Formulation of nanoemulsion: a comparison between

http://informahealthcare.com/drd ISSN: 1071-7544 (print), 1521-0464 (electronic) Drug Delivery, Early Online: 1–12 ! 2013 Informa Healthcare USA, Inc.. DOI: 10.3109/10717544.2013.866992

RESEARCH ARTICLE

Formulation of nanoemulsion: a comparison between phase inversion composition method and high-pressure homogenization method Sabna Kotta1, Abdul Wadood Khan1, S. H. Ansari1, R. K. Sharma2, and Javed Ali1 Department of Pharmaceutics, Faculty of Pharmacy, Jamia Hamdard, New Delhi, India and 2Institute of Nuclear Medicine and Allied Sciences, New Delhi, India

Drug Delivery Downloaded from informahealthcare.com by 59.179.59.122 on 12/13/13 For personal use only.

1

Abstract

Keywords

There is lot of confusion in the literatures regarding the method of production of nanoemulsion. According to some authors, only the methods using high energy like highpressure microfluidizer or high-frequency ultra-sonic devices can produce actual nanoemulsions. In contrast to this concept, one research group reported for the first time the preparation of nanoemulsion by a low-energy method. Later on many authors reported about the lowenergy emulsification method. The purpose of this work is to formulate, evaluate and compare nanoemulsions prepared using high-energy as well as low-energy method. Nanoemulsions formulated were based on the phase inversion composition technique (low energy method) and were selected from the ternary phase diagram based on the criterion of their being a minimum concentration of Smix used in the formulation. For high-pressure homogenization method (high energy method) Design-Expert software was used, and the desirability function was probed to acquire an optimized formulation. No significant difference (p40.05) was observed in the globule size of formulations made by each method, but the value of polydispersibility index between the two methods was found to be extremely significant (p50.001). A very significant difference (p50.001) was observed in the drug release from formulations made by each method. More than 60% of the drug was released from all the formulations in the initial 2 h of the dissolution study.

Condensation method, high-energy emulsification, low-energy method, nanoemulsion, phase inversion composition

Introduction Emulsions with droplet size in the nanometric scale (typically in the range 20–200 nm) are often referred in the literature as mini-emulsions, nanoemulsions, ultra-fine emulsions, submicron emulsions, etc. (Solans et al., 2005). The properties of nanoemulsion depend not only on the composition but also on the method of preparation. As far as the recent applications are considered, studies on optimization methods for the formulation of nanoemulsions are very much required (Nakajima, 1977; Amselem & Friedman, 1998; ElAasser & Sudol, 2004; Sonneville-Aubrun et al., 2004; Solans et al., 2005). One of the major confusion between nanoemulsion and microemulsion is the method of preparation. According to Mason and associates, emulsions with droplet size in the nanometer range obtained only by shear methods or highenergy methods are considered as nanoemulsions. Those nanoemulsions prepared by condensation methods including phase inversion temperature (PIT) or phase inversion composition (PIC) methods and self-emulsifying methods should

Address for correspondence: Javed Ali, Department of Pharmaceutics, Faculty of Pharmacy, Jamia Hamdard, New Delhi 110062, India. Email: [email protected]

History Received 3 October 2013 Revised 14 November 2013 Accepted 14 November 2013

not be considered as nanoemulsions according to these authors (Mason et al., 2006; Gutie´rrez et al., 2008). To produce nanoemulsion, either a large amount of energy or surfactant or the combination of both is essential. It is obvious that the formulation method influences not only different properties like droplet size, stability of nanoemulsions, but also the nature of the final product. The dispersion phase should be same regardless of the method of preparation whether it uses the high-energy method or low-energy method. The size range is different according to different authors. Some consider 200 nm as the upper limit and some others as 500 nm. Many emulsion properties such as stability, rheology, and colour, depend on the emulsion droplet size (EDS) and size distributions (Becher, 2001). ‘‘Low-energy emulsification’’ methods like PIC and PIT technique involve transitional inversion induced by changing factors which affect the HLB of the system, such as temperature or catastrophic inversion induced by increasing the dispersed phase volume fraction, i.e. composition (Forgiarini et al., 2001; Izquierdo et al., 2004; Tadros et al., 2004; Jafari et al., 2007). But ‘‘high-energy emulsification’’ methods like highpressure microfluidization, high-frequency sonication, etc., are suitable for industrial applications since they allow flexible control of globule size distribution, and are capable of producing fine emulsions from a large variety of materials (Jafari et al., 2007).


2

S. Kotta et al.

In this study, nanoemulsions were formulated and evaluated using high-energy as well as low-energy methods. First, we formulated nanoemulsion by the PIC method with the help of a titration chart. High-pressure homogenizer has been successfully used for the production of nanoemulsion by different researchers. Musthafa et al. observed that on increasing the pressure, the particle sizes of the nanoemulsions were reduced accordingly up to a certain level. However, not any significant reduction in particle size was observed after different cyclic treatments. Qian et al. examined the impact of system composition and homogenization conditions on the formation of nanoemulsions using a high-pressure homogenizer (microfluidizer). They found that the mean particle diameter decreased with increasing homogenization pressure and number of passes, with a linear log–log relationship between mean particle diameter and homogenization pressure. The minimum droplet diameter that could be produced after 6 passes at 14 kbar depended strongly on emulsifier type and concentration. Similarly, Yuana et al. formulated oil-in-water nanoemulsions of b-carotene by high-pressure homogenization and investigated the influence of emulsifying conditions including emulsifier type and concentration, homogenization pressure, temperature and cycle on the properties and stability of the nanoemulsions. They found that the mean diameters of the dispersed particles containing b-carotene ranged from 132 to 184 nm and the size distribution was unimodal and extended from 40 to 400 nm. The nanoemulsions produced with tween 20 had the smallest particle sizes and narrowest size distribution. The particle sizes decreased with increases in homogenization pressure and cycle, and also with temperature up to 50 C. The physical stability of the nanoemulsions decreased with the elevation of temperature but increased with pressure (up to 100 MPa) and homogenization cycle (up to three cycles). Jian Zhang studied the effect of homogenization pressure and number of passes on the mean droplet diameter of whey protein concentrates stabilized emulsions. It was found that increasing pressure resulted in decreasing mean droplet diameter. In the study, homogenization pressures ranged from 6000 to 22 000 psi were evaluated to produce nanoemulsions stabilized by different emulsifiers. With increasing pressure from 6000 psi to 14 000 psi, droplet size decreased at 1 pass but did not change significantly at 2 and 3 passes. When the pressure was further increased to 22 000 psi, mean droplet diameter decreased dramatically compared to that at lower pressures. The smallest diameter obtained was 350 nm when homogenized at 22 000 psi and passed through the system 3 times (Yuana et al., 2008; Qian & McClements, 2011; Zhang, 2011; Mustafa et al., 2012). Using high-pressure homogenizer nanoemulsion has been optimized for their characteristics using Box–Behnken experimental design. The optimized nanoemulsion have been characterized and evaluated. A fixed composition of oil, Smix and water is used for the formulation of nanoemulsion by high-energy method. As high energy was used therefore a lower concentration of surfactant was used. Parameters which influence nanoemulsion characteristics were studied which can be classified as composition or preparation variables. For emulsification by low-energy methods composition variables show a much higher influence

