Gelling Lamellar Phases of the Binary System Water

Oct 13, 2017 - To answer these questions we study gelled lyotropic liquid crystals (LLC). In the present study we show that it is possible to gel the ...
4 downloads 16 Views 7MB Size
Download PDF
Article Cite This: Langmuir XXXX, XXX, XXX-XXX

pubs.acs.org/Langmuir

Gelling Lamellar Phases of the Binary System Water− Didodecyldimethylammonium Bromide with an Organogelator Sachi Koitani,† Sonja Dieterich,‡ Natalie Preisig,‡ Kenji Aramaki,† and Cosima Stubenrauch*,‡ †

Graduate School of Environment & Information Sciences, Yokohama National University, Tokiwadai 79-7, Hodogaya, Yokohama 240-8501, Japan ‡ Institute of Physical Chemistry, University of Stuttgart, Pfaffenwaldring 55, 70569 Stuttgart, Germany S Supporting Information *

ABSTRACT: Does the presence of a gel network influence the properties of a lyotropic liquid crystal? Does the replacement of oil by a lyotropic liquid crystal influence the properties of an organogel? To answer these questions we study gelled lyotropic liquid crystals (LLC). In the present study we show that it is possible to gel the lamellar phase of the binary system water−didodecyl dimethylammonium bromide (2C12DAB) with the organogelator 12-hydroxyoctadecanoic acid (12-HOA). We compare various properties of the gelled LLC phases with the “parent systems”, i.e., with the binary organogel consisting of n-decane−12-HOA and with the nongelled LC phases, respectively. Optical and electron microscopy, differential scanning calorimetry (DSC), rheometry, as well as small and wide-angle X-ray scattering (SWAXS) proved the coexistence of an Lα phase and a 12-HOA gel network in the gelled Lα phase. However, a small influence of the Lα phase on the gel properties was seen, namely slightly lower sol−gel transition temperatures and viscoelastic moduli of the gelled Lα phase compared to the binary gel. On the other hand, the presence of the gel also has an influence on the Lα phase: the interlayer spacing of the surfactant bilayers in the gelled Lα phases is slightly larger compared to the nongelled Lα phases, which is due to mixing part of the 12-HOA molecules in the Lα bilayers. Despite this mutual influence the structures of both the Lα phase and the gel network are hardly disturbed in the gelled Lα phase, i.e., that the self-assembly of the surfactant and of the gelator molecules clearly occur in an orthogonal way.

1. INTRODUCTION Gelled lyotropic liquid crystals (LLC) are novel soft materials in which the microstructure of the LLC is combined with the mechanical stability of a gel.1 The unique selling point of gelled LLCs is the fact that the two coexisting structures can take over two different functions. For example, in transdermal drug delivery a gel would guarantee a convenient application, while an LLC is needed for an effective solubilization of waterinsoluble drugs.2,3 A second example is a membrane-based liquid crystalline gel, which could incorporate membraneembedded proteins that are biologically active, thus providing a way in which proteins can be delivered via a stable gel.4 A third example is related to materials science. LLCs are used as templates for the synthesis of highly ordered, nanostructured materials.5 Gelling LLCs would increase their mechanical stability, which may be helpful for the synthesis. Moreover, it should be possible to “arrest” macroscopically aligned LLCs, i.e., monodomains, by gelling them. These gelled monodomains can be used as templates for macroscopically aligned nanoporous monoliths.6 Fourthly, gelled LLCs might be used as anisotropic electrolyte gels enabling efficient ion migration pathways (“ion channels”) for potential applications, e.g., in enhanced lithium battery systems.7 The most prominent example, however, of a lyotropic liquid crystalline gel, i.e., of a gelled LLC, is the cell. We will come back to this example in © XXXX American Chemical Society

the next paragraph. Studying gelled LLCs we thus expect to better understand their properties in order to tailor-make new functional and “smart” materials. Some gelled LLCs are orthogonal self-assembled systems.1 Orthogonal self-assembly is the independent but simultaneous formation of two coexisting self-assembled structures within one system. For orthogonal self-assembly to happen, the noncovalent interactions leading to the two self-assembled structures must be selective and noninterfering. Due to this prerequisite not any gelator but only gelators that form a physical gel whose network forms via self-assembly can be used for the formation of an orthogonal self-assembled system. This class of gelator is known as low molecular weight gelators (LMGs) which can self-assemble into gelator fibers. These fibers, in turn, form entangled networks, so-called selfassembled fibrillar networks, SAFiNs, which convert the liquid (sol) into a gel.8 Both in nature and science there are numerous examples of orthogonal self-assembled systems.9−21 As already mentioned, the most prominent example of a gelled LLC is the cell, which, in turn, forms via orthogonal self-assembly. The cell membrane is a phospholipid bilayer (self-assembled surfacReceived: June 20, 2017 Revised: September 28, 2017

