Surfactant-laden Liquid Crystal Interfaces

Fig. 1: Schematic experimental setup for ellipsometric studies of liquid-crystal/water interfaces. A laser beam is reflected by the interface and the corresponding change of the state of polarization, described by the two parameters Δ and Ψ, is detected. The results yield information about the refractive index profile of the interface.

Surfactant-laden interfaces between thermotropic liquid crystals and aqueous phases enable studies of the surface-induced liquid crystal order under a systematic variation of the strength of the ordering surface field. The surface field is controlled via the molecular structure of the surfactant and the magnitude of the surfactant coverage of the interface. Our group performs ellipsometric studies in order to elucidate the liquid crystal surface order near nematic - isotropic and smectic - isotropic transitions.

Fig. 2: Temperature dependence of the ellipticity coefficient ρ at surfactant-laden 8CB/water interfaces in the vicinity of the nematic − isotropic bulk transition temperature Tb. The concentration of the surfactant CTAB in the aqueous phase amounts to 0.8 μM, 0.7 μM, 0.6 μM, and 0.5 μM (from top to bottom). Small blue dots are experimental values, black open circles correspond to calculated values resulting from a Landau-de Gennes model, insets show the temperature dependence of the Brewster angle θB. Whereas the data at higher surfactant concentration (top figure) show a complete wetting by a more ordered (nematic) phase as Tb is approached from above, the data at low surfactant concentration (bottom figure) indicate that the bulk nematic/water interface is wetted by a less ordered (nearly isotropic) phase as Tb is approached from below. — Fig. 2: Temperature dependence of the ellipticity coefficient ρ at surfactant-laden 8CB/water interfaces in the vicinity of the nematic − isotropic bulk transition temperature Tb. The concentration of the surfactant CTAB in the aqueous phase amounts to 0.8 μM, 0.7 μM, 0.6 μM, and 0.5 μM (from top to bottom). Small blue dots are experimental values, black open circles correspond to calculated values resulting from a Landau-de Gennes model, insets show the temperature dependence of the Brewster angle θB. Whereas the data at higher surfactant concentration (top figure) show a complete wetting by a more ordered (nematic) phase as *T_b* is approached from above, the data at low surfactant concentration (bottom figure) indicate that the bulk nematic/water interface is wetted by a less ordered (nearly isotropic) phase as *T_b* is approached from below.

Fig. 2: Temperature dependence of the ellipticity coefficient ρ at surfactant-laden 8CB/water interfaces in the vicinity of the nematic − isotropic bulk transition temperature Tb. The concentration of the surfactant CTAB in the aqueous phase amounts to 0.8 μM, 0.7 μM, 0.6 μM, and 0.5 μM (from top to bottom). Small blue dots are experimental values, black open circles correspond to calculated values resulting from a Landau-de Gennes model, insets show the temperature dependence of the Brewster angle θB. Whereas the data at higher surfactant concentration (top figure) show a complete wetting by a more ordered (nematic) phase as *T_b* is approached from above, the data at low surfactant concentration (bottom figure) indicate that the bulk nematic/water interface is wetted by a less ordered (nearly isotropic) phase as *T_b* is approached from below.

Nematic wetting of isotropic liquid crystal/aqueous interfaces

Surfactants possessing the usual structure (polar head, nonpolar tail) induce a strong homeotropic anchoring at liquid crystal/aqueous interfaces, provided the coverage of the interface is large enough. At the interface, an ordering surface field exists which leads to the formation of a thin nematic film even at temperatures where bulk liquid crystal is in the isotropic state. When the temperature approches the bulk transition to the nematic phase, the thickness of the nematic film diverges, i. e., the isotropic liquid crystal/aqueous interface is completely wetted by the nematic phase. Decreasing the surfactant coverage results in a decrease of the ordering surface field, and the wetting behavior changes first from complete to partial and finally to a nonwetting situation. This behavior is demonstrated in the following figure showing ellipsometric results for the 8CB/CTAB/water system.

