Nanomaterials in Drug Delivery, Imaging, and Tissue Engineering

By Ashutosh Tiwari and Atul Tiwari

This groundbreaking, multidisciplinary work is one of the first books to cover Nanotheragnostics, the new developmental edge of nanomedicine. Through a collection of authoritative chapters, the book reports on nanoscopic therapeutic systems that incorporate therapeutic agents, molecular targeting, and diagnostic imaging capabilities. An invaluable reference for researchers in materials science, bioengineering, pharmacy, biotechnology, and nanotechnology, this volume features four main parts on biomedical nanomaterials, advanced nanomedicine, nanotheragnostics, and nanoscaffolds technology.

• Language: English
• Publisher: Wiley
• Release date: Feb 19, 2013
• ISBN: 9781118644744

Book preview

Part I

BIOMEDICAL NANOMATERIALS

Chapter 1

Nanoemulsions: Preparation, Stability and Application in Biosciences

Thomas Delmas¹ Nicolas Atrux-Tallau¹,⁴ Mathieu Goutayer¹ Sang Hoon Han² Jin Woong Kim³ Jérôme Bibette⁴

¹Capsum, Marseille, France

²Amore-Pacific Co. R&D Center, Bora-dong, Giheung-gu, Yongin-si, Gyeonggi-do, South Korea

³Department of Applied Chemistry, Hanyang University, Gyeonggi-do, South Korea

⁴ESPCI ParisTech, Lab Colloides and Mat Divises, Paris, France

Abstract

Nanoencapsulation is being thoroughly investigated for the encapsulation and delivery of actives and/or contrast agents. Our approach allows for better solubilization, protection, transportation, and delivery of encapsulated molecules to their biological site of action. This is expected to increase treatment efficiency while reducing possible side effects through dose reduction and/or targeted delivery. Among other nanocarriers, lipid nanoparticles are biocompatible, biodegradable, can be easily produced by up-scalable processes and, depending on the lipid physical state, may allow control of the release of encapsulated molecules.

This chapter will explain how nanoemulsions can be efficiently used as nanocarriers for drug delivery and imaging. We will first emphasize the importance of specific formulation to reach long-term physical stability of the nanoparticles in simple and more complex formula. We will then highlight how the lipids’ physical state dramatically impacts the actives encapsulation and release behaviors. Finally, we will explore the interaction of nanoemulsion with biological media in terms of biocompatibility and targeting possibilities. Differences between the two main application domains envisaged, namely pharmaceutics and cosmetics, are detailed, and implications for nanoemulsion preparation are discussed.

Keywords: Nanoemulsion lipid nanoparticles entropic stabilization trapped species active encapsulation release kinetics passive targeting active targeting

1.1 Introduction

The use of nanostructured materials is envisioned to revolutionize biosciences and biomedical applications through earlier and more acute diagnosis, or personalized and controlled therapy [1, 2]. For this purpose, numerous nanocarrier types have been proposed as delivery vehicles for contrast agents or drug molecules for all possible administration routes [3]. Nanocarriers can indeed significantly improve therapeutic efficiency while limiting possible undesirable effects, through specific delivery to the pathological zone and control over active molecules release [4]. One of the first requirements for these systems is to present an absolute harmless-ness and biocompatibility. It is for this reason that the scientific community has principally focused its work on the development of organic particles.

Among a wide variety of nanocarriers, lipid-based systems have aroused interest because of their composition which is based on natural lipids already widely present in the organism, and therefore confers such carriers with high biocompatibility and biodegradability [5, 6]. As previously shown, these lipid nanocarriers can furthermore be easily produced by versatile and up-scalable processes [7-9], and can provide the possibility to control actives encapsulation and release [10].

Hydrophobic molecules encapsulation favors the use of lipid nanospheres, instead of the originally developed nanocapsule initiated by the liposomes discovery in 1964 [11]. The typical structure of a lipid nanosphere is directly derived from nanoemulsion structure, typically relying on a lipid core surrounded by a membrane of diverse surfactants. The history of lipid nanospheres development has followed the understanding of the importance of lipids physical state:

Classical nanoemulsions were the first lipid nanospheres to be introduced decades ago. These systems were composed of a liquid lipid core, stabilized by a membrane of surfactants. Despite the interest of these initially developed systems for solubilization of lipophilic actives in aqueous phases, few finally reached the market due to formulation issues. Indeed, these systems used to suffer from low colloidal stability (even though this has been dramatically improved as will be shown here), and sustained release of encapsulated actives is difficult to achieve due to the low viscosity of the dispersed phase [12], high surface/volume ratio, and low rigidity of the surfactants membrane. This generally leads to the rapid diffusion of the drugs out of the droplets.

Solid Lipid Nanoparticles (SLN) have thus been proposed to overcome these limitations. These systems present a structure identical to nanoemulsions. However, the internal lipids forming the nanoparticle core are crystalline lipids here, conferring a solid nature to the particle’s core (Figure 1.1). SLN are thus generally composed of pure long chain triglycerides, wax or long chain carboxylic acids [8]. They can be stabilized by all types of surfactants; their choice being principally dictated by the administration route envisaged [8]. SLN fabrication processes are similar to nanoemulsions, but the lipid phase is generally heated above the lipids fusion temperature to favor droplet size reduction. Lipid crystallization then occurs following cooling and storage. However, despite high expectations of such systems for prolonged release of hydrophobic molecules, SLN have shown limited controllability. As will be further explained, crystallization of the lipid phase generally leads to active/lipid phase separation and subsequent expulsion, providing high burst release [13, 14].

