3.1. Influence of Phospholipid Premix Composition on Foaming Properties
All manufactured premixes were turbid liquid dispersions with a water-like consistency. Blends with varying concentrations (0.25%, 0.50% and 1.00%), as well as different ratios of PL1:PL2 (3:1, 1:1 and 1:3), of two phospholipids were prepared. The most suitable total phospholipid concentration and PL1:PL2 ratio for each premix has been determined in a previous study (data not shown).
Table 2 shows a summary of the investigated premixes.
Figure 4 shows the total height of the foamed premixes and its change over time. The negligible differences in the initial foam height (h(t
0)) for all formulations indicate that the foamability is neither influenced by the composition nor the concentration of the phospholipid blend used as a foaming agent.
Most premixes show a minute increase in height during the measuring time. This observation can potentially be explained by an expansion of the individual bubbles due to the internal pressure acting against a decreasing strength of the foam bubble membranes as a result of liquid drainage [
22]. In combination with limited decay, this could lead to a slight increase in the total volume over the course of a measuring time as short as 5 min. The only exception from this general behavior was observed with Premix 2, which was the only premix that contained predominantly LPC90 (75% of the phospholipid blend). It showed a slight decrease in total height over the course of the measurements. Obviously, here, the decay was marginally larger than the post-expansion of the compressed air after the foaming process. However, in summary, all premixes allow to form stable foams.
The foam structure analysis (FSA) is used to determine the structural attributes of foamed formulations and their progression over time. It is conducted with the foam structure module (FSM) of the DFA100, which includes a secondary illumination source, as well as a high-resolution camera and a modified glass column with an attached prism [
23]. Foam structure is expressed as the bubble count per mm
2 (BC) and the mean bubble area (MBA).
Figure 5 depicts these parameters for the foamed premixes over the course of a measurement.
Premix 3 and Premix 5 yielded highly dispersed foams, as indicated by the high bubble count and relatively small mean bubble area in comparison to the other formulations. Both formulations also showed superior foam stability, as changes in either parameter over time were less pronounced than in the rest of the examined formulations, as displayed in
Table 3. While Premixes 3 and 5 were able to maintain 61.93% and 51.69% of their initial bubble count, respectively, the foam structure parameters of the other premixes showed substantial decay. This was most pronounced with Premix 1 maintaining only 23.90% of its initial bubble count after 300 s.
The addition of PL80H, a hydrogenated PL, to either of the LPC compounds (Premix 2 and Premix 4) impaired their foamability, becoming obvious by the lower initial BC and higher initial MBA. Additionally, Premix 1 consisting of phosphatidylcholine (PL90G) and lysophosphatidylcholine (LPC90) displayed relatively low foam dispersity and foam structure stability. Therefore, it can be concluded that blends that include predominantly LPC20 with an addition of a non-hydrogenated secondary phospholipid (LPC90 or PL90G) lead to more disperse foams that can maintain their foam structure better than other premixes. The presence of hydrogenated lysophosphatidylcholine (hLPC) appears to play an important role in producing these results, as neither of the premixes that only included non-hydrogenated lysophosphatidylcholine (LPC) (Premixes 1 and 2) displayed similarly high foam dispersity or stability. This observation is in line with the increased elasticity of LPC-containing phospholipid membranes [
24], since film elasticity plays a vital role in the stabilization of foams [
25]. Taking all of the previously mentioned observations into account, Premixes 3 and 5 are the most suitable candidates for the development of a foamable phospholipid emulsion.
3.3. Influence of the Cosurfactants and Polarity of the Oil Phase on the Foaming Properties of Phospholipid Emulsions
The results of
Section 3.2 show that the incorporation of an additional surface-active component is beneficial in order to obtain stable, foamable emulsions based on phospholipid Premixes 3 and 5 discussed in
Section 3.1. Further systematic experiments use these two premixes and three different oil phases representing a broad range of polarities, namely paraffin oil (PO), medium-chain triglycerides (MCT) and castor oil (CO). The polarity of these oil phases increases in the order: paraffin oil < medium-chain triglycerides < castor oil. Moreover, the emulsions were supplemented with two classical foam surfactants (nonionic Lauryglucoside (LG); zwitterionic Lauramidopropyl betaine (LAPB)).
Table 4 lists all the investigated emulsions.
Figure 7 shows the total height of the foamed formulations and its change over time.
The initial foamability of the formulations was neither influenced by the presence of an oil phase nor the use of a cosurfactant, as indicated by the parameter h(t0) that is almost identical in all investigated emulsions and premixes.
However, in contrast to the aqueous phospholipid premixes, some of the corresponding emulsions display a much lower foam stability. This could have been expected, as the presence of oily substances is known to impede the formation of stable foam structures [
26]. This was especially true for emulsions that included LAPB as a zwitterionic cosurfactant, as indicated by Emulsions 4, 6 and 8. Even worse was the stability of the foam from Emulsion 2, where the foam almost completely decayed after 300 s.
On the other hand, the corresponding formulation variants using LG as a nonionic cosurfactant showed far superior foam stability in comparison. Emulsions 5, 7, 9 and 11 in particular displayed good foam stability close to the level of the pure phospholipid premixes.
Figure 8 shows the respective results for the foam structure analysis. Obviously, the used phospholipid premix largely affects the foam structure. Interestingly, the formulations that were based on Premix 3 (Emulsions 1–6) yielded highly disperse foams whenever LG was used as a cosurfactant, and in contrast, foam formation was almost impossible when LAPB was used as a cosurfactant. In contrast, Premix 5-based emulsions were not influenced by the type of cosurfactant in a similar way.