Find more information:

Surfactant-induced nematic wetting layer at a thermotropic liquid crystal/water interface
Ch. Bahr, Phys. Rev. E 73, 030702(R) (2006).
DOI: 10.1103/PhysRevE.73.030702

Cross-over in the wetting behavior at surfactant-laden liquid-crystal/water interfaces: experiment and theory
E. Kadivar, Ch. Bahr, and H. Stark, Phys. Rev. E 75, 061711 (2007).
DOI: 10.1103/PhysRevE.75.061711

Fig. 3: Ellipticity coefficient ρ as a function of the temperature difference to the smectic-A − isotropic bulk transition temperature Tb at 12CB/water interfaces. The 12CB volume phase is doped with a small amount of the nonionic surfactant monoolein, the values xs give the mol fraction of the surfactant in the bulk liquid-crystal phase. The data for xs = 0.0043 indicate a 0↔2-layer transition and for xs = 0.0040 a 0↔3-layer transition (the inset displays the same data on an expanded temperature scale). — Fig. 3: Ellipticity coefficient ρ as a function of the temperature difference to the smectic-A − isotropic bulk transition temperature T_b at 12CB/water interfaces. The 12CB volume phase is doped with a small amount of the nonionic surfactant monoolein, the values x_s give the mol fraction of the surfactant in the bulk liquid-crystal phase. The data for x_s = 0.0043 indicate a 0↔2-layer transition and for x_s = 0.0040 a 0↔3-layer transition (the inset displays the same data on an expanded temperature scale).

Fig. 3: Ellipticity coefficient ρ as a function of the temperature difference to the smectic-A − isotropic bulk transition temperature T_b at 12CB/water interfaces. The 12CB volume phase is doped with a small amount of the nonionic surfactant monoolein, the values x_s give the mol fraction of the surfactant in the bulk liquid-crystal phase. The data for x_s = 0.0043 indicate a 0↔2-layer transition and for x_s = 0.0040 a 0↔3-layer transition (the inset displays the same data on an expanded temperature scale).

Smectic layering transitions at isotropic liquid crystal/aqueous interfaces

Whereas the nematic surface order increases continuously on approaching the transition to the bulk nematic phase, a smectic surface film increases via a series of layering transitions. Usually, the thickness of the smectic film increases at each layering transition by the formation of an additional single smectic layer. The precise control of the surface field enables the confirmation of long-standing theoretical predictions concerning the occurrence of multiple-layer transitions in the regime of small surface fields.

Find more information:

Surface triple points and multiple-layer transitions observed by tuning the surface field at smectic liquid crystal/water interfaces
Ch. Bahr, Phys. Rev. Lett. 99, 057801 (2007).
DOI: 10.1103/PhysRevLett.99.057801

Fig. 4: Temperature dependence of the ellipticity coefficient ρ of a surfactant-laden liquid-crystal/water interface in the temperature range above the nematic - isotropic transition temperature TNI; ρ is a linear measure of the thickness of the nematic layer at the interface. The curves are obtained with different surfactant concentrations (the number at each curve gives the molar fraction in 10-3) in the liquid crystal volume phase. The discontinuous jump observed for low concentrations indicates a first-order prewetting transition at which the thickness of the nematic wetting layer changes abruptly (see schematic drawings in which the nematic layer is marked in yellow). With increasing surfactant concentration, the prewetting transition is driven towards a critical point beyond which the wetting layer thickness varies continuously. — Fig. 4: Temperature dependence of the ellipticity coefficient ρ of a surfactant-laden liquid-crystal/water interface in the temperature range above the nematic - isotropic transition temperature *T_NI*; ρ is a linear measure of the thickness of the nematic layer at the interface. The curves are obtained with different surfactant concentrations (the number at each curve gives the molar fraction in 10^-3) in the liquid crystal volume phase. The discontinuous jump observed for low concentrations indicates a first-order prewetting transition at which the thickness of the nematic wetting layer changes abruptly (see schematic drawings in which the nematic layer is marked in yellow). With increasing surfactant concentration, the prewetting transition is driven towards a critical point beyond which the wetting layer thickness varies continuously.