Nanostructured Lipid Carriers (NLC) were introduced as a compromise. Composed of a mixture of liquid and solid lipids, the NLC core presents an imperfect crystallization which favors better encapsulation ratio thanks to lower crystallinity, while allowing control over release kinetics through the solid character of the lipid phase [15]. Three different types of NLC have been proposed (Figure 1.1): 1) the imperfect type, whose crystallinity is lowered by creating imperfections in the crystal lattices; 2) the structureless type, which is solid but amorphous; and 3) the multiple O/F/W type, in which small droplets of liquid lipids are phase separated in the solid matrix [16]. Although these 3 types can theoretically be obtained, the mixture of spatially incompatible lipids generally leads to the obtainment of the first type NLC [17, 18]. Several studies have reported the obtainment of the forms II and III, however, there is discrepancy in the conclusions. Some authors thus account for supercooled melt rather than amorphous solid particles concerning the structureless type [15, 19, 20]. Similarly, the same system was described as a typical multiple oil in fat in water type (O/F/W) [21], while a complete demixing of oil from wax occurs in the so-called nanospoon structure [22–24].

Figure 1.1 Different types of solid lipid nanospheres.

The further development of these nanoemulsion-based systems for application in biosciences should rely on the complete understanding of nanoemulsion physicochemical properties, production procedures, stability rules, and on control over the internal lipids physical state. We will present here a general approach that can be followed to formulate such systems aiming for biomedical applications. After giving a thermodynamic definition of nanoemulsions, we will first describe possible production procedures and explain the general rules that need to be followed to formulate stable nanoemulsions. We will then investigate the role of the lipids physical state over particle stability, actives encapsulation and release, along with the biocompatibility of the particles. Next we will show how this understanding allows for finely tuning nanoemulsions properties in order to have control over the applicative properties of biodistribution and actives encapsulation/release. Two main domains of application are then detailed through examples of nanoemulsion-based systems for biomedical imaging and drug delivery in the pharmaceutical field, and topical delivery for cosmetics.

1.2 Nanoemulsion: A Thermodynamic Definition and Its Practical Implications

1.2.1 Generalities on Emulsions

An emulsion is a mixture of two immiscible liquids, one liquid being dispersed in the other as droplets, stabilized by surfactants. The size of the droplets may vary from the characteristic size of micelles (10–20 nm) to diameters larger than the micrometer. This mainly depends on the surface tension of the system and the energy provided by the production process [25]. Although direct (oil in water) and inverse (water in oil) similarly exist, we will focus on direct emulsions.

Interest: Emulsions are largely used to solubilize and transport substances in a continuous phase in which they are normally not soluble: hydrophobic substances can, for instance, be easily solubilized in water without the use of any solvents [25, 26]. Such an approach is of high interest for surface treatment like route surfacing or painting; the aim being to depose hydrophobic substances onto surfaces through a continuous water phase that will evaporate, and form a film by fusion between adjacent droplets. Emulsions are also widely used for their rheological properties: it is indeed possible to change liquid solutions into semi-solid formulations such as gels or creams, property largely used in the food and pharmaceutical/cosmetics industry; or, inversely, to make macroscopic solid become easily spreadable, such as bitumen for route surfacing.

Production: Most emulsion systems generally require energy for their formation. One part of it allows overcoming the surface-free energy required to increase the interface between the two phases (ΔG = γΔA; with ΔG, free energy of the system; γ, surface tension; ΔA, created interface area) and finely disperse one phase into the other. The other important part is also used to overcome the viscous resistance along the scission of large globules into small droplets. Finally, the last part is simply lost through dissipation by the Joule effect. Different methods exist and are already used at a laboratory- and industrial-scale: mechanical agitation, high pressure homogenizer or ultrasonication and microfluidics systems [25, 26].

Stability: Once formed, emulsions can present highly different life-time, depending on their composition and their production procedure. This stability can thus vary from a few hours to more than a year [25, 27, 28]. Among the most important parameters are the mutual solubility of the two phases and the surfactant(s) type(s) and concentration(s) [25, 27]. Indeed, different phenomena can lead the system to destabilize: some of them rely on particles aggregation and gravitation, and are reversible; others, related to droplets size evolution, are irreversible [26]. Reversible flocculation phenomena lead the system to sediment or cream, depending on the relative density of the two phases. In parallel, the irreversible evolution of the droplets size can be due to two different mechanisms: Ostwald ripening and coalescence [29, 30]. Ostwald ripening favors the growth of big globules to the detriment of smaller ones, by dispersed phase transfer across the continuous phase. This phase transfer is due to the Laplace pressure difference between droplets of different sizes (PLaplace ~ 1/r; with r the droplet radius). Meanwhile, coalescence corresponds to the breakage of the surfactants’ film separating two droplets in contact, leading to their fusion.

1.2.2 Nanoemulsion vs. Microemulsion, a Thermodynamic Definition

Among emulsions, nanoemulsions differ from micro- or macroscopic emulsions because of their size, ranging from 20–200 nm, which confers upon them unique visual and rheological properties. Because such nanometric size is well below the visible spectrum wavelength (400–800 nm), light is not significantly diffused, even for large refractive index differences between dispersed and continuous phases, making the dispersion transparent, or at least translucent. These unique optical and rheological properties make nanoemulsions of primary interest in such domains as cosmetics and pharmaceutics for the encapsulation and delivery of actives.

Nanoemulsions are thus specific emulsions of nanometric size. Here we need to clearly differentiate them from microemulsions. Indeed, although the two systems present similitudes in terms of composition (ternary system of oil/water/surfactants), droplets size (20–100 nm), and therefore optical appearance and rheological properties, and may finally present similar apparent stability (system stable at one day, one month or even one year), nanoemulsions and microemulsions are in essence completely different systems.

Nanoemulsions are emulsions, and consequently they are out-of-equilibrium systems. The resulting system is not at the equilibrium because the two phases present a surface tension (>0) which will lead the system to minimize its interfacial area until the obtainment of the thermodynamic equilibrium: two separated phases of oil and water [25, 27].

