Novel Pragmatic Turbidimetric Data Analysis Method for Evaluating the Stability of Emulsions

Abstract

A novel pragmatic turbidimetric data analysis method for evaluating emulsion stability by using a spectrophotometer is proposed. An overview of this method is shown in a flowchart. The advantage of this method is that it uses change in turbidity against time, which allows it to be applied to all emulsions and food scientists to easily calculate half-life time. A high half-life time shows that turbidity is decreased slowly and that the emulsion state can be kept for a long time. Due to this simplicity, it is hoped that this will become a commonly used method.

Keywords:

• Evaluating stability of emulsion
• Spectrophotometer
• Turbidity
• Rate of decreasing turbidity
• Half-life time

INTRODUCTION

Information on the stability of emulsions, for example, how long the emulsion state can be kept, is key information for the quality control of the food emulsion products used in manufacturing and the development of new food emulsifiers. In food manufacturing, checking the stability of emulsions is routine work; in the development of new food emulsion products and new emulsifiers, developers begin by calculating the stabilities of their experimental products and emulsions prepared with their experimental emulsifiers. Economically simple techniques for evaluating the stability of emulsions—that is, methods that are cheap, quick, and accurate—are therefore in great demand.

Currently, the main methods for evaluating the stability of emulsions are visual inspection at constant time intervals,[1 ⁻ 15] microscopic observation,[5] the use of particle size analyzers to measure the size of the oil droplets in the emulsion,[16 ⁻ 22] and devices to measure the absorption of light (e.g., spectrophotometer, turbidimeter, etc.).[23 ⁻ 26]

Visual inspection at constant time intervals is a test to check whether the creaming layer of the emulsion remains or separates into two phases of oil and water. This is currently the standard method for evaluating emulsion stability and is cheap and simple; however, it is a qualitative, not a quantitative, analysis.[1 ⁻ 15] Microscopic inspection and particle size analyzers to measure the size of the oil droplets in emulsions can provide a quantitative analysis. However, changes in particle size do not represent the stability of the emulsion because stability depends on a number of complex factors; rather, they represent the state of the emulsion. Moreover, particle size analyzers are unfit for installation within the flow of the manufacturing process because some time to determine the sizes of the oil droplets takes and the sizes of the oil droplets cannot be obtained immediately.

Devices to measure the absorption of light can be easily installed within the flow of the manufacturing process; moreover, spectrophotometers are already commonly available in research laboratories. Architects of manufacturing flow processes and developers of emulsifiers have, therefore, attempted to quantitatively evaluate emulsion stability by means of devices that measure absorption of light. Various indexes (emulsion activity index [EAI], emulsion stability [ES], stability index [SI; or turbidity ratio]) calculated from the absorption of light have been suggested.[27−37] The EAI is expressed as units of area of interface stabilized per unit weight of emulsifier (EAI = [2 × T] ÷ [φ × C], where T, φ, and C are turbidity calculated from absorption of light at 500 nm, volume fraction of the dispersed phase, and weight of emulsifier per unit volume of aqueous phase, respectively). For ES, as the value decreases, emulsion stability increases (ES = {[EAI(max) − EAI(at temperature)] ÷ EAI(max)} × 100). These two indexes are the most frequently used in food science.[27 ⁻ 37] The stability index (SI = [absorption at 850 nm] ÷ [absorption at 450 nm]) has been suggested based on practical experience in cosmetics and toiletries science.[30,31]

These three indexes do not measure change in turbidity over time; however, since the state of emulsions changes over time, so turbidity must also change. We, therefore, propose that there is a need for a turbidimetric data analysis method for evaluating the stability of emulsions that uses change in turbidity over time. In this article, we present a novel, pragmatic, yet simple method for analyzing change in turbidity against time for evaluating the stability of emulsions.

MATERIALS AND METHODS

Preparation for Emulsification

Materials

 

Materials for buffer solutions

Disodium hydrogen phosphate (Na₂HPO₄•12H₂O) and sodium dihydrogen phosphate (NaH₂PO₄•2H₂O) were obtained from Wako Pure Chemical Industries (Osaka, Japan).

Oils

Corn oil, peanut oil, and soybean oil were obtained from Nacalai Tesque (Kyoto, Japan), and safflower oil was obtained from Sigma-Aldrich Japan Co. (Tokyo, Japan).

Emulsifiers

Bovine serum albumin (BSA) and casein sodium (Casein) were obtained from Wako Pure Chemical Industries, and sodium dodecyl sulfate (SDS) was obtained from Nacalai Tesque.