Drug Delivery, Early Online: 1–12

than preparation variables, however for high-energy emulsification, the influence of preparation variables were also determined (Gutie´rrez et al., 2008). The variables selected for the low-energy method were the surfactant oil ratio and the ratio between surfactants, since a surfactant mixture is used. For high-energy method, optimum pressure, percentage of Smix and number of cycles were selected as process variables.

Experimental parameters Materials and methods Materials Efavirenz was gifted by Lupin Ltd. (Pune, India). Capryol 90 (Propylene glycol monocaprylate) and TranscutolÕ HP (Diethylene glycol monoethyl ether) were gifted by Gattefosse (Saint Priest, Cedex, France). Tween 20 (polyoxyethylene sorbitan monolaurate) was purchased from Merck (Schuchardh, Hokenbrunn, Germany). Water was obtained from Milli-Q-water purification system (Millipore, MA). All other chemicals and reagents were of analytical grade and procured from Merck (Mumbai, India) and S.D. Fine Chem. (Mumbai, India). Screening of oil Solubility of efavirenz was ascertained in different oils. An excess amount of drug was added in 2 mL of selected vehicle in stoppered vials and mixed with the help of a vortex mixer (Nirmal International, Delhi, India). These vials were then kept at 25 1 C in an isothermal shaker (Shel Lab shaker incubator, India) for 72 h. The resulting samples were centrifuged at 3000 rpm for 15 min (REMI International, Mumbai, India). The supernatant was filtered through a 0.22 mm filter. The content of drug was determined using HPLC at 247 nm (Bali et al., 2010). The system used was Shimadzu LC-10AT VP having a UV detector (Shimadzu, Kyoto, Japan) and the software used was Class VP, version 5.032. A RP C18 column (25 cm 4.6 mm, 5 mm particle size) was used as the stationary phase along with a mixture of water and acetonitrile (50:50, v/v) as the mobile phase. A flow rate of 0.5 mL/min was employed along with a run time of 15 min. The retention time of efavirenz was found to be 9.1 0.1 min. Selection of surfactant and co-surfactant Surfactants and co-surfactants were selected based on the miscibility with the selected oil. Various surfactants and cosurfactants were screened for miscibility study. Tween 80, tween 20, labrasol, labrafil, lauroglycol, gelucire44/14, PEG 400, PEG200, transcutol, propylene glycol, and ethanol showed complete miscibility with the selected oil phase, i.e. capryol. Preparation of nanoemulsion by low-energy method (PIC) Based on the results of solubility studies, Capryol 90 having HLB value of 6.0 was used as the oil phase for the development of nanoemulsion. Drug enriched oil was used for the study. Geucire 44/14 was used as surfactant and transcutolÕ HP as the cosurfactant. Surfactant and cosurfactant

Formulation of nanoemulsion

DOI: 10.3109/10717544.2013.866992

were mixed (Smix) in different volume ratios (1:0, 1:1, 2:1, 3:1, 4:1, 1:2, and 1:3). Double distilled water was used as the aqueous phase. After the oil and Smix was thoroughly mixed with the help of a vortex mixture. The aqueous phase was added slowly. The amount of aqueous phase was incremented to provide concentration of aqueous phase above 50% of total volume. After each addition of aqueous phase, physical state of the mixture was marked whether it is transparent or opaque. Oil and specific Smix ratio was mixed in volume ratios ranging from 1:9 to 1:0.1 to obtain 16 different combinations like 1:9, 1:8, 1:7, 1:6, 1:5,1:4, 1:3.5, 1:3, 1:2.3, 1:2, 1:1.5, 1:1, 1:0.7, 1:0.43, 1:0.25 and 1:0.1 (Chang et al., 2004; Bali et al., 2010; Parveen et al., 2011).


Preparation of nanoemulsion by high-energy method (high-pressure homogenization) A Box–Behnken statistical design with 3 factors, 3 levels, and 17 runs was selected for the optimization study using DesignExpert software (Design-Expert 8.0.0.6, State-Ease Inc., Minneapolis, MN). The design is suitable for exploring quadratic response surfaces and constructing second-order polynomial models. The experimental design consists of a set of points lying at the midpoint of each edge and the replicated center point of the multidimensional cube that defines the region of interest. The independent and dependent variables and their levels are listed in Table 1. The polynomial equation generated by this experimental design is as follows. R¼ C0 þC1 A þ C2 B þ C3 C þ C4 AB þ C5 AC þC6 BC þ C7 A2 þ C8 B2 þC9 C2 : where R is the dependent variable, C0 is the intercept, C1 to C9 are the regression coefficients, and A, B and C are the independent variables. The experimental design matrix is shown in Table 1. The data obtained from the design were evaluated using DesignExpert software (Design-Expert 8.0.0.6, State-Ease Inc., Minneapolis, MN). Numerical optimization was carried out Table 1. Box–Behnken experimental design data. Independent factors Exp. run 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Dependent factors

ASmix (%)

B-Pressure (MPa)

C-No. of cycle

8 18 8 18 8 18 8 18 13 13 13 13 13 13 13 13 13

22 22 88 88 55 55 55 55 22 88 22 88 55 55 55 55 55

5 5 5 5 2 2 8 8 2 2 8 8 5 5 5 5 5

Globule size SD (nm)