Microemulsions (also referred to as swollen micelles) are thermodynamically stable systems [31–33]. Microemulsions are also formed by two immiscible liquids (oil and water), one surfactant, and possibly one co-surfactant. Nonetheless, the dissolution of amphiphilic molecules and oil in water leads to the obtainment of a unique phase, optically isotropic, of swollen micelles defining the thermodynamically stable state of the system [34]. It means that, from a thermodynamic point of view, the system is stable, and the phase separation of oil and water will not decrease the free energy of the system. Microemulsions are therefore not emulsions in the thermodynamic sense.

These differences can be highlighted by looking at energy landscape diagrams. Microemulsions, as micelles, are thermodynamically-stable systems, being formed by spontaneous self-assembly (ΔGA→B, < 0) (Figure 1.2a) [33]. Conversely, nanoemulsions are metastable systems that require high energy input to be formed (ΔGA→B > 0) (Figure 1.2b) [29, 30]. They do not possess a thermodynamic stability, but may sometimes present a kinetic stability ranging from a few hours to a few days, or even to several years [27, 29, 35]. This kinetic stability will, or will not, be observed depending on the height of the energetical barrier ΔGbarrier which separates this nanoemulsion state (B) from the thermodynamically favored state of phase separation (A). If this barrier is lower, or close to the thermal energy kBT, the system will be unstable; conversely, the system may be blocked in this nanoemulsion state for a long period of time if the barrier is higher. These differences have still been shown to be maintained whatever the close proximity between a nanoemulsion state and a microemulsion state in the pseudoternary (oil/water/surfactant) phase diagram [36].

Figure 1.2 Thermodynamical differences between the microemulsion state (a) and nanoemulsions; (b): Energy diagrams.

Like common emulsions, nanoemulsions are mainly used to solubilize hydrophobic species in a water continuous phase. Nanoemulsions differ from common macro- or microscopic sized emulsions because of their size ranging from 10–200 nm that confers them with unique optical and rheological properties. They are nonetheless true emulsions, in the thermodynamic sense, and dramatically differ from microemulsion, being metastable systems. This implies that they generally require high energy input for their formation and a specific strategy for their stabilization [36].

1.3 Stable Nanoemulsion Formulation

After introducing the specificity of nanoemulsions compared to classical emulsions and microemulsions, the aim of this section is to define possible production processes, discuss ways of stabilizing these metastable systems, and describe the accessible formulation domain giving nanoemulsions possessing long-term physical stability.

1.3.1 Nanoemulsion Production

1.3.1.1 Fabrication Procedures

As previously shown, the formation of nanoemulsions from two separated phases of oil and water generally requires energy input. Considering the aimed small droplets sizes, and thus the large quantity of interface to generate, a huge amount of energy needs, in fact, to be provided to the system. Droplets are split when the applied shear rate is larger than the Laplace pressure. In the case of very diluted dispersion, the Taylor equation obtains the order of magnitude of the shear rate required to reach droplets of a certain size dp (Eq. 1.1) [37]:

(1.1) equation

with γ, the surface tension (N·m−1); ηc, the continuous phase viscosity (Pa·s) and , the shear rate (s−1).

For instance, the obtainment of 20 nm diameter particles requires the use of shear rates close to 10⁹s−1 [30, 37, 38]. Few devices allow the obtainment of such high shear rates. The most commonly used are the high frequency ultrasonic devices, the high pressure homogeneizers (HPH), and the microfluidizers [30, 39]. Microfluidizers seem to be more efficient than sonication and HPH when aiming for sample homogeneity; nonetheless, production costs are far more important, lowering the viability of its industrial use [40]. The use of ultrasounds for emulsification is increasingly studied for its lower energy consumption, its use of fewer amounts of surfactants, and the obtainment of smaller sizes for more homogeneous products than classical mechanical processes [29, 39, 40]. In addition, sonication is a very flexible procedure that allows working with smaller or more viscous samples rather than with high-pressure homogeneizers [40]. Yet, HPH is an interesting technique for industrial production, as it may allow the easy production of large amounts of nanoemulsion.

It has nonetheless to be noted that nanoemulsions can also be obtained thanks to low energy processes that rely on passages through microemulsion states. The PIT (for Phase Inversion Temperature) approach is among one of the most employed. Its principle relies on the modification of PEG surfactant solubility along a temperature change [41]. It is possible to force the system to continuously pass from a direct microemulsion state to an inverse microemulsion state by a simple temperature increase [42]. Doing a cycle around the HLB temperature, corresponding to the bicontinuous phase of zero spontaneous curvature, it is possible to obtain direct microemulsions that will ultimately lead to nanoemulsion by further decreasing the temperature [35, 43]. A similar principle is used by the CPI (Catastophic Phase Inversion) techniques. In this case, formulations parameters, such as salt concentration or dispersed phase concentration, are used to change the sign of the spontaneous curvature, thus blocking the microemulsion state formed (most of the time through an important dilution) [35]. These low energy methods therefore allow for the obtainment of nanoemulsions. However, the final nanoemulsion properties are dictated by the initial microemulsion, which dramatically restrains the accessible formulation domain [29, 35, 44].

Among all these preparation procedures, we have focused our work on ultrasonic emulsification at laboratory scale, while exploring High Pressure Homogeneization at a larger industrial scale, in order to ensure the preparation of small droplets with easy to implement procedures and low production costs.

1.3.1.2 Nanoemulsion Preparation by Ultrasonication

By using ultrasonication it is also possible to model the size decrease along the preparation procedure. Once a pre-emulsion is formed, the droplet size follows a first-order exponential decay of the sonication time (Figure 1.3a) [36]. The exponential decay suggests that the droplets may reach their final size y0 in a single step during sonication, and do not significantly undergo coalescence as in the surfactant-rich regime defined by Taisne et al. [45].

Figure 1.3 Nanoemulsion preparation by ultrasonication: (a) typical droplet size evolution along sonication; (b) droplet size evolution as a function of the energy input.