A

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir tants) which coexists with a variety of other self-assembled architectures such as protein assemblies that form scaffolding filaments in the cell, i.e., the cytoskeleton.22 The cytoskeleton is one of the most sophisticated gels, and the combination with the fluid cell membrane leads to a membrane-based liquid crystalline biogel or short to a gelled lamellar (Lα) phase. The important points of the cell structure are that each selfassembled structure plays a different role and that both selfassembled structures are needed for a proper functioning of the cell. In other words, the two structures complement each other, however, without influencing one another. This example clearly demonstrates the huge prospects of studying orthogonal selfassembled systems. Based on what we described so far the two questions to answer are straightforward: (1) How to gel lyotropic liquid crystals? (2) Are gelled LLCs orthogonal self-assembled systems? In our previous work we studied gelled bicontinuous microemulsions17−21 as well as gelled LLCs23 and discussed the results in terms of orthogonal self-assembly. In the first case we indeed found that the microemulsion and the gel network form independently and that gelled bicontinuous microemulsions are thus a new example of an orthogonal self-assembled system. In the second case, however, things are different. While both the phase behavior as well as the rheological properties of the LLC phases support the hypothesis that gelled lyotropic liquid crystals are orthogonal self-assembled systems as well, freeze fracture electron microscopy (FFEM) seem to indicate an influence of the gel network on the structure of the Lα phase and vice versa. The FFEM-pictures show that the presence of the Lα phase changes the structure of the gelator fibers and that the presence of the gelator changes the ordering of the Lα phase.23 In our previous study we investigated the ternary system water−n-decane−n-decyl tetraoxyethylene glycol ether (C 10 E 4 ) which we gelled with the organogelator 12hydroxyoctadecanoic acid (12-HOA). The measurements were carried out at surfactant concentrations around 30 wt % since the lyotropic LC phases do not form at lower surfactant concentrations. This study aims at complementing our previous work by further studying the influence of the gel network on the structure of the LLC and vice versa. For this purpose, we simplified the system (a) by choosing a binary system consisting of water and the cationic surfactant didodecyldimethylammonium bromide (2C12DAB) and (b) by looking at the lamellar phase (Lα) only, which is formed at surfactant concentrations as low as a couple of wt %.24 At first we wanted to gel the Lα phase of this binary system with a traditional hydrogelator, namely N,N′-dibenzoyl-L-cystine (DBC), since most of the system is water. However, we failed and tried 12-HOA instead. Counter to what we expected, we managed to gel the binary system H2O−2C12DAB with the organogelator 12-HOA. To the best of our knowledge this is the first lamellar phase of a binary system that has been gelled with an LMG in general and with an organogelator in particular. The systems to study are thus gelled lamellar phases of H2O−2C12DAB−12-HOA (Scheme 1, middle) as well as the “two parent systems”, namely lamellar phases of H2O− 2C12DAB (Scheme 1, left) and gels of n-decane−12-HOA (Scheme 1, right). The techniques we chose to show that the microstructure of our gelled Lα phase looks indeed like the one drawn in Scheme 1 (middle) are visual and microscopic inspections of the samples, differential scanning calorimetry (DSC), rheology, small-angle X-ray scattering (SAXS), and freeze fracture electron microscopy (FFEM).

Scheme 1. Schematic Drawings of the Systems under Questiona

a

(Left) Lamellar phase of H2O−2C12DAB, (middle) gelled lamellar phase of H2O−2C12DAB−12-HOA, and (right) gel of n-decane and 12-HOA.

