Table of Contents
Emulsion Formulation and Stability Study
Most emulsions are naturally unstable and need careful formulation to create dispersions with improved shelf life. Formulators can choose the optimum chemistry from various theories and instrumental techniques to achieve the desired results. This article does not serve as a guide to emulsion formulation, but rather introduces the current analytical techniques that describe how stable emulsions are created.
An emulsion is a mixture of two or more liquids that are not typically miscible. Most emulsions are a two-phase system: a dispersed phase (smaller volume) and a continuous phase (larger volume). There are numerous emulsion types, including water in oil (w/o), oil in water (o/w), and double emulsions such as a water in oil in water (w/o/w) emulsion. In an o/w emulsion, the continuous phase is the water and the dispersed phase is the oil.
An energy source is typically needed to create an emulsion. These can include stirring, shaking, or the use of a homogenizer, Microfluidizer (1), or ultrasound. Destabilization occurs in most emulsions over time, sometimes immediately with the ceasing of the energy input. Addition of chemicals, known as emulsifiers, delays phase separation and extends the stable period.
Emulsifiers are usually surfactants consisting of a hydrophobic R-C chain and a hydrophilic head. The hydrophilic head is oriented towards the water, and the hydrophobic tail towards the organic phase. The emulsifier positions itself in this orientation at the interface, and reduces the surface tension and increases the charge (the zeta potential) on the droplet surface – which results in a stabilizing influence on the emulsion, as shown in Figure 1. Food products such as sodium phosphates, lecithin, and surfactants (both ionic and non-ionic) are also emulsifiers. To increase emulsion stability, viscosity modifiers such as PEG can also be added.
Figure 1. Emulsion with surfactant
Emulsion Formulation and Stability Study
Multiple analytical techniques were employed to study emulsion formulation and stability. For creating oil in water emulsions, two surfactants at varying concentrations were used. The Nicomp dynamic light scattering (DLS) system was used to determine the mean size of the emulsion droplets.
The zeta potential of the droplets for all samples was measured using the Nicomp system, where the zeta potential can serve as a predictor of dispersion stability. The Nicomp DLS measurement results are presented as mean size and polydispersity index, PI (2).
An indication of emulsion stability is the large diameter droplet tail, which was measured using the AccuSizer single particle optical sizing (SPOS) system. The relationship between emulsion stability and the large diameter droplet tail as measured on the AccuSizer is well documented (3, 4). This relationship has been codified into the pharmaceutical test USP 729 “Globule size distribution in lipid injectable emulsions” (5, 6) . In USP 729, the volume percent greater than 5 µm (PFAT5) is used to indicate emulsion stability, with a limit of 0.05%. According to USP 729, the DLS can be used to determine the mean droplet size of emulsions with a limit of less than 500 nm for the intensity mean diameter.
While emulsion stability is indicated by the zeta potential, DLS mean diameter, and large diameter droplet tail, the dispersion stability can be directly measured by Formulaction Turbiscan (7). The Turbiscan detects size change and particle migration in order to quantify destabilization phenomena; in this case, the phase separation (creaming) of the emulsion as a function of time. The Turbiscan is used to place a sample, and an NIR light source scans transmission and backscatter up and down, as shown in Figure 2. The backscatter data was used to characterize the samples, as the emulsions in this investigation were fairly opaque.
Figure 2. Turbiscan transmission (left) and backscatter (right) detectors
The emulsion samples were analyzed at the time of creation using these techniques. The analysis was carried out as a function of time over 15 minutes and several hours. An extensive study of emulsion stability would take longer than the time frame chosen for this study, whose aim was to demonstrate the use of instruments and data, and was not meant to be a reference guide on long-term stability analysis and/or emulsion formulation.
To investigate stability, several oil in water emulsions were created. 1 ml of mineral oil was mixed into 19 ml of DI water, containing a surfactant to create the emulsions. Two surfactants were employed at two concentrations:
A: Anionic surfactant and emulsifier
- A High: 10 g dissolved in 100 ml DI water
- A Low: 2.5 g dissolved in 100 ml DI water
B: Nonionic surfactant and emulsifier
- B High: 5 ml in 100 ml DI water
- B Low: 1 ml in 100 ml DI water
In each formulation, the surfactant was added to the water, stirred for 10 minutes, and heated to 50 °C. The mineral oil was added to the water/surfactant solution, after being heated to 40 °C. An ultrasonic probe was used to sonicate the oil/water mixture for two minutes.