Stock solutions

 

Stock protein-emulsifier solutions

Stock BSA solution (pH 8.0, ionic strength 0.1) was prepared by dissolving BSA (1.5 g), Na₂HPO₄•12H₂O (3.513 g), and NaH₂PO₄•2H₂O (0.086 g) in 300 mL of water. Stock Casein solution (pH 8.0, ionic strength 0.1) was prepared by dissolving Casein (1.5 g), Na₂HPO₄•12H₂O (3.513 g), and NaH₂PO₄•2H₂O (0.086 g) in 300 mL of water. Stock 0.1% SDS solution was prepared by dissolving SDS (0.15 g) in 150 mL of water.

Emulsification and Dilution of Emulsions with Stock 0.1% SDS Solution

Pre-emulsion solutions were prepared by pipetting the oil and stock protein-emulsifier solution into a jar according to the compositions shown in Table 1. Air dissolved in the pre-emulsifying solutions was removed by aspiration for 20 min. The pre-emulsion solutions were emulsified by using a homogenizer (Ultra-disperser LK-21; Yamato Scientific Co. Ltd., Tokyo, Japan) for 5 min at 293 K. In accordance with the frequently used method of Pearce and Kinsella,[33] 75 μL of each emulsion was gently dissolved into 15 mL 0.1% SDS solution to make a diluted emulsion capable of being analyzed with a spectrophotometer.

Table 1  Emulsion compositions

Fraction of oil (φ)	Oil (mL)	Stock protein-emulsifier solution (mL)
0.02	0.6	29.4
0.06	1.8	28.2
0.10	3.0	27.0

Measurement

Each diluted emulsion was gently poured into a 1-cm-path crystal cell. The cell was installed into a spectrophotometer (Shimadzu UV-1600; Shimadzu, Kyoto, Japan). Absorbance was measured at a wavelength of 500 nm at 30-s intervals for 1800 s. Absorbance at each interval was automatically recorded on a computer. Turbidity was then calculated from each absorbance value using the following equation:^[27−37]

(1)

A graph of turbidity against time was then plotted.

RESULTS AND DISCUSSION

Light basically passes through an emulsion. However, if the light hits oil droplets at a high frequency and scatters, the emulsion can be considered to have a high turbidity. In contrast, if the light hits oil droplets at a low frequency, the emulsion can be considered to have a low turbidity. Thus, turbidity is related to the size and number of oil droplets present in the emulsion. The change in turbidity over time relates to the change in the state of the emulsion over time. Graphs of turbidity against time for all tested diluted emulsions are shown in Fig. 1. All of the graphs exhibit gradually decreasing curves for turbidity against time. According to the author's experience, this pattern approximates that of the curve of the following equation for a first-order reaction:[38]

(2)

where T ₀ is start time, A ₀ is turbidity at T ₀ when the curve of turbidity against time starts decreasing, and Am is minimum turbidity; T ₀, A ₀, and Am were read from the graph of turbidity against time (Fig. 2a). RDT is the rate of decreasing turbidity. Half-life time (HLT) was calculated from the RDT using the following equation:

Figure 1 Graphs of turbidity against time for the diluted emulsions.

Figure 2 (a) Method for reading T ₀ and A ₀ from the graph. (b) Graph of turbidity against time for a diluted corn oil–BSA emulsion (Corn oil fraction, 0.1; emulsifier, BSA). The red line is a curve fitted by using Eq. (2) for first-order reactions.

(3)

Table 2  Fraction of oil and turbidity decreasing parameters for the emulsions emulsified with BSA and coefficient of determination (R²) of the least square-fit calculation

	Corn oil emulsion			Peanut oil emulsion
Fraction of oil (φ)	RDT (ms^−1)	HLT (s)	R^2	RDT (ms^−1)	HLT (s)	R^2
0.02	1.39	499	0.79	1.05	660	0.98
0.06	2.55	272	1.00	4.13	168	0.99
0.10	1.91	363	1.00	2.84	244	1.00
	Safflower oil emulsion			Soybean oil emulsion
Fraction of oil (φ)	RDT (ms^−1)	HLT (s)	R^2	RDT (ms^−1)	HLT (s)	R^2
0.02	1.26	550	0.94	2.52	275	0.92
0.06	2.88	241	1.00	2.35	295	0.95
0.10	3.72	186	1.00	4.62	150	0.99

Table 3 Fraction of oil and turbidity decreasing parameters for the emulsions emulsified with Casein and coefficient of determination (R²) of the least square-fit calculation

	Corn oil emulsion			Peanut oil emulsion
Fraction of oil (φ)	RDT (ms^−1)	HLT (s)	R^2	RDT (ms^−1)	HLT (s)	R^2
0.02	1.20	578	0.94	0.98	707	0.91
0.06	2.70	257	1.00	2.93	237	0.99
0.10	4.61	150	1.00	3.97	175	1.00
	Safflower oil emulsion			Soybean oil emulsion
Fraction of oil (φ)	RDT (ms^−1)	HLT (s)	R^2	RDT (ms^−1)	HLT (s)	R^2
0.02	1.23	563	0.86	1.05	660	0.96
0.06	0.91	762	0.96	0.47	1474	0.97
0.10	1.64	423	0.97	2.42	286	0.99