Drug release SD (%)

398.92 7.37 78.0 8.23 340.35 7.2 65.2 4.5 204.76 6.83 24.09 6.08 393.1 9.0 23.43 7.2 70.2 8.3 79.4 5.8 189.1 8.9 131.2 12.5 129.2 9.8 129.3 9.6 31.85 11.5 29.04 9.0 32.13 5.8

50.89 12.5 78.26 6.83 187.0 14.5 292.2 9.0 54.96 6.2 79.94 5.92 53.81 7.38 74.77 7.30 62.4 7.22 65.87 6.51 61.28 6.71 63.06 8.62 65.91 7.92 66.92 9.21 68.19 5.98 68.19 8.62 65.04 7.01

3

for unveiling the optimized formulation. The optimized formulation was prepared and evaluated. The prediction error was determined for the design from the data of the dependent factors yielded from the experimental and predicted values. The emulsion composition was selected based on a preassumption from the low-energy method composition. Process variables were pressure, amount of surfactant and cycles of homogenization. The dependent variables considered for optimization were same as that for low-energy method that is globule size and percentage drug release. First, a coarse emulsion is prepared by mixing the oil and Smix with the help of a magnetic stirrer. Then the aqueous phase was added to it and stirred for 10 min. The prepared coarse o/w emulsion was passed through a high-pressure homogenizer (Stansted, Essex, UK) operating from 22 to 88 MPa. This system could be operated continuously or recycled. Emulsions were homogenized at different pressure and cycles as shown in Table 2. The sample (100 mL) was taken away after each cycle for size analysis and the remaining volume was passed again for the next cycle (Jafari et al., 2007). Evaluation parameters Accelerated physical stability studies Stability of nanoemulsions was studied using heating–cooling cycles, centrifugation, and freeze–thaw cycle stress tests. Heating–cooling cycles between 45 C temperature and room temperature (25 2 C) with storage time of 24 h at each temperature (six cycles each) followed by centrifugation (5000 rpm for 30 min) and then Freeze–thaw cycles at 20 C in a deep freezer (Vest frost, Hyderabad, India) and room temperature (25 2 C) for 24 h were carried out six times (six cycles each) (Chang et al., 2004). The nanoemulsions that were stable were considered for further studies. Dispersibility test Dispersibility studies were performed to evaluate the efficiency of dispersibility of oral nanoemulsion. Each formulation (2 mL) was added to 500 mL of distilled water and 0.1 N HCl in a standard USP dissolution apparatus 2 (Veego, Mumbai, India). Speed of the paddle was adjusted to 50 rpm and the temperature was maintained at 37 0.5 C (Ping et al., 2005). The formulations were visually evaluated for any precipitation. Table 2. Percentage composition of oil, Smix and water of the selected formulations.

Formulation

% of oil

% of Smix

% of water

Surfactant: co-surfactant

Oil: Smix

PIC 1 PIC 2 PIC 3 PIC 4 PIC 5 PIC 6 HPH1 HPH 2 HPH 3 HPH 4 HPH 5 HPH 6

2.86 2.86 2.86 2.86 2.86 2.86 2.86 2.86 2.86 2.86 2.86 2.86

14.3 17.16 14.3 17.16 17.16 17.16 18 18 18 18 13 13

82.84 79.96 82.84 79.96 79.96 79.96 79.14 79.14 79.14 79.14 84.14 84.14

2:1 2:1 3:1 3:1 4:1 1:0 4:1 4:1 4:1 1:0 4:1 4:1

1:5 1:6 1:5 1:6 1:6 1:6 1:6 1:6 1:6 1:6 1:5 1:5

4

S. Kotta et al.

Characterization of nanoemulsion Viscosity determination The viscosity of the formulations (0.5 mL) was determined without dilution using a Brookfield DV III ultra V6.0 RV cone and plate rheometer (Brookfield Engineering Laboratories, Inc., Middleboro, MA) using spindle #CPE40 at 25 0.5 C. The software used for the viscosity calculations was Rheocalc V2.6. For each sample, continuous variation of shear rate (80–400 s1) was applied and the resulting shear stress was measured. Viscosity of nanoemulsions with Newtonian flow properties was calculated.


Refractive index (RI) The RI of the system was measured by an Abbe refractometer (Bausch and Lomb Optical Company, Rochester, NY) by placing one drop of the formulation on the slide in triplicate at 25 C. Percentage transmittance Percentage transmittance of the prepared nanoemulsion formulations was determined spectro photometrically using Shimadzu UV–VIS spectrophotometer (Shimadzu, Japan). A total of 1 mL of the formulation was diluted 100 times using distilled water and analyzed at 247 nm. Droplet size analysis Droplet size of the nanoemulsion was determined by photon correlation spectroscopy using Zetasizer 1000 HS (Malvern Instruments, Worcestershire, UK). The formulation was diluted with distilled water and sonicated in a bath sonicator for 10 min. prior to analysis. Light scattering was monitored at 25 C at a scattering angle of 90 . Effect on dilution Formulation made by high-pressure homogenization method is studied for the effect of dilution on globule size and PDI. Formulation is diluted 1000 times with distilled water and analyzed by dynamic light scattering technique. Transmission electron microscopic (TEM) analysis The morphology of the oil droplets in the nanoemulsion formulations was visualized with TEM analysis. TEM analysis was also important in order to visualize any precipitation of the drug upon addition of the aqueous phase. The nanoemulsion was diluted 100 times and a drop was applied to 300-mesh copper grid. The grid was left for 1 min. The grid was inverted and a drop of phosphotungstic acid (PTA) was applied to the grid for 10 s. Excess of PTA was removed by absorbing on a filter paper and the grid was analyzed using Morgagni 268D (FEI Company, OR, USA) operated at 60– 80 kV at 1550 magnification. In vitro drug release Dissolution studies were performed to compare the release of drug from the formulations which were formulated by lowenergy and high-energy methods. In vitro release test was


performed in 900 mL of distilled water containing 2% SLS (sodium lauryl sulphate) using dissolution apparatus type II, (Veego Scientific, Mumbai, India) at 50 rpm and 37 0.5 C (USP, 2006). Nanoemulsion formulation (5 mL) was placed in treated dialysis bag (MWCO 12000) g/mole, Sigma Aldrich, St. Louis, MO). A total of 5 ml of samples were withdrawn at regular time intervals at 0, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 6 and 8 h. An aliquot amount of dissolution medium was replaced. The samples were analyzed for the drug content using HPLC (Shimadzu LC-10AT VP, Kyoto, Japan) having a UV detector at 247 nm.