Two techniques were used to measure the particle size:
- Dynamic light scattering (DLS) using the PSS Nicomp Z3000 for submicron particle size and zeta potential
- Single Particle Optical Sizing (SPOS) using the PSS AccuSizer 780 APS for particle size 0.5 – 400 µm
Using DLS and SPOS to measure the stability and size of emulsions is a well-documented approach, and is specified into the pharmaceutical USP test 729 for lipid emulsions (6).
The Formulaction Turbiscan was used to measure emulsion stability.
DLS size: The following settings were used to measure PI and mean size on the Nicomp:
- Channel width: Automatic; typical value was 38 µs
- Liquid viscosity: 0.933 cP
- Temperature: 23 °C (Peltier to cool sample before analysis)
- Laser wavelength: 658 nm
- Intensity setpoint: Automatic
- Cell type: Disposable square cuvette
- Measurement angle: 90 deg
- Algorithm: Gaussian
- Baseline adjust: Automatic
Zeta Potential: The following setup conditions were used to program the zeta potential measurements:
- Temperature: 23 °C
- Scattering angle: -14.14 deg
- Liquid viscosity: 0.933 cP
- Cell type: Dip cell in square cuvette
- Dielectric constant: 78.5
- E-Field strength: 4 V/cm
- Electrode spacing: 0.4 cm
- Analysis type: PALS (not constant current)
- κa: Smoluchowski
SPOS size: The following setup conditions were used to program the AccuSizer APS measurements:
- Data collection time: 60 seconds
- Diluent flow rate: 60 ml/seconds
- Number channels: 128
- Background threshold: 100 part/seconds
- Target concentration: 4500 part/ml
- Calibration: Summation mode
- Sensor: LE400
- Syringe volume: 1 ml
- Injection loop: 0.5 ml
- Sample flow time: 5 seconds
- Sample equilibration time: 60 seconds
- Initial Df2: 1200
Stability: The following setup conditions were used to program the Formulaction Turbiscan measurements:
- Scan rate: Every 30 seconds
- Measurement time: 15 minutes*
- Data reporting: Backscatter and TSI (global)**
- Temperature: 40 °C
*This is a very short measurement time that would typically be extended in such studies.
**The Turbiscan stability index (TSI) is a one-click calculation that compares the variances in the signals from scan to scan. A high TSI means that there are a lot of variances in the scans and therefore a lot of particle movement/size increase and an unstable sample. A low TSI is just the opposite - there are very few variances in the scans and therefore a more stable emulsion.
Figure 3 illustrates the collected data, and shows the destabilization kinetics value of TSI (global) as a function of time over 15 minutes for the four samples.
Figure 3. Turbiscan TSI plots for all samples
These results indicate the following:
Both High A and Low A were extremely unstable emulsions, though High A was more stable than Low A.
Both High B and Low B were much more stable emulsions, though High B was slightly more stable than High B.
For better visual interpretation of the data, Figures 6-10 show the individual backscatter results for these samples. While High A and Low A suffer from very large creaming phenomena at the top of the vial, Low B and High B suffered from only slight size change and particle movement.
Figure 4 shows the volume distribution from the AccuSizer APS system, which is another easy way to understand the collected data.
Figure 4. Relative volume % distributions
As per visual examination, the order of decreasing percentage of large diameter tails is Low A> High A> Low B> High B. This tracks the Turbiscan results seen in Figure 3.
A higher volume percentage of large diameter droplets indicates a less stable emulsion, as documented in references (3, 4, 5). In USP 729, the value for setting specifications is the percentage greater than 5 µm. Figure 4 shows the AccuSizer results that were generated in less than 10 minutes of preparing each emulsion. No droplets appeared in the greater than 5 µm range. Hence, a value such as volume percent greater than 1 µm might be a better calculation to focus on to distinguish these samples at the initial creation time. Sample High A was re- analyzed four hours after the result shown in Figure 4. As shown in Figure 5, the droplets have dramatically increased in size, and there is a distinct population found greater than 5 µm.
