Rayner et al., 2014). Emulsions are quite unstable and can be destroyed as a result of phase separation due to coalescence of droplets, creaming, and Ostwald ripening (Mao & Miao, 2015). Emulsions are mostly stabilized by surfactants or hydrocolloids. However, among all stabilizers, emulsion stabilization using particles (Pickering emulsions) have been of substantial and increasing research interest over the last decade, as they showed long-term stability against coalescence and Ostwald ripening compared to surfactants (Rayner, Sjöö et al., 2012). Particles often used to stabilize emulsions stated in the literature are: fat crystals, globular proteins, silica, latex, clay, and hydrocolloids (Timgren et al., 2013). For example, chemically modified starch granules are suitable as Pickering emulsion stabilizers. Among all the starch sources, quinoa starch is a preeminent one because of its relatively small granular size (0.5–3 µm in diameter), gluten-free properties, and superior emulsion-stabilizing capacity (mg starch per mL oil basis) (Rayner, Timgren et al., 2012). Furthermore, earlier studies showed that quinoa starch granules stabilized emulsion remains stable over a two-year storage period without any phase separation or droplet size changes (Timgren et al., 2013).

Pickering emulsion droplets can be regulated by the careful application of heat in the system by in situ gelatinization of starch granules that are absorbed at the oil-water interface. The gelatinized starch is then further structured by recrystallization after subsequent storage. This phenomenon can be applied for the encapsulation of oil and other components inside emulsion droplets. Sjöö et al. (2015) stated that the thickness of the barrier layer at the interface depends on the level of gelatinization of starch. The degree of gelatinization of starch increased with increasing temperature and time and an optimal degree of starch gelatinization for the barrier function was found.

Dehydration of emulsions facilitates the use of encapsulated ingredients in a multidimensional way in food, pharmaceuticals, and the cosmetic industry by means of increasing shelf life. Freeze drying is a common and convenient way of drying compared to other methods. This is because there is less damage to the sensitive structures of heat-sensitive food materials and other biological products (Marefati et al., 2013). Additionally, there is huge demand for dry ingredients in the food, pharmaceutical, and cosmetic industries. Dry ingredients facilitate storage and transportation by reducing water as well as increasing shelf-life (Cano-Higuita et al., 2015, Gharsallaoui et al., 2007). Currently, more than 50% of industrial raw materials are powders. The increased use of powder in the industry with increasing variety and quantity make it necessary to gather more information about the handling and processing features of powders. Powder materials have a variety of diverse physical properties (de Freitas & da Silva, 2007). Powder properties measurement is essential because it affects its performance throughout storage, packing, handling, transportation, and formulation (Fitzpatrick et al., 2004), which can reduce costs. Powder compressibility (compaction) tests have drawn researchers’ attention because they can explain the attrition tendency, flow ability, resistance to tension measurement, and agglomeration resistance (de Freitas & da Silva, 2007). Therefore, the aim of this study was to develop a method for textural properties analysis and investigate textural properties of dried emulsions prepared by heat treatment and freeze drying, using physically and chemically modified quinoa starch respectively, for stabilization.

Material and Methods


Physically modified and octenyl succinic anhydride (OSA) modified quinoa starches (OSA) were used as emulsion stabilizers. Native quinoa starch granules (NS) were isolated from quinoa grains (Biofood, Sweden) by dry milling and washing, and further treated by heating in a convection oven at 120°C for 150 minutes. OSA (2.9%) modified starch (EQMULSETM) was obtained from Speximo AB. Distilled water was used as a continuous phase and rapeseed oil as a dispersed phase. Rapeseed oil was purchased from a local supermarket (Willys AB). Two freeze-dried emulsion powders (with encapsulated orange flavor and prickly pear oil, respectively) were collected from Speximo AB.

Emulsions preparation

Emulsions were prepared in glass beakers (DURAN, GmbH, Wertheim/Main Germany). Oil (13%, v/v) and 83% distilled water were used as dispersed and continuous phases respectively. Starch was used as an emulsion stabilizer in a concentration of 300 mg/mL oil. The procedure in brief was: starch was dispersed into water and mixed with an Ystral mixture (D-79282, Ballrechten-Dottingen, Germany) at 16,000 rpm speed for 1 minute. After dispersion, the beaker was set up with an overhead stirrer and stirred with a propeller at 700 rpm. Oil was added and continuously stirred for 5 more minutes. Then, emulsions were homogenized by the Ystral mixture at 22,000 rpm for 1 minute.