The characteristic decay time is mainly governed by the energy input, while the saturated size y0 is determined by the characteristic surface tension of the system. The predominant factor affecting the kinetics of droplet size reduction is the energy density provided to the system [36]. It is thus possible to rescale all obtained kinetics to a unique master curve when plotting the size reduction as a function of the total energy provided to the system along sonication (Figure 1.3b). This concept should be generalizable to all preparation procedures, as long as the energy provided is sufficient to counteract the Laplace pressure. The obtained saturated size y0 is then mainly governed by the types and concentrations of the introduced surfactants. Increasing the PEG chain length of PEG-stearate surfactants leads, for instance, to larger particles, as such a change increases the corresponding characteristic surface tension.

As detailed, nanoemulsions being true emulsions, in a thermodynamic sense, they require large energy input for their formation. In addition, this metastable state also additionally needs to be stabilized to prevent it from reaching its favored thermodynamic state of phase separation.

1.3.2 Nanoemulsion Stability Rules

1.3.2.1 Nanoemulsions Destabilization Mechanisms

As previously introduced, nanoemulsions are a thermodynamically unstable system, even though a kinetic stability may sometimes be observed [27, 44, 46]. The small sizes of nanoemulsion droplets prevent them from undergoing reversible destabilization mechanisms linked to gravity. Indeed, the migration velocity of the droplets is governed by the Stockes law (Eq. 1.2):

(1.2) equation

Equation 1.2 Stockes Law: v, droplet speed of migration (m·s−1); g, gravity acceleration (m·s−2); Δρ, volumic density difference between continuous and dispersed phases (kg·m−3); r, droplet radius (m); ηc, dynamic viscosity of the continuous phase (Pa·s).

The rate of creaming or sedimentation is thus proportional to the radius square, and is thus dramatically reduced in case of nanometric droplets. In addition, flocculation, which may lead to the obtainment of particle aggregates of sizes favorable to this force, are not favored for such small sizes as adhesion decreases with the droplet radius [47, 48]. The Brownian motion therefore becomes predominant, favoring the dispersion homogeneity. The destabilization of such a system thus mainly relies on the irreversible mechanisms of Ostwald ripening and coalescence [29, 30, 35].

Ostwald ripening is defined by the growth of the largest droplets at the expense of the smallest ones. This phenomenon is due to the oil chemical potential difference between droplets of different sizes and thus different curvature radius. As the chemical potential increases when the radius decreases, because of the Laplace pressure (PLaplace ~ 1/r), the smallest droplets tend to give material to the largest ones by diffusion through the continuous phase. The time evolution of a droplets population undergoing Ostwald ripening is described by the LSW theory (Lifshitz-Slyozov-Wagner) [49, 50]. The droplet size increase is in this case proportional to the cube root of time and the size distribution is autosimilar (Eq. 1.3).

(1.3) equation

Equation 1.3 Time evolution of nanoemulsions size due to Ostwald ripening (LSW theory) with ω, the Ostwald ripening; and r, the droplets radius; C(∞) and Vm, being respectively the infinite solubility in the continuous phase and the molar volume of the component forming the dispersed phase; D and ρ, respectively being its diffusion coefficient and its density; finally R and T, respectively being the ideal gas sonstant and the temperature.

Coalescence occurs when two droplets collide and finally merge to become a larger droplet. In case of coalescence being the main destabilization phenomenon, the time evolution of the average droplet size can follow very different behaviors from perfectly homogeneous growth (monodal distribution whose average size increases with time) to strongly heterogeneous growth (plurimodal distribution with the possibility of very early phase separation). In the case of homogeneous growth, a mean field law can be used to describe the growth of the droplets [51–53] (Eq. 1.4). By integrating this equation, a decreasing linear variation of 1/r² as a function of time is obtained.

(1.4) equation

Equation 1.4: Time evolution of nanoemulsions size due to coalescence (homogeneous case). With r, the droplet radius, α, a geometric parameter; ω, the frequency of canal opening between 2 adjacent droplets per surface unit (fusion phenomenon at the origin of coalescence) et dr, the droplet size increment during the time interval dt.

In a nutshell, the small size of nanoemulsions protects them from undergoing reversible phenomena such as creaming and sedimentation. This small size also has a deep implication on the irreversible mechanisms of Ostwald ripening and coalescence. Indeed, the adhesion between droplets significantly decreases with their diameter [47, 48], therefore decreasing the time of contact when it occurs. The probability of coalescence is thus dramatically reduced for small droplets, especially when the interface is charged, favoring electrostatic repulsion, or covered by long hydrophilic polymer chains, favoring steric repulsion. Conversely, Ostwald ripening is favored by elevated Laplace pressure, which can reach several atmospheres for droplets of a few decades of nanometer (PLaplace ~ 1/r) [27, 29].

1.3.2.2 Ostwald Ripening as Main Destabilization Mechanism

An example can be given using a simple system composed by short triglycerides (hydrocarbon chains C8–C10) stabilized by a PEG40-stearate surfactant. Once formed, the system quickly undergoes destabilization which leads to an average droplet size increase. As shown in Figure 1.4a, the cube of the radius follows a linear increase over time, thus highlighting the predominant role of Ostwald ripening in the destabilization of this system. As previously demonstrated, Ostwald ripening highly depends on the temperature, following an Arrhenius law (Figure 1.4b) [36].

Figure 1.4 Classical nanoemulsion destabilization by Ostwald ripening: (a) size evolution along time; (b) Ostwald ripening rate as a function of temperature.

Therefore, for nanoemulsions, the most important destabilization phenomenon to overcome is related to Ostwald ripening. We will now describe a strategy that can be used to limit its practical implications on nanoemulsion stability thanks to the trapped species approach.