Figure 5. High A 10 minutes (blue) and 4 hours (red) after creation
Figure 6. Turbiscan backscatter result for Low A
Figure 7. Turbiscan backscatter result for High A
Figure 8. Turbiscan backscatter result for Low B
Figure 9. Turbiscan backscatter result for High B
Figure 10. Turbiscan TSI global result Low B (pink) high B (orange) enlarged to show differences
It is important to understand that the AccuSizer results shown in this article represent only the large tail and not the entire distribution. The vast majority of the droplets are below the detection limit as the LE400 sensor used for this investigation has a dynamic range of 0.5-400 µm. Therefore, the mean size of the emulsion distributions was determined by using DLS. Table 1 shows the DLS mean size, PI, and zeta potential results.
Table 1. DLS size and zeta potential results
The smaller mean size for surfactant B means a more stable emulsion, as also indicated by the Turbiscan data and AccuSizer data shown in Figures 3 and 4, respectively. For both A and B, a smaller size and PI value and a higher zeta potential value are a result of a higher surfactant concentration. However, since both zeta potential values for A are greater than B indicates that zeta potential alone does not decide the optimum formulation conditions, and the more important consideration is the surfactant that is actually a better emulsifier for a given emulsion type. The hydrophile-lipophile balance (HLB) denotes an empirical relationship between the hydrophilic and hydrophobic parts of a surfactant (8) . A better understanding about surfactant choice and expected emulsion stability is obtained by applying the HLB calculations to a given emulsion formulation. Generally, w/o emulsions require lower HLB surfactants and o/w emulsions require higher HLB surfactants.
Useful information is provided by all instruments used in this investigation to guide the formulation and stability analysis of the emulsions studied. Emulsion stability is directly measured by Formulaction Turbiscan, which provides easy-to-interpret data that quantify the relative stability of the emulsions. In order to accentuate the ability of the Turbiscan to quantify the differences very quickly, emulsions Low A and High A were deliberately created to be very unstable. However, in reality, the Turbiscan measurements would typically require more time scale than used in this brief investigation.
The PSS Nicomp provided quick, easy zeta potential data and mean droplet size, and generated ideal initial information for the emulsions studied. Greater stability is indicated by greater zeta potential. However, to achieve some level of stability, most emulsions only require some zeta potential value more than at least 10 mV. When more than one surfactant is involved, zeta potential alone will not answer all questions regarding the stability of a range of formulations.
To estimate and monitor emulsion stability, the PSS AccuSizer quickly generated unambiguous data. The pharmaceutical industry uses this standard technique to determine the stability of lipid emulsion, and the technique should also see wider usage in general emulsion formulation studies.
1. PSS Application Note - 744 - Size Reduction by a Microfluidizer
2. PSS Technical Note - 730 - DLS Data Interpretation
3. Driscoll, D. et. al., Physicochemical assessments of parenteral lipid emulsions: light obscuration versus laser diffraction, International Journal of Pharmaceutics 219 (2001) 21–37
4. Driscoll, D. et. al., Fat-globule size in a propofol emulsion containing sodium metabisulfite, Am J Health-Syst Pharm—Vol 61 Jun 15, 2004
5. USP <729>, Globule Size Distribution in Lipid Injectable Emulsions
6. PSS Application Note - 736 - USP 729
7. Formulaction, http://www.formulaction.com/
8. Vaughan, C.D. Rice, Dennis A.; Predicting O/W Emulsion Stability by the “Required HLB Equation”; Journal of Dispersion Science and Technology; 1990. Vol. 11 (1), pp 83 — 91
Referenced PSS application and technical notes are available for download at: http://pssnicomp.com/documentation-download-center/
This information has been sourced, reviewed and adapted from materials provided by Particle Sizing Systems.
For more information on this source, please visit Particle Sizing Systems.