2. EXPERIMENTAL SECTION 2.1. Materials. 12-Hydroxyoctadecanoic acid (12-HOA, >80.0%, no specification about stereochemistry) and didodecyldimethylammonium bromide (2C12DAB, >98%) were purchased from Tokyo Chemical Industry co., Ltd. Bidestilled water was used for all sample preparations. Note that we studied very extensively the binary system n-decane−12-HOA and showed that our 12-HOA is (nearly) pure (R)-12-HOA.25 2.2. Methods. 2.2.1. DSC. In order to determine the sol−gel transition temperatures of the gelled samples, DSC measurements were carried out using a differential scanning calorimeter DSC 8000 from PerkinElmer. The calorimeter was calibrated for each of the used heating rates (1, 5, 10, and 20 K min−1) by measuring the melting temperatures of zinc and indium. About 10−20 mg of the samples was filled into aluminum pans (PerkinElmer, Part No. B016−9321), which were sealed and placed in the furnace of the instrument. The temperature program consisted of keeping the temperature constant at 10 °C for 1 min and then heating to 80 °C with the mentioned different heating rates. The heat flow versus temperature curves were recorded and analyzed using the software Pyris from PerkinElmer. Note that sol−gel transition temperatures were measured solely on heating since phase separation may occur at high temperatures and a reformation of the original gel upon cooling cannot be guaranteed. 2.2.2. Optical Microscopy. Polarized optical microscopy was used to study the textures of the binary gel, of the lyotropic lamellar phase, and of the gelled lamellar phase. The samples were placed between untreated glass plates and observed at constant temperature (T = 25 °C) using a Leica DMLP polarization microscope. Images of the textures were taken with a Nikon D40 reflex camera. 2.2.3. Rheometry. The rheological measurements were performed on an AG-R2 rheometer (TA Instruments). A plate−plate assembly with an upper (moving) plate of 4.0 cm diameter was used. The temperature was set and kept constant at 25 °C. Then the samples (2.0 g) were transferred to the plate with a spatula. Subsequently, the upper plate was lowered to the measuring position at a gap width of 1 mm. Initially, oscillation-strain sweeps were performed with a constant oscillation frequency of 1 Hz in order to identify the linear viscoelastic (LVE) ranges of each sample. Then oscillation-frequency sweeps (0.01−100 rad/s) were performed at a constant strain within the LVE ranges. From these measurements the storage modulus G′ and the loss modulus G″ were determined. We repeated all measurements with good reproducibility and calculated the average values of three oscillation-frequency sweep measurements. In order to determine the sol−gel transition temperatures, oscillation temperature sweeps were carried out at a constant frequency of 1 Hz and a shear stress of 0.80 Pa. The temperature was ramped up with a heating rate of 1 K min−1. The software collected data every 78 s until the measurements were stopped after the dropdown of G′ and G″ which indicated the transition from the solid-like gel to the liquid sol. 2.2.4. SWAXS. SWAXS measurements were performed on a Kratkytype camera (SAXSess, Anton Paar, Austria) with a PW3830 laboratory X-ray generator (Philips, Netherlands). The SAXSess has a long fine focus sealed glass X-ray tube (Cu Kα wavelength of 0.1542 nm). The apparatus was operated at 40 kV and 50 mA, and the B

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir samples were irradiated for 10 min. Samples were loaded in a cell with a Kapton-film window specially designed for the SAXSess camera. 2.2.5. FFEM. Freeze Fracture and Etching System EM BAF060 from Leica was used for replica preparation of the gelled and nongelled lamellar phases (H2O−2C12DAB with and without 12-HOA) and the binary organogel (n-decane−12-HOA). The samples were prepared as described in section 2.3. Two copper plates (4.5 mm × 3.0 mm) and two copper grids were assembled in a so-called sandwich. The specimens were placed on the grids and plates using a spatula and scalpel at room temperature. Subsequently the specimen sandwiches were quickly frozen in liquid ethane. The frozen fractured surface of the specimens was shadowed with platinum−carbon (∼2 nm) at 45° and covered by a layer of pure carbon (∼20 nm) at 90° in the vacuum chamber of the BAF060, whose specimen stage was cooled to −150 °C. The replicas were cleaned with ethanol, dried and inspected with the transmission electron microscope Tecnai G2 Sphera from FEI. 2.3. Sample Preparation and Samples. The surfactant mass fraction is defined as m2C12DAB γ= mH2O + m2C12DAB (1) and the gelator mass fraction as m12 − HOA η= mH2O + m2C12DAB + m12 − HOA