Results and discussion Selection of components Oil represents one of the most important excipients in the nanoemulsion formulation, since a very good solubility of the drug in the oil phase is important for the nanoemulsion to maintain the drug in solubilized form. The solubility of efavirenz in different oils was determined and was found to be highest in Capryol 90 (749 4.04 m g/mL). Since the drug exhibited highest solubility in Capryol 90, it was selected as the oil phase for the development of nanoemulsion. In the absence of surfactant generally the oil phase remains in the upper side since it has a low density than the water phase. So, the system will be in thermodynamic equilibrium in the absence of any surfactants if we add surfactants, that remain at the interface for reducing the interfacial tension. Surfactants that are highly soluble in any one of the phase, particularly in the dispersed phase are generally chosen since they can reduce the interfacial tension considerably. When the oil– water interface that is coated with surfactants brought in close to each other a thin film of water will remain at the interface, as a result the interface repels each other due to the like or similar charges of the surfactants. The surfactant chosen must be able to lower the interfacial tension to a very small value to aid the dispersion process during the preparation of the nanoemulsion, provide a flexible film that can readily deform around droplets, and be of the appropriate lipophilic character to provide the correct curvature at the interfacial region for the desired nanoemulsion type (i.e. oil/water, water/oil, or bicontinuous) (Myers, 1999; Lawrence & Rees, 2000; Meleson et al., 2004; Kotta et al., 2012; Zainol et al., 2012). Non-ionic or zwitterionic surfactants are often considered for pharmaceutical applications and nanoemulsion formulation since these are less toxic and less affected by pH anionic strength changes. Sometimes it is not possible to achieve the required interfacial area with the use of single surfactant. If, however, a second amphiphile is added to the system, the effects of the two surfactants can be additive provided that the absorption of one does not adversely affect the absorption of the other and that mixed micelle formation does not reduce the available concentration of surfactant molecule. The second amphiphile is referred to as the cosurfactant. Nanoemulsion is formulated with capryol as oil phase; tween 20, Gelucire 44/14 or Labrasol as surfactant and TranscutolÕ HP, PEG 400 or ethanol as co-surfactant.

DOI: 10.3109/10717544.2013.866992

Construction of phase diagram

5

possible. Therefore, selection of formulations was based on the criterion of their being a minimum concentration of Smix used in the formulation. For each percentage of oil selected, only those formulations that used a minimum concentration of Smix were taken from the phase diagram. Phase diagrams were constructed separately for each ratio of Smix, so that o/w nanoemulsion regions could be identified. The combination of Capryol 90 as oil, Gelucire44/14 as surfactant, and TranscutolÕ HP as cosurfactant yielded broad regions of nanoemulsion. Only these combinations are shown in diagram. Gelucire44/14 and TranscutolÕ HP are non-ionic surfactants having HLB of 14 and 4.2 respectively. Gelucire44/14 (Lauroyl macrogol-32 glycerides EP or Lauroyl polyoxyl-32 glycerides NF) is non-ionic water dispersible surfactant composed of well-characterized


Studies on phase behavior for optimization of nanoemulsion properties are very much important in case of condensation or low-energy emulsification method, because the phases involved during emulsification are determinant in order to obtain nanoemulsions of small droplet size and low polydispersity. In contrast for shear methods or high energy using methods, the final composition are important since there is not a composition emulsification path. Hundreds of formulations can be prepared from the nanoemulsion region of the phase diagram (Figure 1). It is already reported that large amounts of surfactants, mainly ionic surfactants, cause irritation, therefore for drug delivery, nonionic surfactants are preferred in a small concentration as


Figure 1. Pseudo ternary phase diagrams for nanoemulsioin using capryol as an oily phase, Gelucire 44/14 as a surfactant and TranscutolÕ HP as a co-surfactant.


6

S. Kotta et al.

PEG-esters, a small glyceride fraction and free PEG. TranscutolÕ HP is diethylene glycol monoethyl ether EP/NF. The nanoemulsions described in the table are formed with more than 75% of water and they are stable even after infinite dilution. Composition of each component is ultimately determined based on the dose of efavirenz which is to be incorporated or dissolved in a fixed amount of the selected oil that is capryol. When Gelucire alone was used in the Smix an appreciable nanoemulsion region was obtained. The maximum amount of oil that could be emulsified was found to be 23% v/v using 66% v/v of Smix. When cosurfactant was incorporated along with the surfactant in equal proportion, i.e. Smix ratio 1:1 nanoemulsion was not formed even in 1:9 ratio of oil: Smix. On further increasing the proportion of cosurfactant in the Smix from 1:2 to 1:3, it was observed that the nanoemulsion region was further reduced. The observation is similar when the concentration of surfactant was increased to 2:1. On further increasing the concentration of surfactant in the Smix to 3:1, it was observed that there was little increase in the nanoemulsion region but the maximum amount of oil that could be emulsified still remained the same as that of 1:1. No appreciable increase in the nanoemulsion region was observed on further increasing the proportion of surfactant in the Smix to 4:1 and the maximum amount of oil that could be emulsified was found to be 23% v/v using 66% v/v of Smix. While studying phase diagrams it is found that the free energy of nanoemulsion formation depends on the level to which the surfactant lowers the surface tension of the oil–water interface and the change in dispersion entropy. As a result, a negative free energy of formation is achieved when large reduction in surface tension is accompanied by major favorable entropic change. In such cases, nanoemulsification is spontaneous and the resulting dispersion is stable (Myers, 1999; Lawrence & Rees, 2000; Chang et al., 2004; Meleson et al., 2004; Ping et al., 2005; USP, 2006; Bali et al., 2010; Parveen et al., 2011; Kotta et al., 2012). Optimization using high-pressure homogenization The nanoemulsions were prepared as described in the experimental section. All the 17 batches proposed by the experimental design were formulated and evaluated for Globule size, and percentage drug release. The data obtained for the experimental design are shown in Table 1. Screening of variables The emulsion composition was selected based on a preassumption from the low-energy method composition. Based on the resultant data, the lower, middle and upper levels of the four independent variables were determined. Nanoemulsions showed globule size range between 24 and 400 nm and % drug release in between 50 and 80 by restraining the amount of surfactant, pressure and cycles of homogenization at levels of 8–18%, 22–88 MPa and 2–8 times, respectively. Fitting the response surface models The variation in the globule size and percentage drug release was predicted by employing response surface methodology as the responses were the function of the emulsion composition