Ostwald ripening relies on material exchange between droplets through the continuous phase. The dispersed phase solubility in the continuous phase is thus of high importance. Reducing oil solubility in water can thus be envisaged to significantly lower Ostwald ripening. It can be achieved by using apolar oil and/or adding salt to the continuous phase. Another approach can nonetheless be dramatically more efficient while giving access to a larger stable formulation domain: entropic stabilization through trapped species [54–56]. This approach consists of adding insoluble species to the dispersed phase. Emulsions formed by a unique component A will evolve to a larger size via Ostwald ripening as long as A possesses even a very low solubility in the continuous phase. This is due to chemical potential differences between droplets of different sizes arising from Laplace pressure differences (Eq. 1.5a). When a second component B, insoluble in the continuous phase, is added to the dispersed phase, a new term appears in the difference of chemical potential (Eq. 1.5b) [54, 56–58]. This term is an entropic contribution of the A/B mixture favoring their mixture. As B is blocked inside the droplets, it will limit A transfer among droplets in order to preserve the A/B mixture. The previously unstable system can consequently acquire a true thermodynamic stability. This approach can be efficiently implemented by addition of a multitude of insoluble species in both the core (Fig. 1.5a) and at the droplet membrane (Fig. 1.5b) (Eq. 1.5c) [36]. In this case, numerous entropic terms are added to the equation, further improving the droplets stability (Eq. 1.5c). The example described in Figure 1.5 uses a complex wax (containing insoluble species such as long chain triglycerides) and lecithin as trapped species in the core and at the membrane respectively.

Figure 1.5 Nanoemulsion stabilization against Ostwald ripening: trapped species in core (a); and at the membrane (b) through addition of wax and lecithin, respectively.

(1.5)

equation

Equation 1.5 Chemical potential difference of an encapsulated component: case of a unique component (a), and cases of the addition of one (b) or multiple (c) insoluble species.

It is possible to obtain an order of magnitude of the required amount of insoluble species for droplet stabilization against Ostwald ripening by equaling the osmotic pressure to the Laplace pressure (Eq. 1.6).

(1.6) equation

Equation 1.6 Calculation of the concentration limit of insoluble species required to stabilize droplets against Ostwald ripening with R = 8,314 J.K−1mol−1, the ideal gas constant; γ, the surface tension; vm, the molar volume of the oil; r, the droplet radius and T, the temperature.

For a particle of 25 nm radius, presenting a characteristic surface tension of γ = 10−3 N.m−1 and a molar volume of encapsulated species vm = 1000 cm³, the molar fraction of insoluble species required to counter-balance the Laplace pressure is equal to 3.3 % à T = 298 K, which corresponds to the experimentally observed weight fraction of 10% w/w (Figure 1.5) [36].

Nanoemulsions are thus true emulsions of nanometric size, and are therefore thermodynamically metastable systems. This means that they require energy input for their formation and a specific strategy for their stabilization. We have shown that it is possible to: 1) produce nanoemulsions with a simple and easily up-scalable procedure relying on high energy processes such as ultrasonication or HPH, and 2) entropically stabilize the formed droplets through the addition of trapped species in both the core and at the membrane. Stabilization through specific membrane composition opens up new formulation opportunities by allowing fine tuning of the core composition for solubilization/release properties of the encapsulated active, but also for modulating the sensory profile of the final product. It is indeed well known that low molecular oils provide, for instance, a light and dry touch, whereas high molecular weight oils are usually experienced as heavy and greasy [59]. We will now investigate the accessible formulation domain of a nanoemulsion system designed with the aforementioned rules and explore its abilities for active encapsulation and release.

1.3.3 Nanoemulsion Formulation Domain

Understanding both nanoemulsion production and stabilization, we can now exploit the defined rules to evaluate the largest formulation domain giving small nanoemulsions presenting long-term stability.

1.3.3.1 Components Choice

The choice of the components used for nanoemulsion formulation has a deep influence on nanoemulsion stability, and active encapsulation and release, but also on its biocompatibility [60, 61]. The description of the investigated system is given in Figure 1.6.

Figure 1.6 Description of the model system (from ref [60]).

Aiming at biomedical applications, the component choice has been restrained to components already widely used and accepted for human application in both pharmaceutics and cosmetics [60]. A vegetable oil and a phospholipid surfactant have been chosen for their recognized biocompatibility. The addition of a semisynthetic wax allows for the obtainment of a core mixture of oil and wax, composed of saturated and unsaturated long chains of triglycerides, which ensures a limited solubility in the continuous water phase and therefore limits the degradation by Ostwald ripening. As will be shown, this liquid and solid lipids mixture also allows for fine-tuning the lipid core physical state, which provides an opportunity to modulate the actives encapsulation and release properties. Finally a hydrophilic co-surfactant is chosen among the PEG-stearate, aiming for a steric barrier preventing coalescence and favoring small particle diameters. The combination of PEG stearate and phospholipids surfactants at the membrane further improves stabilization of the droplets against Ostwald ripening.

1.3.3.2 Nanoemulsions Formulation

The most important parameters for controlling applications are the size and the stability of the formed nanoemulsions. All formulation parameters can in theory influence both parameters. A one-variable-at-a-time approach may be used to explore the formulation domain. Nonetheless, this empirical approach can be very time-consuming and can lead to misinterpretations. The methodology based on the design of experiments overcomes these issues and presents many more advantages [62, 63]. This approach maximizes the number and quality of obtained information, while minimizing the required experimental effort, explaining its growing use for both academic and industrial purposes. In addition, the obtained experimental design further allows particle optimization through questioning with specific requirements linked to size, stability, or others evaluated by physicochemical parameters.

That is why we exploited a specific design of experiment to investigate this system formulation domain [60]. Frontiers have been chosen to restrain the study to systems possessing a mixture of both surfactants (for stability purposes), over a domain allowing the highest encapsulation possibility (the biggest core), while keeping small nanoemulsions sizes. Formulation parameters (weight fractions of core, hydrophilic surfactant and lipophilic surfactants, and continuous phase) have consequently been simultaneously changed, and the effect on particle size, polydispersity, and also on the dispersion appearance (homogeneity, transparency) and viscosity have been followed (Figure 1.7a). At this stage the lipid core mixture has been fixed to 75% wax/25% oil. The Figure 1.7b details the results obtained for particle size. It highlights the accuracy of the model to a priori determine the size of the droplets formed on the size range 20–200 nm.