Fig. 4: Temperature dependence of the ellipticity coefficient ρ of a surfactant-laden liquid-crystal/water interface in the temperature range above the nematic - isotropic transition temperature *T_NI*; ρ is a linear measure of the thickness of the nematic layer at the interface. The curves are obtained with different surfactant concentrations (the number at each curve gives the molar fraction in 10^-3) in the liquid crystal volume phase. The discontinuous jump observed for low concentrations indicates a first-order prewetting transition at which the thickness of the nematic wetting layer changes abruptly (see schematic drawings in which the nematic layer is marked in yellow). With increasing surfactant concentration, the prewetting transition is driven towards a critical point beyond which the wetting layer thickness varies continuously.

Nematic prewetting transitions

Prewetting transitions are generally expected for all systems showing a first-order wetting transition. A prewetting transition should be observable as a discontinuous change of the wetting layer thickness occurring before the bulk transition is reached. For most experimental systems, such a behavior is difficult to observe because the ordering surface field must be tuned such that the system is in close vicinity to the wetting transition. The precise control of the surface field at surfactant-laden isotropic liquid crystal/aqueous interfaces enables the experimental observation of a prewetting transition in a nematic wetting layer.

Find more information:

Experimental study of prewetting transitions by systematic variation of the surface field at nematic liquid crystal/water interfaces
Ch. Bahr, EPL 88, 46001 (2009).
DOI: 10.1209/0295-5075/88/46001

Interfaces to liquid phases with different polarities

The surfactant coverage at the interface between two immiscible liquids is controlled not only by the surfactant concentration in the bulk phase(s) but also by the difference between the polarities of the two liquids. The polarity difference determines the affinity of the surfactant molecules, which usually possess a polar and a nonpolar part, to the interface. Thus, the variation of the polarity of a liquid phase at constant surfactant concentration, should have the same effect as the variation of the surfactant bulk concentration. We have experimentally demonstrated this behavior by studying the smectic layering transitions and nematic prewetting transitions at interfaces between isotropic liquid crystals, doped with a small constant amount of the surfactant monoolein, and various water/glycerol mixtures with different glycerol concentration.

With increasing glycerol concentration, the static permittivity of the water/glycerol mixture decreases, resulting in a decrease of the monoolein coverage of the interface. Thus, we observe the same effect as we would have decreased the monoolein bulk concentration:

Fig. 5: Influence of a variation of the glycerol content on smectic layering transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 12CB (doped with monoolein, mole fraction xs = 0.016) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. Tb denotes the bulk smectic-A - isotropic transition temperature of the sample. The stepwise increase of ρ with decreasing temperature indicates the successive formation of molecular smectic layer. With increasing glycerol content the layering transitions shift towards the bulk smectic-A - isotropic transition. The same behavior is observed when the bulk concentration of monoolein is decreased. — Fig. 5: Influence of a variation of the glycerol content on smectic layering transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 12CB (doped with monoolein, mole fraction x_s = 0.016) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. T_b denotes the bulk smectic-A - isotropic transition temperature of the sample. The stepwise increase of ρ with decreasing temperature indicates the successive formation of molecular smectic layer. With increasing glycerol content the layering transitions shift towards the bulk smectic-A - isotropic transition. The same behavior is observed when the bulk concentration of monoolein is decreased.

Fig. 5: Influence of a variation of the glycerol content on smectic layering transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 12CB (doped with monoolein, mole fraction x_s = 0.016) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. T_b denotes the bulk smectic-A - isotropic transition temperature of the sample. The stepwise increase of ρ with decreasing temperature indicates the successive formation of molecular smectic layer. With increasing glycerol content the layering transitions shift towards the bulk smectic-A - isotropic transition. The same behavior is observed when the bulk concentration of monoolein is decreased.

Fig. 6: Influence of a variaition of the glycerol content on nematic prewetting transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 9CB (doped with monoolein, mole fraction xs = 0.005) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. Tb denotes the bulk nematic - isotropic transition temperature of the sample. With increasing glycerol content a first-order prewetting transition emerges [discontinuity of ρ at T - Tb = 0.6 K in (c)]. The same behavior is observed when the bulk concentration of monoolein is decreased. — Fig. 6: Influence of a variaition of the glycerol content on nematic prewetting transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 9CB (doped with monoolein, mole fraction x_s = 0.005) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. T_b denotes the bulk nematic - isotropic transition temperature of the sample. With increasing glycerol content a first-order prewetting transition emerges [discontinuity of ρ at T - T_b = 0.6 K in (c)]. The same behavior is observed when the bulk concentration of monoolein is decreased.