Heat treatment of emulsions (Encapsulation)

To improve the barrier properties of in situ gelatinized starch at the oil droplet surface, emulsions were heat treated at 60°C and 75°C, respectively, for 10 min. During heat treatment, emulsions were continuously stirred at a speed of 500 rpm with an overhead stirrer for uniform heat distribution and to obtain individual encapsulated oil droplets. After heat treatment, emulsions were cooled to room temperature (∼20°C) while continuously stirred.


Heat-treated emulsions were freeze dried using a laboratory freeze dryer (CD 12, Hetosicc, Denmark) for 4 days in the following conditions: drying chamber temperature 20°C, cooling unit temperature -50°C, and under 10–2 mbar vacuum pressure. Punctured aluminum foil covers were placed over the sample trays during freeze drying.

Characterization of rehydrated emulsions

Dried emulsions were rehydrated with distilled water by vortex mixing at 600 rpm for 2 minutes. A laser light scattering (Mastersizer 2000, Ver.5.60, Malvern, UK) was used to analyze emulsion droplet size distribution. The pump velocity was set to 2,000 rpm. The refractive indexes (RI) were set to 1.54 and 1.33 for the starch particles in the oil water interface and for the continuous phase (water) respectively. Furthermore, samples were observed under a light microscope (Olympus BX50, Japan). For the microscopic analysis, samples were diluted 30 times with distilled water. A few drops of samples were spread on a microscope slide and no glass cover was used. Images were taken with 10X, 20X, and 50X objects magnification using a camera (DFK 41AF02, Image source, Germany) connected to the microscope. Images were processed by the software ImageJ (NIH, Version 1.42m).

Textural properties analysis

The strength of dried emulsions was analyzed through a compression test with a slight modification of the method used by de Freitas and da Silva (2007). A TVT 6700 texture analyzer (Perten Instruments AB, Sweden) was used for the analysis. Single cycle compression tests were performed using a P-CY 18R probe. Weight as well as height was calibrated using 2 kg standard weight and a P-CY 18R probe over the empty platform of the instrument. Dried emulsion cakes were cut with a sharp pipe shape cutter (17 mm diameter) as shown in Figure 1 (a, b). Filter paper, approximately 21 mm in diameter, was used to measure released oil prior to compression Figure 1 (c). Filter paper was weighed and placed in the center of the instrument platform. The sample was placed over the filter paper center. The sample was maintained in the center of the probe for equal force distribution. Samples were compressed using three forces: 4,500 g, 5,500 g, and 6,500 g. After compression, the filter papers were reweighed. The amounts of oil released were reported as the weight difference of the filter paper before and after compression. Powder samples were analyzed using an additional sample holder (20 mm in diameter and 15 mm in depth) Figure 1 (d) with a P-CY 20S probe for a specific force (6,500 g). The overall parameters of TVT texture analyzer were set as shown in Table 1.

Figure 1: (a) Pipe shape sample cutter, (b) round shaped cut sample, (c) filter paper, and (d) powder sample holder.

Table 1: Set values for compression test
Single Cycle Compression


Sample height (mm) 7.0 15.0
Starting distance from sample (mm) 5.0 5.0
Initial speed (mm/s) 1.0 1.0
Test speed (mm/s) 1.0 1.0
Retract speed (mm/s) 10.0 10.0
Trigger force (g) 50. 5.0
Data rate (pps) 500 500
Probe diameter (mm) 180. 20.0

Statistical analysis

Data were analyzed using the statistical software package R. Analysis of Variance (ANOVA) was used to analyze significant differences. TUKEY HSD tests were performed to confirm the significance between differently treated dried emulsion cakes and powders. Pearson correlation was studied to find the effect of particle size on the amount of oil released.

Material and Methods


Heat-treated (60°C and 75°C) emulsions of native (NS) and OSA-modified (OSA) quinoa starch were freeze dried, as mentioned in the freeze-drying section. A powder was expected after freeze drying. However, the result was cake-like matrixes (Figure 2 a, b).

This was probably due to a thick emulsion layer during drying in combination with the concentration of oil (13%) and starch to oil ratio (300 mg starch/mL oil). This may have caused aggregation of emulsion droplets. Therefore, to obtain powder for analysis, the 60°C heat-treated OSA stabilized dried emulsion cakes were diluted in different ratios of water, 1:2, 1:4, and 1:6. Low dilution provided more cake-like and less powdery matrixes, whereas 1:4 dilution was in between powdery and cakey, and high dilution delivered an almost powder-like matrix (Figure 2c). These powders have been used for strength analysis.