Figure 1.7 Formulation domain definition through an experimental design: (a). design of experiment definition; (b). correlation plot between experimental and predicted particle diameters (from ref [60]).

Important formulation parameters governing the droplet size have been shown to be: i) the relative viscosities of dispersed and continuous phase (here it mainly depends on the oil/wax mixture of the core), ii) the types and relative concentrations of both surfactants, and iii) the relative proportion of surfactants over the dispersed phase.

This model has finally been used for formulation optimization in order to obtain reproducible droplets population of a specific size and presenting long-term stability. Figure 1.8 presents the obtained monodisperse populations of standard size ranging from 30–120 nm (FXX, with XX defining the average particle size) and their subsequent stability. The use of the aforementioned rules for stability has allowed for the obtainment of very stable formulations, as these nanoemulsions present more than 18 months stability at 4°C, room temperature and 40°C.

Figure 1.8 Standard formulations: (a) size distribution (FXX, with XX the average size of the particles); and (b) accelerated stability (40°C).

Further formulation studies have additionally proven the long-term stability of nanoemulsions possessing different core composition. Nanoemulsions were thus seen to be stable over the 4–40°C range, whether their core ranged from pure oil (NCO) to pure wax (NC100) (Figure 1.9) [60]. Finally, increasing the dispersed phase weight fraction Φ allows for the modification of the dispersion viscosity from highly liquid to gel-like suspensions, while maintaining nanoemulsion stability (Figure 1.10) [60]. Similarly, nanoemulsions can be dispersed in aqueous gels or even biphasic systems, such as cream or macroscopic emulsion, without significant destabilization of the system.

Figure 1.9 Core composition modification: F50 standard formulation with varying wax/oil ratio (NCXX, with XX defines the %w/w of wax in the oil/wax mixture) and stability.

Figure 1.10 Dispersed phase weight fraction modification: (a) rheological behavior: photograph of samples of different viscosities depending on the dispersed phase weight fraction: from highly diluted suspension (left) to gel-like behavior (right); (b) accelerated stability (40°C); (b) nanoemulsion stability as liquid (Φ = 10% w/w) and gel-like (Φ = 40% w/w) forms.

Using the introduced stability rules, it is therefore possible to prepare stable nanoemulsions over a large formulation domain. Sizes can be varied from 20–200 nm by simple variation of the surfactant types, concentrations, and through the weight ratio of surfactants/core. The design of experiments allows for easy optimization of nanoemulsion formulation to obtain stable particles of different sizes and even under different galenic forms, from liquid to highly viscous solutions.

1.3.4 Conclusion on the Formulation of Stable Nanoemulsions

Nanoemulsions are thus true emulsions of nanometric size, and are therefore thermodynamically metastable systems. This means that they require energy input for their formation and a specific strategy for their stabilization. We have shown that it is possible to: 1) produce nanoemulsions with a simple and easily up-scalable procedure relying on high energy processes such as ultrasonication or HPH, and 2) entropically stabilize the formed droplets through the addition of trapped species in both the core and at the membrane. Stabilization through specific membrane composition opens up new formulation opportunities by allowing fine-tuning of the core composition for solubilization/release properties of the encapsulated active, but also for modulating the sensory profile of the final product. We will now evaluate the system’s ability to encapsulate actives molecules and control their release.

1.4 Nanoencapsulation in Lipid Nanoparticles

Once formulated, these nanocarriers can be used for application in biosciences, especially for actives encapsulation. Here we will detail the aim of such encapsulation and describe how the lipids physical state governs both active encapsulation and release properties. Particular emphasis will be put on amorphous solid lipid particles.

1.4.1 Aim of Active Encapsulation

The nanoencapsulation of actives for drug delivery has been a prolific domain of research since its first conceptualization by the Nobel Prize winner Paul Ehrlich at the end of the 19th century. He was then dreaming about the magic bullet that could deliver a drug specifically to its site of biological action. The practical implementation of this idea was made possible with the advent of nanotechnology and the possibility of producing nanoparticles capable of delivering, targeting, and releasing the active ingredients to their place of biological activity.

This new approach thus relies on the nanoencapsulation of actives inside nanoparticles in order to: 1) protect, 2) target, and 3) increase the active ingredients intrinsic properties to significantly improve treatment efficiency while reducing their possible undesirable side effects (Figure 1.11). To better understand the interest of such an approach let’s detail the objectives of the ideal nanoparticulate carrier:

Figure 1.11 Aims of active encapsulation in drug delivery systems.

Protection: Molecular encapsulation inside nanocarriers allows physically separating the active ingredients from direct contact with the biological medium, therefore limiting their interactions [64–67]. This first allows limiting active degradation by the biological medium through chemical (pH…) or enzymatic (hydrolases, lipases…) means. This also allows limiting the possible toxicity of active molecules on the living environment before having reached its biological target.

Targeting: Each active ingredient possesses a biological target it must reach to perform its therapeutical action. The molecule of interest must therefore rejoin its biological target from its place of administration. This implies passing the numerous biological barriers and concentrating at the pathological site. To fulfill this aim, passive and/or active strategies can be implemented [68–71]. The nanometric size of the carriers generally facilitates passways through the different biological barriers and the passive targeting of actives to their site of activity. In parallel, this passive targeting can be significantly improved by active targeting properties through nanoparticles surface grafting of molecules or biomolecules (antibody, peptides, sugar…) that possess recognition abilities for certain cellular receptors [64–66]. The attachment of nanoparticles to the target cells can therefore be facilitated thanks to the biological action of these surface molecules, thus improving their attachment or even internalization inside those cells.