Fig. 6: Influence of a variaition of the glycerol content on nematic prewetting transitions. The figures show the temperature dependence of the ellipticity coefficient ρ of the interface of 9CB (doped with monoolein, mole fraction x_s = 0.005) to aqueous phases with different glycerol content; the volume ratio between water and glycerol is indicated in each panel. T_b denotes the bulk nematic - isotropic transition temperature of the sample. With increasing glycerol content a first-order prewetting transition emerges [discontinuity of ρ at T - T_b = 0.6 K in (c)]. The same behavior is observed when the bulk concentration of monoolein is decreased.

Find more information:

Surface order at surfactant-laden interfaces between isotropic liquid crystals and liquid phases with different polarities
X. Feng and Ch. Bahr, Phys. Rev. E 84, 031701 (2011).
DOI: 10.1103/PhysRevE.84.031701

Interfaces laden with semifluorinated alkanes

All studies mentioned above concern liquid crystal/aqueous interfaces, at which surfactants, possessing the common polar head/nonpolar tail structure, readily form Gibbs films by adsorption from the bulk phase. Semifluorinated alkanes (SFAs) are known to be surface active for oil/air interfaces. Thus, they should form Gibbs films on LC/air interfaces and it should be possible to extend the studies of the surfactant-laden LC/aqueous interfaces to liquid crystal/air interfaces.

There are several differences between these two types of interfaces: One important difference is the anchoring condition of the bare (without surfactant) interfaces which is planar for water but homeotropic for air. So far, no study has addressed the question how the intrinsic anchoring and the accompanying surface ordering at LC/air interfaces is influenced by surfactants (since usual surfactants do not adsorb at oil/air interfaces). A second difference concerns the relevant interactions: Whereas at aqueous interfaces, laden with ionic surfactants, electrostatic interactions are important, interactions at LC/SFA/air interfaces can be expected to be mainly steric or of the van der Waals type. Another important point is that SFA monolayers can show structural phase transitions: For SFA Gibbs films on the surface of alkanes, a first-order transition with decreasing temperature from a dilute to a condensed state is observed [P. Marczuk, P. Lang, G. H. Findenegg, S. K. Mehta, and M. Möller, Langmuir 18, 6830 (2002)].

The following Figures show results for the nematic surface order of 8CB and the smectic surface order of 12CB at air interfaces laden with the semifluorinated alkane C₁₈H₃₇-C₁₂F₂₅(abbreviated as H₁₈F₁₂). We find that SFA films on LC surfaces show a similar transition as observed on alkane surfaces and that it has a striking effect on the anchoring and surface ordering behaviour of nematic and smectic LCs.

Fig. 7: Temperature dependence of the ellipticity coefficient ρ of the free surface of 8CB doped with different amounts of H18F12; TNI designates the bulk nematic - isotropic transition temperature of 8CB. The mole fraction of H18F12, xHF, is indicated at each curve. The inset shows the data for xHF = 1.1 x 10-4 with an expanded y-scale; note the decrease of ρ as TNI is approached from above. The schematic drawings on top illustrate the different proposed surface structures: dense H18F12 film with a planar aligned nematic layer on an isotropic bulk phase (a), and dilute H18F12 film with a homeotropic aligned nematic layer on an isotropic bulk phase (b). — Fig. 7: Temperature dependence of the ellipticity coefficient ρ of the free surface of 8CB doped with different amounts of H₁₈F₁₂; *T_NI* designates the bulk nematic - isotropic transition temperature of 8CB. The mole fraction of H₁₈F₁₂, *x_HF*, is indicated at each curve. The inset shows the data for *x_HF* = 1.1 x 10^-4 with an expanded y-scale; note the decrease of ρ as *T_NI* is approached from above. The schematic drawings on top illustrate the different proposed surface structures: dense H₁₈F₁₂ film with a planar aligned nematic layer on an isotropic bulk phase (a), and dilute H₁₈F₁₂ film with a homeotropic aligned nematic layer on an isotropic bulk phase (b).