A high HLT means that turbidity is slowly decreasing and that the emulsion state can be kept for a long time. Because the graph of turbidity against time for the diluted corn oil–BSA emulsion (fraction of corn oil, 0.1; emulsifier, BSA) was a clear, gradually decreasing curve, the author attempted to fit Eq. (2) to the data by using a least square-fit calculation by means of a program that was custom-made by using Visual Basic 6.0 programming language.[37 ⁻ 42] Equation (2) provided the best fit for the data. The RDT, HLT, and coefficient of determination (R ²) of the least square-fit calculation were 1.96 ms⁻¹, 354 s, and 0.98, respectively. The result, therefore, indicates that turbidity against time is a gradually decreasing curve according to the law of first-order reactions (Fig. 2b). However, almost all of the graphs of turbidity against time did not exhibit a clear gradually decreasing curve; instead they exhibited an s-shaped decreasing curve (Fig. 1) despite the author repeating all measurements. The author attempted to fit Eq. (2) to all tested data; however, it did not provide the best fit for the data over the whole range (results not shown). Almost all of the curves of turbidity against time did not exhibit a clear gradually decreasing curve according to the law of first-order reactions.

A further consideration is whether or not most food scientists are able to calculate the direct least squares fit from Eq. (2). Because the computer algorithm for the least square-fit calculation for Eq. (2) is not available in standard spreadsheet applications (e.g., MS-Excel), food scientists would have to make their own computer programs and we cannot expect the average scientist to be able to make their own program. Therefore, using the direct least square-fit calculation for Eq. (2) is not an adequate solution for most scientists. Nevertheless, it is necessary for food scientists to obtain the numerical values on the stability of emulsions from the s-shaped decreasing curves of turbidity against time to verify the stability of their emulsions and further their studies. A pragmatic method for analyzing the data is therefore in great demand. Equation (2) can be converted to the following equation:

(4)

The graph of ln(A − Am) against T − T ₀ should be a proportionately diagonally decreasing line. RDT can be calculated from the slope of this line, from which HLT can then be calculated. Graphs of ln(A − Am) against T − T ₀ for all the diluted emulsions are shown in Fig. 3, along with the lines showing the least squares fit using Eq. (4). The fraction of oil and the turbidity decreasing parameters RDT and HLT of the emulsions, and the coefficient of determination (R ²) of the least square-fit calculation are listed in Tables 2 and 3, respectively. At some point within the range of ln(A − Am) against T − T ₀, all of the graphs exhibited a diagonally decreasing line from which RDT could be calculated (Fig. 3). This method of calculating RDT was applied to all of the graphs from the diluted emulsions. Even if the graphs of turbidity against time did not exhibit a clear gradually decreasing curve, the turbidity decreasing parameters RDT and HLT were calculated by using this method and a standard spreadsheet application. This method, therefore, is a pragmatic and simple way of evaluating emulsion stability. Finally, an overview of this method is shown in the flowchart in Fig. 4. Because of the limited number of methods currently available for evaluating the stability of emulsions, the author hopes that this method will become accepted as a pragmatic and simple way of evaluating emulsion stability.

Figure 3 Graphs of ln(A − Am) against (T − T ₀) for the emulsions, along with the lines (blue) used to calculate the rate of decreasing turbidity. Red circles indicate the points used to determining the lines by using a least squares calculation.

Figure 4 Flowchart for the novel turbidimetric data analysis method for evaluating emulsion stability.

CONCLUSION

A pragmatic turbidimetric data analysis method for evaluating emulsion stability by using a spectrophotometer is proposed. An overview of this method is shown in the flowchart in Fig. 4. Absorbance of a diluted emulsion is measured at constant time intervals and turbidity is calculated. A graph of turbidity against time is plotted, and starting time (T ₀), turbidity at T ₀ (A ₀) when the curve of turbidity against time in the graph starts decreasing, and minimum turbidity (Am) are read from the graph. T − T ₀ and ln(A − Am) are calculated. A graph of ln(A − Am) against T − T ₀ is plotted and the rate of decreasing turbidity (RDT) is determined from the slope of the curve. Finally, half-life time (HLT) is calculated. A high HLT shows that turbidity is slowly decreasing and that the emulsion state can be kept for a long time. This method will enable all food scientists and engineers to calculate the turbidity decreasing parameters RDT and HLT independently by means of standard spreadsheet applications. The author hopes that this method will become accepted as a pragmatic and simple way of evaluating emulsion stability.

ACKNOWLEDGMENT

The author wishes to thank Mariko Tada, a graduate of the Faculty of Pharmaceutical Sciences, Setsunan University, for performing the experiments.

Notes

Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ljfp.