Improvement of the activity: This last aim gathers all methods that allow for the improvement of the therapeutic efficiency of the active molecule, or the imaging properties of a contrast agent for diagnosis. The first objective of protection already enables keeping the original properties of the encapsulated molecules to their biological site of action by avoiding their degradation. Nonetheless, nanoencapsulation allows going even further. Indeed, numerous molecular actives often possess limited aqueous solubility challenging their formulation in aqueous systems. The use of nanoparticulate carriers can enable encapsulating very hydrophobic species; their solubilization in aqueous media therefore rely on the particles’ surface properties which are easily tunable [64, 66, 67, 72]. The therapeutic efficiency of a molecular active can also be dramatically improved through nanoformulation, since this can allow modulating active release kinetics and subsequent biodisponibility [4]. Finally, as will be further detailed later on, the nanoformulation of contrast agent may significantly improve their diagnosis efficiency through contrast enhancement and better accumulation at the pathological site [73].

Originally developed for pharmaceutics, especially for cancer treatment, this approach can similarly be developed to increase treatment efficiency through the dermic route, and can therefore be implemented to control the delivery of pharmaceutical or cosmetic actives by topical application.

1.4.2 Lipid Complexity and Influence of Their Physical State

When using lipids as core matrix for nanocarriers, great importance has to be put on understanding their physical state, since it can have a dramatic influence on particle properties and subsequent active encapsulation/release properties. As most compounds with long alkyl chains, lipids can crystallize in different forms [74]. These different polymorphs result from various accessible spatial arrangements of the lipid molecules. Three different polymorphs are classically described for lipids: the α, β and β’ polymorphs (Figure 1.12). The a polymorph corresponds to the less organized arrangement, and the β polymorph to the most ordered and the thermodynamically favored state (Fig. 1.12). Lipid crystallization generally starts by an α organization, but then evolves toward the thermodynamically favored β form, passing through an intermediate β’ form (Fig. 1.12) [75]. These different polymorphs can present highly different physicochemical properties, for instance, in terms of temperature and enthalpies of crystallization and fusion, or in terms of solubility [75]. This behavior complexity is even more pronounced in the case of lipid mixtures, and can lead to the advent of very long kinetics of polymorphic transitions [74–76].

Figure 1.12 Schematic representation of the typical molecular arrangements associated with triglycerides polymorphs (adapted from [19]): triglycerides tend to crystallize, passing from the low stable α form, to the thermodynamically favored β form, after a possible intermediate β’ polymorphic state.

The formulation of crystalline lipids as nanoparticles complicates even more these behaviors because of large surface effects that could become predominant [13, 15, 77, 78]. Crystalline lipids may stay under a liquid supercooled state during weeks or even months when they are nanoformulated [15, 20, 79]. Yet this amorphous state generally evolves towards crystalline states that may further differentiate through polymorphic transformation(s) [15]. The mixture of chemically highly different and spatially incompatible lipids may lead to the obtainment of very different structures as previously described for the NLC. Non-perfectly crystallized states are generally obtained. These various accessible lipid physical states and the subsequent lipid nanoparticles crystallinity are of critical significance when considering possible applications. They can indeed directly influence not only the dispersion stability and the encapsulation/release properties of the encapsulated molecules, but also the interactions between nanoparticles and biological medium [8, 14, 80].

Stability: As previously introduced, liquid lipid nanospheres present numerous destabilization pathways, especially by Ostwald ripening and coalescence [29, 30, 81]. Lyophilisation may allow stabilizing these particles through water removal. Nonetheless, this technique complicates the fabrication procedure and can lead to new stability issues during the particles resuspension step [82–85]. The solid state of lipids may enable limiting destabilization by Ostwald ripening in solution, but may destabilize the colloidal system by gelation. Indeed, lipid crystallization generally leads the lipid molecules to pass from a random isotropic organization, favoring spherical nanoparticulate form, to a well-ordered anisotropic organization, leading to platelet-like nanoparticles [14, 86]. This morphological change is generally associated with a dramatic increase of the particles’ specific surface area, the spherical form being the geometric morphology of lowest surface/volume ratio. As a consequence, the surface density of surfactants may be dramatically lowered if not enough free surfactant is available in solution [14]. This favors irreversible contacts between particles, ultimately leading the system to form a gel in place of the original liquid dispersion [8, 14].

Encapsulation/Release: The core crystallinity also directly influences the encapsulation and release of active molecules. A solid matrix is necessary to promote slow diffusion kinetics of the encapsulated molecules from the particles to the surrounding continuous medium. Nonetheless, depending on the overall system composition (lipids, surfactants, active molecule), the lipids molecular arrangement as crystalline forms may lead to active molecules/lipids phase separation, leading to structures with en heterogeneous active distribution as enriched core or membrane (Figure 1.13) [80, 87]. The lipids crystalline arrangement indeed considerably reduces the free volume, subsequently dramatically limiting the insertion of active molecules in between the lipid molecular network. This generally leads to the expulsion of the foreigner active molecules outside of the particles, then being adsorbed on the particle surface or dispersed as free molecular aggregates in solution [87]. This consequently dramatically lowers the effective accessible active loading ratio and the associated release behaviors [8, 80, 87]. As aforementioned, this is the reason why much attention has now been gained by NLC following the initial SLN development.

Biological properties: Finally, the lipid physical state may also influence the particles biological properties in terms of enzymatic degradation kinetics, passage through the biological barriers, cellular internalization or toxicity [61]. Lipid crystallinity for instance generally slows down the kinetics of enzymatic degradation [88–91]. That is why it is highly important to fully characterize the internal physical state of the nanoformulated lipids and the limits of this state.

Figure 1.13 Possible active/lipid molecules distribution inside lipid nanospheres.

As a consequence, nanoformulating lipids under a solid amorphous state may represent the best option to get stable and flexible active molecule encapsulation, while preserving ability to control active release. We will therefore explore the nanoformulation of amorphous lipids, exploit the obtained lipid nanoparticles for active encapsulation, and determine what parameters control active release.

1.4.3 Amorphous Lipids for a Large Range of Encapsulated Molecules

We have just shown that solid amorphous lipid particles may represent the best candidates for control over active encapsulation and release. Nonetheless, the thermodynamically favored state for most lipids is the crystalline state for solid systems. How can we thus reproducibly obtain stable amorphous solid lipid nanoparticles for stable encapsulation of a large range of molecules?