Figure 2: Freeze dried cakes of emulsion of (a) NS – native starch, (b) OSA-modified starch, and (c) powder from different dilutions.

Textural properties of dried emulsions

To our knowledge, there is no suitable existing work about textural properties analysis of dried emulsions. In this study, we tried to develop a method to analyze textural properties, mainly the strength of dried emulsions, with a TVT 6700 texture analyzer (Perten Instruments AB, Sweden). For the analysis, dried emulsion cakes of about 17 mm in diameter and 7 mm in height were used, and compressed under certain forces until oil was released. Primarily we were looking for the exact force where the sample released oil, with an expectation that different samples may release oil at different forces, and from that be able to differentiate samples. Unfortunately, it was difficult to detect the exact force where the oil release is visualized. Different samples absorbed the same force until they released a detectable amount of oil (Figure 3).

Figure 3: Plot of compression force (g) against time.

Based on these results, another technique was applied, i.e. keeping the force constant for all samples and then determining the amount of released oil after compression. Based on the amount of released oil, the samples were distinguished as strong or weak. With this technique, filter paper was used under the sample during compression, which adsorbed released oil prior to compression, as shown in Figure 4. Amounts of released oil have been calculated by weighing the filter paper before and after compression. Reproducible data have been obtained by applying this procedure.

To validate the method, two fine powders of freeze-dried emulsion (orange flavor and prickly pear oil) were analyzed using 6,500 g forces, which gives remarkably consistent results with very small standard deviation. These standard deviations indicate the accuracy of the developed method, Table 2

Figure 4: Dried emulsion sample and filter paper before compression (a), compressed sample and filter paper with released oil (b).

Table 2: Released oil after a single cycle compression of orange flavor and prickly pear oil emulsion-based powders
Sample name

Average released oil

Std dev.
Orange flavor 0.70 0.1
Prickly pear 1.25 0.2

Using this new technique, NS-stabilized 60°C and 75°C heat-treated dried emulsion plus OSA-stabilized 60°C and 75°C heat-treated dried emulsion samples were tested by the application of three different compression forces: 4,500 g, 5,500 g, and 6,500 g. The amount of released oil (mean ± SD, n=3) of four different dried emulsions are presented in Table 3.

Table 3: Released oil after a single cycle compression at different forces (g)
Sample name  4,500 g force  5,500 g force  6,500 g force
Different letters in the same column show significant difference (p<0.05)

There was no significant difference (0.05% level) in the amount of released oil among the two starches used for stabilization for 60°C and 75°C heat-treated dried emulsions for 4,500 g force compression. This means they have the same strength against this force. At 5,500 g force compression, 60°C heat-treated OSA-modified dried emulsion presented as the weakest by releasing a significantly (p<0.05) higher amount of oil (2.5 ± 0.3 mg) compared to 60°C and 75°C heat-treated NS and 75°C heat-treated OSA stabilized dried emulsions. When the 6,500 g force compression was applied, it was found that none of the samples released significantly different (0.05% level) amounts of oil. Though, the amount of released oil tended to be higher in the 60°C heat-treated OSA-stabilized emulsion, followed by 60°C heat-treated NS-stabilized emulsion. This study showed that higher temperature (75°C) heat treatment makes a stronger dried cake of emulsion than lower temperature (60°C) heat treatment.

Textural properties of powders

Powders from 60°C heat-treated OSA-stabilized dried emulsion cakes through 1:2, 1:4, and 1:6 cake to water ratio dilution were analyzed with the developed method and compressed under 6,500 g force. There was no significant (0.05% level) difference in the released oil of the primary cake or 1:2 ratio diluted cakey powder. However, 1:4 and 1:6 ratio diluted powders released notably (0.05% level of significance) higher amounts of oil compared to the primary cake and less diluted sample (Table 4).

Table 4: Released oil after a single cycle compression at 6,500 g force of dried emulsion cake and powders
Sample name                                                              Released oil at 6500 g force
Different letters in the same column show significant difference (p<0.05)

The above result showed that more cake-like samples (i.e. aggregated powders) were stronger against compression. This may also happen with emulsion cakes.