and preparation variables. The experimental data were statistically analyzed. The statistical analysis was used to determine the best fitted model for the three independent variables. A positive value in the regression equation represents an effect that favors optimization due to a synergistic effect, while a negative value indicates an inverse relationship or antagonistic effect between the factor and the response. It should be mentioned that non-significant (p 40.005) linear terms were included in the final reduced model if quadratic or interaction terms containing these variables were found to be significant (p 50.05) (Copra et al., 2008; Mirhosseini et al., 2009; Woitiski et al., 2009; Zainol et al., 2012). The Model F-value of 177.60 implies the model is significant. There is only a 0.01% chance that a ‘‘Model F-Value’’ this large could occur due to noise. Values of ‘‘Prob4F’’ less than 0.0500 indicate model terms are significant. In this case A, B, C, AC, BC, A2, B2 are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. If there are many insignificant model terms (not counting those required to support hierarchy), model reduction may improve your model. The ‘‘Lack of Fit F-value’’ of 0.58 implies the Lack of Fit is not significant relative to the pure error. There is a 66.11% chance that a ‘‘Lack of Fit F-value’’ this large could occur due to noise. The ‘‘Pred R-Squared’’ of 0.9742 is in reasonable agreement with the ‘‘Adj R-Squared’’ of 0.9900. ‘‘Adeq Precision’’ measures the signal to noise ratio. A ratio greater than 4 is desirable. A ratio of 37.954 indicates an adequate signal. This model can be used to navigate the design space. The model proposed the following polynomial equation for globule size: Effect of independent factors on globule size (R1).

R1 ¼ þ36:29 143:30A 15:01B þ 44:80C þ 11:44AB 47:25AC 16:78BC þ 114:10A2 þ70:23B2 þ10:96C2 : where R1 is the globule size of nanoemulsions, A is the amount of Smix used (%), B is the homogenization pressure and C is the number of cycles of homogenization. It was observed that only one independent variable (C) exhibited a positive effect while both A and B showed a negative effect on the response of globule size (R1). Factor A and B which is the concentration of Smix and homogenization pressure used, affected the globule size in opposite direction to that observed with factor C (number of cycles). The negative coefficient value of factor A and B indicated the decrease in globule size with an increase in Smix concentration and homogenization pressure. Coefficients with more than one factor, or higher order terms in the regression equation, represent the interaction between terms or the quadratic relationship, respectively which suggest a non-linear relationship between factors and responses (Joglekar & May, 1987; Zainol et al., 2012). In this condition, factors can produce a different degree of response than is predicted by the regression equation if they are varied at different levels or more than one factor is changed simultaneously (Tang et al., 2012; Zainol et al., 2012).


DOI: 10.3109/10717544.2013.866992

Effect of independent factors on drug release (R2). The Model


F-value of 88.43 implies the model is significant. There is only a 0.01% chance that a "Model F-Value" this large could occur due to noise.Values of "Prob4F" less than 0.0500 indicate model terms are significant. In this case A, B, C, AB, A2, C2 are significant model terms. Values greater than 0.1000 indicate the model terms are not significant. The "Lack of Fit F-value" of 0.50 implies the Lack of Fit is not significant relative to the pure error. There is a 70.14% chance that a "Lack of Fit F-value" this large could occur due to noise. Non-significant lack of fit is good as the model is required to be fit. The "Pred R-Squared" of 0.9520 is in reasonable agreement with the "Adj R-Squared" of 0.9801. "Adeq Precision" measures the signal to noise ratio. A ratio greater than 4 is desirable. A ratio of 30.108 indicates an adequate signal. So, the model can be used to navigate the design space. The model proposed the following polynomial equation for drug release: R2 ¼ þ66:19 þ 11:84A þ 1:70B 1:28C 1:50AB 1:01AC 0:42BC þ 1:59A2 1:12B2 1:91C2 : where R2 is the % drug release of nanoemulsions, A is the amount of Smix used (%), B is the homogenization pressure and C is the number of cycles of homogenization. According to the equation factor both factor A and B is showing a positive on drug release in opposite to C which is showing a negative effect. That means increase in the level of Smix concentration and homogenization pressure will results in higher percentage of drug release. Since number of homogenization cycle presented a negative effect, increase in this factor will result in lower % of drug release. Since the response was affected by the interaction of independent variables, presenting a quadratic relationship suggest a non-linear relationship between factors and responses. Response surface analysis For the optimization of nanoemulsions, response surface analyses were plotted in three dimensional model graphs. The response surface plots for globule size and percentage drug release which are used to interpret the interaction effect of the variables are presented in Figure 2(a). As shown in the figure, decrease of particle size by increase in the Smix level is due to the fact that the emulsifier plays a vital role in the formation of emulsion as it lowers the interfacial tension, thereby the Laplace pressure, p is reduced and the stress required for droplet deformation is reduced (Vyas & Khar, 2002). It is in agreement with the work of Rabea et al. (Parveen et al., 2011). In the homogenizer, the energy input can be increased by increasing the operating pressure or by passing the emulsion a number of times (cycles), in other words, by increasing the homogenization time. As the pressure increases, there was a decrease in emulsion size however the size is not decreasing with increase in number of cycles alone. This may be due to over processing. There are different reasons related to the incidence of over-processing. One of the most important reasons is related to the role of the emulsifiers. The concentration of the present emulsifier

Effect on globule size.