with visible light, the system n-decane−12-HOA (binary gel) is birefringent indicating local agglomerations of the gelator fibers. The POM images obtained for the gelled Lα phases do not match either of the textures of the parent systems. Hence, no statement regarding the structure of the gelled Lα phase can be made by polarized optical microscopy. However, as can be seen in Figure 1d, the homogeneity of the gelled Lα phases was confirmed by taking images without having the analyzer in the light path. 3.2. DSC and Rheology. The sol−gel transition temperature Tsol−gel is a measure for the strength of a gel and increases with increasing gelator concentration until it reaches a plateau. In this study three complementary methods were carried out for measuring Tsol−gel. (1) A “table-top” technique probes the most characteristic property of a gel, namely the fact that it exhibits no flow in the steady state.8 An inversed test tube is placed in a water basin, and the temperature is slowly increased until the sample starts flowing. (2) With temperature dependent rheology measurements one can examine the mechanical stability of the gel. A sudden decrease of the storage modulus G′ and of the loss modulus G″, which are representing the solid-like and liquid-like character of the gel, respectively, as well as G′ < G″ indicate the loss of the solid-like character of the sample. (3) With DSC the melting point of the gel is measured. All samples are measured four times with the above-mentioned heating rates, namely 1, 5, 10, and 20 K min−1. The melting of the gel is an endothermic process which takes place in a broad temperature range.8 Hence, broad gel melting peaks were obtained, and Tsol−gel was set equal to the maximum of the peak and not to its onset, which is common for phase transitions of first order. For each sample, the sol−gel transition temperatures obtained by the different heating rates were very similar to a maximum deviation of ±1.5 K which is quite good regarding the broad peaks (see Figure S2). For the

(2)

We used γ = 0.10 or 0.20 for preparing the lamellar liquid crystalline phases. After preparing the Lα phases, we added 3 or 5 wt % of the gelator (η = 0.03 or 0.05). We weighed all compounds in screw-capped glass tubes. For homogeneous mixing, the gelator-containing samples first had to be heated in an aluminum block heater to ∼90 °C to melt the solid 12-HOA. Subsequently, the samples were gelled by putting them into an ice bath. All investigated samples including the binary parent systems H2O−2C12DAB and n-decane−12-HOA, respectively, are listed in Table 3.

3. RESULTS AND DISCUSSION 3.1. Optical Microscopy. In Figure 1 images taken by polarized optical microscopy are shown. The oily streak texture, which is characteristic for lyotropic lamellar phases, was observed for the nongelled Lα phase (see Figure 1a: POM image of the system H2O−2C12DAB at γ = 0.10). Although the thickness of the gelator fibers (12−60 nm according to a previous FFEM study25) is below the limit for an observation

Figure 2. Sol−gel transition temperatures Tsol−gel measured by DSC for the system n-decane−12-HOA (binary gel) and the systems H2O− 2C12DAB−12-HOA at γ = 0.10 and 0.20 with gelator mass fractions of η = 0.03 and 0.05. The DSC data for the binary gel from ref 25 are also shown. The solid line is a guide to the eye.

binary gel n-decane−12-HOA, Tsol−gel increases with increasing gelator amount. The obtained values for Tsol−gel are in good agreement with the results reported in ref 25. For the gelled lyotropic lamellar phases, however, an increase of Tsol−gel with increasing gelator amount could not be observed within the error. It might be that the plateau of Tsol−gel is reached at lower gelator concentrations for the gelled Lα phases than for the

Figure 1. Polarized optical microscopy images taken from (a) the nongelled Lα phase H2O−2C12DAB (γ = 0.10), (b) the binary gel ndecane−12-HOA (η = 0.05), and (c) the gelled Lα phase H2O− 2C12DAB−12-HOA (γ = 0.10, η = 0.05). (d) An image taken from the same sample at the same position as in panel c without having the analyzer in the beam path. The scale bars are 200 μm. C