Emulsions 1, 3 and 9 performed best with respect to the foam structure with Emulsion 9 showing the most reproducible results.
Table 5 summarizes the results of the foam structure analysis for these emulsions and the corresponding premixes.
This data shows that the differences in foam structures between the emulsions and their respective premix formulations become more pronounced over the course of a measurement, indicating that the presence of an oil phase influences the foam stability more severely than the initial foam structure.
PO emulsions (1, 2, 7 and 8) displayed similar results to the respective MCT emulsion (3, 4, 9 and 10), with the latter being the more reproducible choice. CO emulsions (5 and 11) using LG as a cosurfactant generally were capable of producing foams of good overall quality in regard to their stability and structural attributes. This indicates that the selection of a medium- to high-polarity oil phase was beneficial for the foam quality of these formulations. However, macroscopically, the CO emulsions were far less stable than most of the PO and MCT emulsions and showed creaming within hours after the preparations. The droplet size distribution of an emulsion is directly related to its creaming rate.
Table 6 shows the results of the droplet size measurements of the tested emulsions.
None of the investigated CO approaches (Emulsions 5, 6, 11 and 12) showed sufficiently small droplets (mean droplet size: 20.165 ± 3.900 µm) for the formulation to be considered stable.
On the other hand, approaches that employed low-polarity PO (Emulsions 1, 2, 7 and 8) and medium-polarity MCT (Emulsions 3, 4, 9 and 10) yielded satisfying results. With a mean droplet size of 3.042 ± 0.344 µm across all formulations, the PO emulsions showed higher values than the MCT emulsions, with a mean droplet size of 2.055 ± 0.223 µm. The same observations applied for the d10 and d90 values.
The respective LAPB and LG approaches of each premix/oil combination displayed very similar results. It can therefore be assumed that the type of cosurfactant does not influence the droplet size distribution and, therefore, the stability of the emulsions in a similar way as the type and the polarity of the oil phase does.
Figure 9 shows the droplet size distribution of Emulsion 9 over a storage period of 12 weeks at 25 °C in the dark.
During this period, the droplet sizes are unaffected, as indicated by the near-constant results for the d10, d50 and d90 values and the overall volume-based droplet size distribution. It can therefore be assumed that the investigated emulsions consisting of Premix 5, LG and MCT remain physically stable for a period of at least three months when stored at room temperature.
Characterization of Liquid–Air Interface
According to Gibbs Equation (1), a system with a higher dɣ (lower surface tension compared to water) displays a higher concentration of adsorbed surface-active molecules in the adsorption layer, thus indicating higher activity of the foaming agent(s) at the liquid–air interface [
1,
27].
Γ—Surface concentration of a foaming agent (mol/m2);
ɣ—Surface tension of the liquid (N/m);
c—molar concentration of the foaming agent (mol/L);
R—gas constant (J/mol*K);
T—temperature (K).
The liquid–air interface is of particular interest when dealing with foam generation and foam stability [
28]. Surface tension and its variations during formulation optimization can give the first hints on the suitability of a formulation approach [
29].
To further study the roles of the various surface-active ingredients that are part of the investigated premix and emulsion formulations, surface tension experiments by means of a drop profile analysis using the pendant drop technique were conducted.
In order to acquire comparable results, Premixes 3 and 5 and the corresponding emulsion approaches (Emulsions 7, 9, 10 and 11), as well as aqueous LG (0.5% (w/w)) and LAPB (0.5% (w/w)) solutions, were included in these experiments.
Figure 10 depicts the extrapolated equilibrium surface tension of these samples. Generally, the pure aqueous cosurfactant solutions showed the lowest surface tensions of the investigated samples, while the PL premixes showed the highest values. The surface tension for the premix/cosurfactant mixtures was almost identical to the surface tension of the corresponding MCT and CO emulsions and by far closer to the cosurfactant solution than to the PL premix. Premix 5, for example, showed a high surface tension value of 36.7 mN/m, while the LG solution lowered the surface tension to 27.6 mN/m. The mixture of both yielded a surface tension of 28.3 mN/m, which was much closer to the pure LG solution. Obviously, LG with its more pronounced interfacial activity dominated the behavior of the surfactant/phospholipid blend. This suggests a preferred presence of LG at the liquid–air interface, leading to enhanced foaming properties of the formulations.
This assumption is underlined by
Figure 11, which shows polarized light microscopic images of foams generated from the LG/Premix 3 and LG/Premix 5 blends, indicating the presence of liquid–crystalline LG structures at the plateau border region and foam lamellae.
Consequently, the phospholipids are likely to be associated with the stabilization of the emulsion. This is also in accordance with the vastly different foaming behaviors of the LG and LAPB approaches discussed in
Section 3.3.
Emulsion 7 (28.2 mN/m) and Emulsion 9 (28.0 mN/m) displayed the lowest surface tension among the emulsions, with the respective PO (30.3 mN/m) and LAPB (30.0mN/m) approaches showing noticeably higher values. This could partially explain the more favorable foaming properties of Emulsion 9 when directly compared to Emulsions 7 and 10 because of the tendency of systems with lower surface tensions to produce more stable foams, according to Gibbs equation.
The high stability of the foams generated from premixes regardless of the higher surface tension can be explained by stabilization through solid particles present in the premix formulation [
30], as well as the possible existence of mixed micellar structures consisting of LPC and PC in such systems [
31].
Figure 12 depicts dried-up foam generated from Premix 5. It can clearly be seen that the phospholipids present in the premix form a particulate residue located where the lamellae of the foam were located. This indicates that solid particles from the premix primarily aggregate at the liquid–air interface, leading to more rigidity and further promoting foam stabilization [
32,
33,
34].