7

should be enough to coat the entire fresh interface and protect them against re-coalescence since specific surface area of the droplets is increasing dramatically during homogenization. This is because the final globule size of an emulsion is as result of the competition between two opposite processes, droplet breakage and droplet coalescence. Sometimes the timescale of collision is shorter than the timescale of emulsifier absorption; the fresh interface will not be completely covered by emulsifier molecules and leads to re-coalescence and thereby, an increase in the droplet size (Jafari et al., 2007). Drug release It has been reported that drug release can be improved with nanosized formulations. From Figure 2(b) it is clear that amount of Smix and homogenization pressure plays a positive role in enhancing drug release. This may be due to the fact that high level of these two variables resulted a fine nanoemulsion having comparatively small globule size. It is obvious that as the size decreases the release and solubility will be improved. But factor C that is number of cycles of homogenization showed a negative impact on drug release. This could be due to the problem of over processing as mentioned in the above section. Optimization of responses for formulating nanoemulsion by high-energy method By using Design-Expert software, the desirability function was probed to acquire an optimized formulation. After analyzing the effects of the independent variables on the responses the optimized formula was determined. An optimum nanoemulsion is that with smallest globule size and which showed maximum % of drug release. The response surface and contour plot were used to visualize the interaction between the independent variables. By investigating the interaction effect between the independent variables and evaluating the optimization constraints, the optimum nanoemulsion was prepared with a composition as shown in the Table 2. Verification of the reduced models Experimental and predicted values of the responses were compared to check the adequacy of the response surface equations. The numerical optimization method was employed for optimization of the nanoparticulate system. The experimental values were obtained by preparing the nanoparticles using the optimized formula suggested by the software. It was observed that the percentage prediction error was very low (55%) which indicated the accuracy of prediction by the software and the utility of the experimental design for tailoring the nanoparticles with desired parameters. The sufficiency of the corresponding response surface models was verified based on the observations. Accelerated physical stability studies The elimination of meta-stable states can be difficult to ensure in practice because the time constraints impose a physical limit on the length of time systems which can be left

S. Kotta et al.



8

Figure 2. (a) Response surface plots for globule size (R1) (i) surface plot of factor B versus A against R1; (ii) Response surface plot of factor C versus B against R1 (iii) Response surface plot of factor C versus A against R1. (b) Response surface plots for drug release (R2) (i) Response surface plot of factor C versus B against R2; (ii) Response surface plot of factor C versus A against R2 (iii) Response surface plot of factor B versus A against R2.

to equilibrate the system. The centrifugation test showed that the tested nanoemulsions had good physical stability. Through freeze–thaw cycle stress test, turbidity was observed when the nanoemulsions were stored at 21 C. Coagulation of the internal phase at low temperature might have led to this instability; however, these nanoemulsions were easily

recovered by storing at ambient temperature. Chang et al. [2004] reported that nanoemulsions should be kept above 15 C at least. Formulations, which did not pass the accelerated physical stability tests, were dropped out and the remaining were subjected to dispersibility test. In case of macroemulsions, the interfacial energy is much larger than the


DOI: 10.3109/10717544.2013.866992

entropy and hence the process of emulsification is nonspontaneous, i.e. energy is needed to produce the emulsion by the use of high-speed mixture, whereas in case of nanoemulsion the interfacial tension is made sufficiently low so that interfacial energy become comparable or even lower than the entropy of dispersion, and hence the free energy of the system becomes zero or negative. This explains the stability of nanoemulsion. Based on the observation of accelerated physical stability studies, six different combinations were selected from PIC method in which the Smix is minimum. Table 2 shows the % composition of oil, Smix and water of the selected six formulations by PIC method. Similarly, 6 formulations from HPH method which passed all the accelerated physical stability test and dispersibilty test.


Characterization of nanoemulsion Viscosity determination The viscosity of the nanoemulsions (PIC 1-6 and HPH 1-6) are given in Table 3. The viscosity of nanoemulsion formulation was found to be low as expected as one of the characteristic. Viscosity of all the formulations was less than 39 mPas. There is no significant difference (p40.05) in viscosity of nanoemulsion prepared by each method. It is also observed that the viscosity is directly proportional to the concentration of surfactants used in the formulation. It can be observed that, in general, viscosity of all formulations was very low. Refractive index RI is an optical property which can be used to describe the isotropic nature of the nanoemulsion and fundamentally signifies the chemical interaction among drug and excipients. The results are shown in Table 3. No significant difference (p40.05) was observed in the refractive indices of formulations made by each method. In all the nanoemulsion formulations, the refractive index was closer to 1.42, the refractive index of water. Similarity of the refractive index value is a sign of the uniform nanoemulsion structure. From these observations, we can conclude that the optimized nanoemulsion formulations were not only stress stable but also isotropic in nature.

9

Percentage transmittance The percentage transmittance of the optimized formulations was determined. The results are shown in Table 3. A significant difference (p50.001) was observed in the refractive indices of formulations made by each method. This indicates that low-energy method produces more transparent nanoemulsions than high-energy method. A value of percentage transmittance closer to 100% indicated that all of the optimized formulations were clear and transparent. Globule size and size distribution analysis For understanding the behavior of nanoemulsion, the information on droplet size and size distribution is very important since the release and absorption of the drug is related to globule size. In addition to composition, the bioacceptability of the delivery system is also influenced by the particle size (Chang et al., 2004; Porter & Pouton, 2008; Khan et al., 2012). It was observed that the ratio of Smix affected the average droplet size of the nanoemulsion formed. Droplet sizes of the selected nanoemulsions were determined and the results are shown in Table 3 along with the polydispersity indices. Figure 3(b) shows the difference in size distribution of formulations made by each method. No significant difference (p50.05) was observed in the globule size of formulations made by each method. Among nanoemulsions made by PIC method PIC1 and PIC3 is showing little bit higher globule size than others. This may be due to a comparative less amount of Smix. Polydispersity is the ratio of standard deviation to the mean droplet size and denotes the uniformity of droplet size within the formulation (Shakeel et al., 2007; Bali et al., 2011; Khan et al., 2012). The lower the polydispersity value, higher is the uniformity of the droplet size in the formulation. The value of poly dispersibility index between two methods was found to be extremely significant (p50.001). The higher value of PDI may be due to coalescence of some droplets because of highenergy inputs. The graph compares the difference in particle size and PDI before and after homogenization, i.e. only one peak before homogenization and too many peaks after homogenization (Figure 3a). From the graph it can be confirmed that globule size is reduced but PDI increased after homogenization.