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

measures the temperature at which the gelator fibers start to melt. Figure 3 shows the results of rheological measurements carried out with (left) the binary gel at two different gelator concentrations, (middle) the gelled Lα phase at the very same gelator concentrations and a surfactant concentration of γ = 0.10, and (right) the nongelled Lα phase at γ = 0.10. The corresponding data for the higher surfactant concentration of γ = 0.20 are shown in the Supporting Information (Figure S3). All results show that the storage modulus G′ is higher than the loss modulus G″ at any oscillatory shear frequency ω indicating elastic or gel properties. For the binary gel, both G′ and G″ are almost independent of ω, which is typical for a strong gel.26 Laupheimer et al.25 reported similar data for the system ndecane−12-HOA at η = 0.025 and 0.05, and we found that the values of our samples with η = 0.03 are in between those reported in ref 25. On the other hand, the G′ and G″ values of the nongelled and the gelled Lα samples are lower compared to those of the binary gel, and they slightly depend on ω indicating a weaker gel. Note that G′ and G″ of the binary gel are 1 to 2 orders of magnitude higher than those of the gelled Lα samples. Although the solvents of the binary gel and of the gelled Lα phases are not the same, large differences in the G′ and G″ values indicate a different structure of the gel network or a lower density of 12-HOA fibers. As already mentioned, one reason for the lower G′ and G″ values of the gelled Lα samples might be the fact that 12-HOA is partly used as a cosurfactant leading to a lower effective gelator concentration and thus less densely packed 12-HOA fibers. Another possible reason for the lamellar gel being weaker than the binary gel might be different intermolecular interactions between the solvent (lamellar phase or n-decane) and the 12-HOA molecules. It might be that interactions between water and the 12-hydroxy group of the gelator molecules hamper the interactions between the gelator molecules themselves thus leading to a weaker gel. We recall that the DSC study revealed lower sol−gel transition temperatures of the gelled Lα than those of the binary gel. This observation is in accordance with the finding that the gelled Lα phase is a weaker gel than the binary parent system. 3.3. SWAXS. SAXS and WAXS curves recorded at 25 °C are shown in Figure 4. At low q values there is no distinct peak for

binary gel. Note that the sol−gel transition temperatures of the gelled Lα phases are 10−15 °C lower than those of the binary gel. A quantitative comparison of the gelled Lα phase with the binary gel is not meaningful, since the solvent is an oil (ndecane) and an aqueous surfactant solution, respectively. It is expected that gelling an oil leads to a stronger gel, since 12HOA is an organogelator and that different intermolecular interactions between 12-HOA, H2O, and 2C12DAB weakens the gel. Another explanation for the lower Tsol−gel values is the fact that 12-HOA partly acts as a cosurfactant, which, in turn, lowers the amount of gelator that forms the gel network. The sol−gel transition temperatures of the gelled Lα phases prepared with 20 wt % surfactant are slightly lower than those of the gelled Lα phases prepared with 10 wt % surfactant. This observation indicates that the H2O−2C12DAB mixture becomes a better solvent for 12-HOA with increasing surfactant concentration. The sol−gel transition temperatures obtained by DSC and temperature dependent rheology measurements are listed in Table 1 (examples of a rheology temperature scan and a DSC Table 1. Sol−Gel Transition Temperatures for the System H2O−2C12DAB−12-HOA at γ = 0.10 and 0.20 with Gelator Mass Fractions of η = 0.03 and 0.05 Obtained by DSC and Temperature Dependent Rheology Measurements Tsol−gel/°C system

η

DSC

rheology

H2O−2C12DAB−12-HOA (γ = 0.10)

0.03 0.05 0.03 0.05

52 49 49 48

47 47 not measured not measured

H2O−2C12DAB−12-HOA (γ = 0.20)

curve are shown in the Supporting Information (Figures S1 and S2)). Regarding the experimental error, the values for Tsol−gel obtained by the different techniques are in good agreement. Temperature dependent rheology measurements provide slightly lower values for Tsol−gel than DSC measurements. The reason for this is that rheology measures the temperature at which the gel fibers loose their interconnectivity, while DSC

Figure 3. Storage modulus G′ (filled symbols) and loss modulus G″ (open symbols) as a function of oscillatory shear frequency (ω) in the systems n-decane−12-HOA (binary gel), H2O−2C12DAB−12-HOA (gelled Lα at γ = 0.10), and H2O−2C12DAB (nongelled Lα at γ = 0.10) at 25 °C. D

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Table 2. SAXS Peak Positions and Interlayer Spacing (d Spacing) of Gelled and Nongelled Lα Phases at Different Gelator (η = 0.03 and 0.05) and Surfactant Concentrations (γ = 0.10 and 0.20) nongelled Lα γ = 0.10 gelled Lα γ = 0.10

η

q1/nm−1 (d/nm)

q2/nm−1

q3/nm−1

q1/q2/q3

0

0.261 (24.1)

0.522

0.789

1/2/3

0.03

0.248 (25.3) 0.241 (26.1)

0.502

1/2/−

0.489

1/2/−

0

0.515 (12.2)

1.04

0.03

0.509 (12.3) 0.502 (12.5)