1.4.3.1 How to Favor Lipids Amorphicity

Crystallization starts by the nucleation of a first nucleant which then grows and evolves till the growth of a crystal. Nucleation could occur through two different mechanisms: i) homogeneous nucleation, and ii) heterogeneous nucleation. Let’s thus explore the role of nucleation on lipid nanoparticles crystallization.

In the homogeneous case, nucleation starts by the direct local organization of lipid molecules, leading to the formation of a first nucleant that ultimately evolves toward the formation of a crystal. As previously introduced, nano-compartmentalization first helps by preventing lipid crystallization through the increased importance of surface effects and the Kelvin effect. The Kelvin effect corresponds to the decrease of probability to get nucleants in the particle core whenever its diameter decreases. This effect has been well documented, but the obtained supercooled lipid nanoparticles do generally crystallize after a few days or weeks at 4°C or even at room temperature [13, 24, 92]. To further decrease homogeneous nucleation complex lipid mixtures need to be used. Lipids purity is indeed of high importance since the lipids chemical structure’s homogeneity and similitude favors their molecular arrangement, therefore favoring the apparition of order and their crystallization. Figure 1.14 shows the importance of lipids complexity in avoiding crystallization: the use of pure and well-defined triglycerides (triglycerides with an alkyl chain length of C14 (Dynasan 114) or C16 (Dynasan 116)) directly leads the system to crystallize (Fig. 1.14b;c), whereas in similar conditions, the complex mixture of a wax (mono-, di- and triglycerides of alkyl chain length varying from C6 to C18) is still under an amorphous state (Fig. 1.14a). The different endothermic events on the thermogramms correspond to the different lipid polymorphs: the 43–44°C peak corresponds to the less ordered α state, while the more ordered and thermodynamically favored β state leads to a peak at higher temperature.

Figure 1.14 Role of homogeneous and heterogeneous nucleation on lipid crystallinity: DSC thermogramms on lipid nanoparticles after 2 weeks storage at 4°C for NC100 particles (pure Suppocire NC particles) (a); C14 triglycerides (Dynasan 114) (b); C16 triglycerides (Dynasan 116) (c), and; NC100 particles with hydrogenated phospholipids at the membrane (d).

In the heterogeneous case, nucleation is favored by the introduction of foreign molecules in the lipid mixture that lower the activation barrier leading to the nucleants formation. This heterogeneous nucleation can thus start from the presence of impurities in the lipid mixture or by the contact of membrane molecules when formulated as nanoparticles. Figure 1.14 highlights, for instance, the role of the phospholipid types on lipids crystallization. The use of hydrogenated phospholipids (phospholipids with hydrogenated alkyl chains), for instance, favors crystallization (Fig. 1.14d), whereas similar particles with nonhydrogenated phospholipids (unsaturated alkyl chains) allow keeping amorphous lipids (Fig. 1.14a). Hydrogenated phospholipids therefore act as nucleants, favoring the heterogeneous nucleation and subsequent crystallization of the internal lipids [79, 92].

Following these rules, it is therefore possible to prepare highly stable amorphous lipid nanoparticles from wax/oil mixtures (Figure 1.15). In the described system, the used lipids are a wax (Suppocire NC, Gattefossé, France) and an oil (Super Refined Soybean oil, Croda, France). At room temperature the wax is a semi-crystalline solid whereas the oil is a liquid. Yet, the oil undergoes a partial crystallization when cooled down below 10°C. Nonetheless, once nanoformulated, no thermal event is observed on the nanoemulsion DSC thermogramms, highlighting the amorphous state of the lipids (Fig. 1.15). A similar observation is made after 8 months of storage whatever the temperature (4°C or room temperature) and whatever the core composition (oil/wax ratio) (Fig. 1.15) [10, 60]. As explained, both nanocompartmentalization and high lipid complexity are the keys to preventing crystallization. Mixing oil and wax dramatically increase the lipids chemical heterogeneity hindering crystallization through ordering. The wax is indeed a complex mixture of mono-, di- and triglycerides of various alkyl chain lengths (C8 to C18), with an overall hydroxyl value of 20–30%. Short alkyl chain lengths are known to disfavor crystallization, presenting more free ends enabling movement. In addition, a high amount of partial glycerides, corresponding to a high hydroxyl value, is known to lead to less crystalline material [76]. Finally, the unsaturated alkyl chains present in soybean oil also tend to disfavor ordering by double-bond blocked cis/trans conformations [75].

Figure 1.15 Evaluation of nanoparticle crystallinity: DSC thermogramms of macroscopic lipids (oil and wax) and lipid nanoparticles right after preparation and after a 8 months storage at room temperature for pure oil particles (NCO), a mixture of 5°% oil/50% wax (NC50) and pure wax particle (NC100).

It is therefore possible to prepare stable amorphous lipid particles by using nanocompartimentalisation and complex lipid mixtures (glycerides of different hydroxylation state (mono-, di- and triglycerides) and alkyl chain lengths (alkyl chains varying from C6 to C18)), and additionally avoiding heterogeneous nucleation by membrane molecules. An example is here given by the use of highly complex lipid mixtures, its nanoformulation as 30–120 nm particles, and the use of non-hydrogenated phospholipids.

1.4.3.2 Amorphous Lipid Nanoparticles for Stable Active Molecules Encapsulation

Using these amorphous lipid nanoparticles it is thus possible to encapsulate a wide range of molecules without lipids crystallization and expulsion of the encapsulated molecules (Figure 1.16). The loading efficiency of a molecule mainly relies on its solubility on the lipid mixture. The more lipophilic an active, the highest its solubility in the lipids and the highest the maximum loading ratio is, without active phase separation and/or expulsion from the nanoparticles [10]. Highly lipophilic compound such as DiO, DiI, DiD fluorophores (log P = 10–12) [73,