Fig. 7: Temperature dependence of the ellipticity coefficient ρ of the free surface of 8CB doped with different amounts of H₁₈F₁₂; *T_NI* designates the bulk nematic - isotropic transition temperature of 8CB. The mole fraction of H₁₈F₁₂, *x_HF*, is indicated at each curve. The inset shows the data for *x_HF* = 1.1 x 10^-4 with an expanded y-scale; note the decrease of ρ as *T_NI* is approached from above. The schematic drawings on top illustrate the different proposed surface structures: dense H₁₈F₁₂ film with a planar aligned nematic layer on an isotropic bulk phase (a), and dilute H₁₈F₁₂ film with a homeotropic aligned nematic layer on an isotropic bulk phase (b).

Fig. 8: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H18F12; TAI designates the bulk smectic-A - isotropic transition temperature of 12CB. The mole fraction of H18F12, xHF, ranges from 0 (pure 12CB) to 1.1 x 10-3 and is indicated at each curve in units of 10-3. — Fig. 8: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H₁₈F₁₂; *T_AI* designates the bulk smectic-A - isotropic transition temperature of 12CB. The mole fraction of H₁₈F₁₂, *x_HF*, ranges from 0 (pure 12CB) to 1.1 x 10^-3 and is indicated at each curve in units of 10^-3.

Fig. 8: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H₁₈F₁₂; *T_AI* designates the bulk smectic-A - isotropic transition temperature of 12CB. The mole fraction of H₁₈F₁₂, *x_HF*, ranges from 0 (pure 12CB) to 1.1 x 10^-3 and is indicated at each curve in units of 10^-3.

Fig. 9: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H18F12; TAI designates the bulk smectic-A - isotropic transition temperature of 8CB. The mole fraction of H18F12, xHF, ranges from 1.1 x 10-3 to 1.5 x 10-3 (upper panel) and from 1.5 x 10-3 to 2.2 x 10-3 (lower panel) and is indicated at each curve in units of 10-3. The schematic drawings on top illustrate the different proposed surface structures: dense H18F12 film on an isotropic bulk phase (a), dilute H18F12 film and one smectic layer on an isotropic bulk phase (b), dilute H18F12 film on an isotropic bulk phase (c). — Fig. 9: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H₁₈F₁₂; *T_AI* designates the bulk smectic-A - isotropic transition temperature of 8CB. The mole fraction of H₁₈F₁₂, *x_HF*, ranges from 1.1 x 10^-3 to 1.5 x 10^-3 (upper panel) and from 1.5 x 10^-3 to 2.2 x 10^-3 (lower panel) and is indicated at each curve in units of 10^-3. The schematic drawings on top illustrate the different proposed surface structures: dense H₁₈F₁₂film on an isotropic bulk phase (a), dilute H₁₈F₁₂ film and one smectic layer on an isotropic bulk phase (b), dilute H₁₈F₁₂ film on an isotropic bulk phase (c).

Fig. 9: Temperature dependence of the ellipticity coefficient ρ of the free surface of 12CB samples doped with different amounts of H₁₈F₁₂; *T_AI* designates the bulk smectic-A - isotropic transition temperature of 8CB. The mole fraction of H₁₈F₁₂, *x_HF*, ranges from 1.1 x 10^-3 to 1.5 x 10^-3 (upper panel) and from 1.5 x 10^-3 to 2.2 x 10^-3 (lower panel) and is indicated at each curve in units of 10^-3. The schematic drawings on top illustrate the different proposed surface structures: dense H₁₈F₁₂film on an isotropic bulk phase (a), dilute H₁₈F₁₂ film and one smectic layer on an isotropic bulk phase (b), dilute H₁₈F₁₂ film on an isotropic bulk phase (c).

The behavior of the ellipticity coefficent ρ and the Brewster angle θ_B (not shown in the Figures) show that formation of the dense H₁₈F₁₂ film causes an anchoring transition of the LC phase from homeotropic at the dilute H₁₈F₁₂film to planar at the dense H₁₈F₁₂ film.

The change of the anchoring condition from homeotropic to planar has a striking effect on the nematic or smectic order which is present at the surface of the isotropic LC bulk phase. For 8CB, which shows nematic surface order, the ellipsometric data suggest that a thin planar nematic film remains when the dense H₁₈F₁₂ film forms. The planar nematic film grows in thickness as the bulk transition to the nematic phase is approached from above (cf. inset in Figure 7).