Influence of particle size

The previous result (Table 3) showed that OSA-modified dried emulsion cakes released more oil than NS-stabilized dried emulsion cakes. This could be explained by the particle size distribution of droplets forming the dried emulsion cakes (Figure 5). The PSD curve showed that larger particles sizes for NS, which were the samples that released less oil.

Therefore, a Pearson correlation was conducted to find out the impact of particle size on the amount of released oil for different compression forces. Results are presented in Table 5. The table proves that the particle diameter has a negative correlation (r = -0.576) with the amount of oil released at 5,500 g compression force. Which means that when the particle size was larger, it released less oil and vice versa. Micrographs (Figure 6) of NS- and OSA-stabilized rehydrated emulsions clarify the structures causing different oil release.

Figure 5: Particle size distributions of rehydrated emulsions.

Table 5: Correlation of particle size diameter with released oil in different forces (g)

The micrograph showed that in NS-stabilized dried emulsion cakes, aggregated emulsion droplets were entrapped by gelatinized free starch, which protected the oil droplet during compression, and thus less oil was released. On the other hand, OSA-stabilized dried emulsion cakes presented only a minor aggregation of droplets which therefore broke more easily and released higher amounts of oil.

Figure 6: Micrograph of NS-stabilized rehydrated emulsion (upper), and OSA-modified starch stabilized rehydrated emulsion (lower).


A textural property analysis method has been developed for the investigation of dried emulsion cake and powder. Compression tests gave reproducible results for both cake and powder. Moreover, two fine powders of freeze-dried emulsion of orange flavor and prickly pear oil provided sound reproducible results with very small standard deviation (0.10 and 0.15), which indicates the precision of the method. Emulsion cakes produced using different starches for emulsion stabilization (native- and OSA-modified quinoa starch) and different temperatures for heat treatment before freeze drying gave similar textural strength in terms of lower compression force. However, dried emulsions treated at a higher temperature exhibited superior strength in terms of higher compression force. When the cake was reconditioned into powder by dilution, it resulted in less strength. The strength and weakness of dried emulsions highly depended on their particle size. A negative correlation (r= -0.576) was found between the particle size diameter and the amount of released oil after compression in a Pearson correlation test. This was due to aggregation of droplets and gelatinized starch, rather than the individual droplet size. Further studies are recommended to validate the methods of texture analysis of emulsion cakes and powders by analyzing a large number of samples.


The authors are grateful for materials and technical support of Speximo AB and Perten Instruments AB. The authors would also like to thank Jasmine Bedi and the Swedish Institute (SI) for their support throughout the study.


Aberkane, L., Roudaut, G. & Saurel, R. (2013). Encapsulation and oxidative stability of PUFA-rich oil microencapsulated by spray drying using pea protein and pectin. Food and Bioprocess Technology, 7(5): 1505–1517.

Cano-Higuita, D., Malacrida, C. & Telis, V. (2015). Stability of curcumin microencasulated by spray and freeze drying in binary and ternary matrices of maltodextrin, gum arabic and modified starch. Journal of Food Processing and Preservation, 39(6): 2049–2060.

de Freitas, E. M. & da Silva, L. S. (2007). Use of texture analysis to determine compaction force of powders. Journal of Food Engineering, 80(2): 568–572.

Fitzpatrick, J., Iqbal, T., Delaney, C., Twomey, T. & Keogh, M. (2004). Effect of powder properties and storage conditions on the flow ability of milk powders with different fat contents. Journal of Food Engineering, 64(4): 435–444.

Gharsallaoui, A., Roudaut, G., Chambin, O., Voilley, A. & Saurel, R. (2007). Applications of spraydrying in microcapsulation of food ingredients: An overview. Food Research International, 40(9): 1107–1121.

Mao, L., & Miao, S. (2015). Structuring food emulsions to improve nutrient delivery during digestion. Food Engineering Reviews, 7(4): 439–451.

Marefati, A., Rayner, M., Timgren, A., Dejmek, P. & Sjöö, M. E. (2013). Freezing and freeze-drying of Pickering emulsions stabilized by starch granules. Colloid and Surfaces A: Physicochemical and Engineering Aspects, 436: 512–520.

Rayner, M., Marku, D., Eriksson, M., Sjöö, M. E., Dejmek, P. & Wahlgren, M. (2014). Biomass-based particles for the formulation of Pickering type emulsions in food and topical applications. Colloid and Surfaces A: Physicochemical and Engineering Aspects, 458: 48–62.