Table 3. Viscosity, refractive index and percentage transmittance of selected formulations.

Formulation PIC 1 PIC 2 PIC 3 PIC 4 PIC 5 PIC 6 HPH1 HPH 2 HPH 3 HPH 4 HPH 5 HPH 6

Viscosity (mPas) SD (n ¼ 3)

Refractive index SD (n ¼ 3)

Percentage transmittance SD (n ¼ 3)

Mean globule size (nm) SD (n ¼ 3)

Polydispersity index (PDI) SD (n ¼ 3)

37.13 2.21 32.24 2.18 34.98 3.01 37.44 1.34 33.08 2.53 38.45 3.11 31.87 2.17 30.68 2.72 31.89 3.18 29.88 2.41 29.01 3.72 31.03 1.54

1.361 0.018 1.353 0.009 1.356 0.017 1.362 0.011 1.358 0.017 1.357 0.008 1.353 0.018 1.350 0.013 1.354 0.012 1.358 0.015 1.365 0.017 1.361 0.006

98.36 0.02 99.48 0.01 98.65 0.05 99.32 0.04 98.98 0.08 99.16 0.07 97.91 0.20 98.13 0.21 98.57 0.17 98.27 0.23 96.01 0.13 96.95 0.19

55.96 15.8 28.7 12.7 65.17 16.6 27.04 11.4 49.43 9.9 26.58 12.3 29.2 18.6 29.3 19.9 24.09 22.1 23.43 18.4 78.0 20.8 65.2 22.4

0.150 0.012 0.407 0.011 0.143 0.017 0.262 0.021 0.175 0.019 0.109 0.013 0.808 0.025 0.637 0.032 0.793 0.037 0.748 0.029 0.644 0.034 0.812 0.037

S. Kotta et al.



10

Figure 3. (a) Comparison of size distribution of globule size before and after homogenization. (b) Comparison of size distribution of globule size. (c) Comparison of size distribution of globule size before and after dilution.

Effect on dilution Figure 3(c) compares the globule size and PDI after and before dilution of the formulation made by homogenization method. There is no significant difference in globule size and PDI after dilution. This confirms that the formulation is stable even after infinite dilution. TEM analysis Transmission electron microscopy is the most important technique for the study of microstructures, since it directly

produces images at high resolution and it can capture any coexistent structures and microstructure transitions (Chang et al., 2004). Combination of bright field imaging at increasing magnification and of diffraction modes was used to reveal the form and size of the nanoemulsion. The droplets in the nanoemulsion appear dark, and the surroundings are bright; a ‘‘positive’’ image was seen using TEM. Some droplet sizes were measured using TEM, as it is capable of point-to-point resolution. The droplet size was in agreement with the results obtained from droplet size analysis using the Zetasizer. TEM image of PIC6 is shown in Figure 4.


DOI: 10.3109/10717544.2013.866992


In-vitro release profile The release of drug from the nanoemulsion formulations was compared and a very significant difference (p50.001) was observed in the drug release from formulations made by highenergy and low-energy method. Percentage drug release from the formulations prepared by high-pressure homogenization method was found to be comparatively less than that prepared by low-energy emulsification technique (Figure 5). This can be correlated with the high polydispersibility index of nanoemulsion formulations prepared by high-energy method. More than 60% of the drug was released from all the formulations in the initial 2 h of the dissolution study. This could be attributed to the small globule size of nanoemulsion which provided large surface area for the release of drug and thus permitting faster rate of drug release. In the present study, the dialysis bag with pore size having molecular cut-off 12 000 g/mole was used to check the formulation release potentials based on size distribution. The nano formulations have variations in the droplet sizes. Therefore, instant release

Figure 4. Transmission electron microscopic (TEM) analysis.

11

followed by the sustained release profile might be the outcome of variable size distribution. The smaller globules passage through the dialysis bag pores result in instant release whereas the sustained release may be due to the drug release from the larger globules which later pass through the pores of dialysis bag. The dispersibility studies results proved that the formulated nanoemulsion is easily and completely dispersed in the simulated gastric medium and also has the ability to remain as nanoemulsion upon dispersion in the aqueous environment of the gastro intestinal tract (GIT). The rate of drug release from the formulation PIC6 is the highest nearly 90%; this may be due to the small particle size.

Conclusion Studies on phase behavior for optimization of nanoemulsion properties are important when low-energy emulsification method is used, because the phases involved during emulsification are determinant in order to obtain nanoemulsions of small and uniform droplet size. The importance of the phase behavior, namely crossing microemulsion or lamellar liquid crystalline phase regions during emulsification is important and can be concluded that by slow addition of water to a lamellar liquid crystalline phase nanoemulsions can be obtained. Regarding optimization in the preparation of nanoemulsions by low-energy methods, it is confirmed that crossing bicontinuous or aqueous continuous phases during emulsification allowed obtaining o/w nanoemulsions of small droplet size and low polydispersity. In contrast, for high-pressure method there is not a composition emulsification path and only phases at the final composition are important. Or we can say optimization is experimentally carried out by selective variation of one variable. Optimizations by HPH using selective variation of experimental parameters or designs agree to conclude that, with respect to composition variables, generally there is an

Figure 5. Percentage drug release from formulations PIC1-6 and HPH 1-6.

12

S. Kotta et al.

optimum Smix composition. It is also observed that the higher the oil to surfactant ratio, the greater the droplet size. Highpressure homogenization was also capable of producing nanoemulsions in almost all experiments even with 8% of surfactant. But in almost all cases the PDI is found to be high. Therefore, selection of optimum range of process variables like homogenization pressure and number of cycles of homogenization are critical for the production of efficient nanoemulsion. So, it can be concluded that low-energy method can also produce efficient nanoemulsion and they are more uniform when compared with nanoemulsion produced by high-energy method. Moreover this can be scaled-up, from lab to industrial, and similar results can be expected.


Acknowledgements The authors are grateful to the Defense Research Development Organization (DRDO) Government of India for providing a fellowship as financial assistance to Sabna Kotta.

Declaration of interest The authors declare no conflicts of interests. The authors alone are responsible for the content and writing of this article.