In the case of 12CB, which shows smectic surface order, our data suggest that the few (essentially one or two) smectic layers, which exist on the surface of the isotropic bulk phase, vanish when the dense H₁₈F₁₂ film forms, i.e., the smectic surface order is destroyed when the LC molecules change their alignment from homeotropic to planar.

In addition to the ellipsometry measurements, we have conducted AFM studies of the free surface of 8CB and 12CB droplets doped with H₁₈F₁₂ in order to clarify the structure of the dense film of the semifluorinated alkane. The AFM images show clearly that the dense film consists of a crystalline-like hexagonal packing of surface micelles, similar to those observed in transferred Langmuir films and spin-coated films of semifluorinated alkanes (see, e.g., M. P. Krafft, Acc. Chem. Res. 45, 514 (2012)). In addition to the hexagonal pattern of the surface micelles, we observe in some cases linear steps in the surface (arrows in Figure 10b) the height of which is clearly smaller than the smectic layer thickness. So far, we do not have an explanation for these steps. The formation of the surface micelles might be the reason for the observed anchoring transition from homeotropic to planar (see Figure 12).

Fig. 10: AFM height images of the surface of droplets of 12CB (a) and 8CB (b) doped with H18F12. The conditions are chosen such that a dense film of the semifluorinated alkane forms on the surface of the liquid crystals. The images demonstrate the presence of surface micelles (lateral diameter approximately 40 nm) which arrange themselves in a dense hexagonal packing. The inset in (a) shows a FFT plot of the height image. Arrows in (b) indicate three terrace steps with a height difference of 1 nm, the blue line indicates the location of the cross section shown in the following figure. The area of each image corresponds to 0.5 x 0.5 µm2. — Fig. 10: AFM height images of the surface of droplets of 12CB (a) and 8CB (b) doped with H₁₈F₁₂. The conditions are chosen such that a dense film of the semifluorinated alkane forms on the surface of the liquid crystals. The images demonstrate the presence of surface micelles (lateral diameter approximately 40 nm) which arrange themselves in a dense hexagonal packing. The inset in (a) shows a FFT plot of the height image. Arrows in (b) indicate three terrace steps with a height difference of 1 nm, the blue line indicates the location of the cross section shown in the following figure. The area of each image corresponds to 0.5 x 0.5 µm².

Fig. 10: AFM height images of the surface of droplets of 12CB (a) and 8CB (b) doped with H₁₈F₁₂. The conditions are chosen such that a dense film of the semifluorinated alkane forms on the surface of the liquid crystals. The images demonstrate the presence of surface micelles (lateral diameter approximately 40 nm) which arrange themselves in a dense hexagonal packing. The inset in (a) shows a FFT plot of the height image. Arrows in (b) indicate three terrace steps with a height difference of 1 nm, the blue line indicates the location of the cross section shown in the following figure. The area of each image corresponds to 0.5 x 0.5 µm².

Fig. 11: Cross section through three surface micelles as obtained from the AFM height image shown in Figure 10b (blue line).

Fig. 11: Cross section through three surface micelles as obtained from the AFM height image shown in Figure 10b (blue line).

Fig. 12: Schematic of two surface micelles formed by the semifluorinated alkane molecules on the surface of an isotropic liquid crystal. Above the bulk transition to the nematic phase, a thin nematic film with planar anchoring is present.

Fig. 12: Schematic of two surface micelles formed by the semifluorinated alkane molecules on the surface of an isotropic liquid crystal. Above the bulk transition to the nematic phase, a thin nematic film with planar anchoring is present.

Find more information:

Surface ordering and anchoring behaviour at liquid crystal surfaces laden with semifluorinated alkane molecules
X. Feng, A. Mourran, M. Möller, and Ch. Bahr, Soft Matter 8, 9661 (2012).
DOI: 10.1039/c2sm26177d

AFM study of Gibbs films of semifluorinated alkanes at liquid crystal/air interfaces
X. Feng, A. Mourran, M. Möller, and Ch. Bahr, ChemPhysChem 14, 1801 (2013).
DOI: 10.1002/cphc.201300173