References Amselem S, Friedman D. (1998). Submicron emulsions as drug carriers for topical administration. In: Benita S, ed. Submicron emulsions in drug targeting and delivery. London: Harwood Academic Publishers, 153–73. Bali V, Ali M, Ali J. (2010). Study of surfactant combinations and development of a novel nanoemulsion for minimising variations in bioavailability of ezetimibe. Coll Surf B: Biointerf 76:410–20. Bali V, Ali M, Ali J. (2011). Nanocarrier for the enhanced bioavailability of a cardiovascular agent: in vitro, pharmacodynamic, pharmacokinetic and stability assessment. Int J Pharm 403:1246–56. Becher P. (2001). Emulsions: theory and practice. Oxford (UK): Oxford University Press. Chang X, Weng T, Zhao X, et al. (2004). A study of microemulsion systems for transdermal delivery of triptolide. J Contr Rel 98:427–36. Copra SK, Talegaonkar S, Kohli K, et al. (2008). Chitosan-sodium alginate nanoparticles as submicroscopic reservoirs for ocular delivery: formulation, optimization and in vitro characterization. J Pharm Biopharm 68:513–25. El-Aasser MS, Sudol ED. (2004). Miniemulsions: overview of research and applications. JCT Res 1:21–31. Forgiarini A, Esquena J, Gonzalez C, Solans C. (2001). Formation of nanoemulsions by low-energy emulsification methods at constant temperature. Langmuir 17:2076–83. Gutie´rrez JM, Gonza´lez C, Maestro A, et al. (2008). Nano-emulsions: new applications and optimization of their preparation. Curr Opin Colloid Interface Sci 13:245–51. Izquierdo P, Esquena J, Tadros Th. F, et al. (2004). Phase behavior and nano-emulsion formation by the phase inversion temperature method. Langmuir 20:166594–98. Jafari SM, He Y, Bhandari B. (2007). Optimization of nano-emulsions production by microfluidization. Eur Food Res Technol 225:733–41. Joglekar AM, May AT. (1987). Product excellence through design of experiments. Cereal Foods World 32:857–68.


Khan AW, Kotta S, Ansari SH, et al. (2012). Potentials and challenges in self nano emullsifying drug delivery system (SNEDD). Expert Opin Drug Deliv 9:1305–17. Kotta S, Khan AW, Pramod K, et al. (2012). Exploring oral nanoemulsions for bioavailabilty enhancement of poorly water soluble drugs. Expert Opin Drug Deliv 5:585–98. Lawrence MJ, Rees GD. (2000). Microemulsion-based media as novel drug delivery systems. Adv Drug Deliv Rev 45:89–121. Mason TG, Meleson K, Wilking J, et al. (2006). Nanoemulsions: formation, structure, and physical properties. J Phys Condens Matter 18:R635–66. Meleson K, Graves S, Mason TG. (2004). Formation of concentrated nanoemulsions by extreme shear. Soft Mater 2:109–23. Mirhosseini H, Tan CP, Taherian AR, Boo HC. (2009). Modeling the physicochemical properties of orange beverage emulsion as function of main emulsion components using response surface methodology. Carbohyd Polym 75:512–20. Mustafa G, Kumar N, Singh T, et al. (2012). Effect of homogenization on the fate of true nanoemulsion in brain translocation. Sci Adv Mater 4: 1–10. Myers D. (1999). Surfaces, interfaces, and colloids, 2nd ed. New York (NY): Wiley. Nakajima H. (1977). In: Solans C, Konieda H, eds. Industrial applications of microemulsions. New York (NY): Marcel Dekker. Parveen R, Babootaa S, Ali J, et al. (2011). Oil based nanocarrier for improved oral delivery of silymarin: in vitro and in vivo studies. Int Journal Pharmaceutics 413:245–53. Ping L, Ghosh A, Wagner RF, et al. (2005). Effect of combined use of nonionic surfactant on formation of oil-in-water microemulsions. Int J Pharm 288:27–34. Porter CJ, Pouton CW. (2008). Adv. Formulation of lipid-based delivery systems for oral administration: materials, methods and strategies. Drug Deliv Rev 60:625–37. Qian C, McClements DJ. (2011). Formation of nanoemulsions stabilized by model food-grade emulsifiers using high-pressure homogenization: factors affecting particle size. Food Hydrocoll 25:5;1000–08. Shakeel F, Baboota S, Ahuja A, et al. (2007). Nanoemulsions as vehicles for transdermal delivery of aceclofenac. AAPS Pharm Sci Tech 8: E1–9. Solans C, Izquierdo P, Nolla J, et al. (2005). Nano-emulsions. Curr Opin Colloid Interface Sci 10:3102–10. Sonneville-Aubrun O, Simonnet JT, Alloret FL. (2004). Nanoemulsions: a new vehicle for skincare products. Adv Colloid Interface Sci 145: 9108–9. Tadros T, Izquierdo P, Esquena J, Solans C. (2004). Formation and stability of nano-emulsions. Adv Colloid Interface Sci 108,109: 303–18. Tang SY, Manickam S, Wei TK, Nashiru B. (2012). Formulation development and optimization of a novel cremophor EL-based nanoemulsion using ultrasound cavitation. Ultras Sonochem 19: 330–45. USP. (2006). Rockville (IL): The United States Pharmacopoeial convention Inc. Vyas SP, Khar RK. (2002). In: Vyas SP, Khar RK, eds. Targeted and controlled drug delivery: novel carrier systems, Chapter 7. New Delhi: CBS publishers. Woitiski CM, Veiga F, Ribiero A, Neufeld R. (2009). Design for optimization of nanoparticles integrating biomaterials for orally dose insulin. Eur J Pharm Biopharm 73:25–33. Yuana Y, Gaoa Y, Zhaob J. (2008). Characterization and stability evaluation of b-carotene nanoemulsions prepared by high pressure homogenization under various emulsifying conditions. Food Res Int 41:61–8. Zainol S, Basri M, Basri HB, et al. (2012). Formulation optimization of a palm-based nanoemulsion system containing levodopa. Int J Mol Sci 13:13049–64. Zhang J. (2011). Novel emulsion-based delivery systems. Available at: http://conservancy.umn.edu/bitstream/117545/1/Zhang_umn_0130E_ 12216.pdf [last accessed Nov 2013].