1.04

1/2/−

1.02

1/2/−

0.05

nongelled Lα γ = 0.20 gelled Lα γ = 0.20

0.05

1.56

1/2/3

spacing was found via SAXS in our previous work25 as well as in the present study (broad peak around q = 1.2 nm−1 in Figure 4). From the first-order Bragg peak the interlayer spacing d of the lamellar structure was calculated (see Table 2). The dspacing of the nongelled Lα is lower than that of the gelled Lα at η = 0.03 and 0.05. As mentioned in the previous section, it is very likely that some 12-HOA molecules are incorporated in the surfactant bilayer. If this is the case, the d-spacing of the lamellar phase must be affected. The d-spacing can be described with eq 3 by considering a geometrical molecular packing model28,29 2vL d= aSϕL (3)

Figure 4. SAXS and WAXS curves of the binary gel n-decane−12HOA, the gelled and the nongelled Lα samples in the system H2O− 2C12DAB−12-HOA at (top) γ = 0.10 and (bottom) γ = 0.20. (a) nongelled Lα (η = 0); (b) gelled Lα with η = 0.03; (c) gelled Lα with η = 0.05; (d) binary gel n-decane−12-HOA (η = 0.03). Note that the peak at around q = 3.8 nm−1 comes from the kapton film of the sample cell. All measurements were carried out at T = 25 °C.

where vL is the averaged volume of the hydrophobic part of the surfactant, ϕL is the volume fraction of the hydrophobic part of surfactant, and aS is the averaged molecular area occupied at the hydrophilic−lipophilic interface. In this study, we obtained the gelled Lα phase by adding 12-HOA to the lamellar phase of the system H2O−2C12DAB, which leads to an increase of ϕL and a decrease of vL. Those changes must lead to a decrease in the dspacing according to eq 3. On the other hand, adding 12-HOA, which has a smaller molecular area at the interface compared to 2C12DAB, decreases aS and thus increases the d-spacing. We roughly estimated the d-spacing change which results from incorporating 12-HOA molecules in a 2C12DAB lamellar phase (see the Supporting Information). The resulting d-spacing of the gelled Lα phase is lower than that of the nongelled Lα phase if we assume ideal mixing of the molecules at the interface. However, the experimentally determined d-spacing of the gelled Lα phase is larger and not smaller. One explanation for this discrepancy is the shift of the scattering peak to lower q caused by an increasing lamellar stacking disorder or polydispersity.30 Another explanation is nonideal mixing of the two molecules in a lamellar bilayer. Synergistic effects are known in lamellar bilayers composed of two different molecules caused by reduced repulsion between the hydrophilic head groups,31,32 which could also be the case in the present system. Therefore, we speculate that synergistic attractive interactions between 2C12DAB and 12-HOA molecules are rather strong and that the

the binary gel, whereas several Bragg peaks are found for the nongelled Lα sample with peak position ratios of q1/q2/q3 = 1/ 2/3, which is characteristic for a lamellar structure (see Table 2). For the gelled Lα phase, a lamellar peak pattern was also observed at low q values. Thus, a lamellar structure still forms after addition of 12-HOA, although the Bragg peaks are broadened compared to those of the nongelled Lα sample which is why only the first two peaks are clearly visible. The broadened and less intense Bragg peaks indicate a lower translational order of the lamellar bilayers since a broader peak is a result of a shorter correlation length. In addition to the broadened Bragg peaks, we also found a broad SAXS peak at around q = 1.2 nm−1, which can also be seen in the curve of the binary gel. Tachibana et al.27 reported for the binary system benzene−12-HOA that the SAXS peak corresponds to a spacing of 4.67 nm and thus to a 12-HOA double layer in which two 12-HOA molecules face each other by their carboxylic headgroup with a tilt angle of 21° from the normal to the layer plane. In the system n-decane−12-HOA the same E

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

because 12-HOA forms dimers in a bent conformation.44,45 All of these features were also found for the binary gel prepared in this study (Figure 5, left). For the sake of clarity, one schematic

resulting decrease of as overwrites the effect of the other parameters. The WAXS curve of the binary gel has an amorphous peak between q = 10−17 nm−1, which can be attributed to the solvent n-decane. If we look at the details of the WAXS curve of the binary gel, we see a shoulder at about q = 13.3 nm−1 (0.472 nm) on the amorphous peak, indicating an angstrom-sized ordered structure. The peak at q = 13.3 nm−1 is more pronounced in the WAXS curves of the gelled Lα samples and does not appear for the nongelled Lα. A short spacing in the gel fibers of the system benzene−12-HOA27 was found at 0.46 nm which is in agreement with the value found for the present system, namely 0.47 nm. In general, the short spacings of bimolecular layers composed of hydrocarbon chains correspond to the cell constant of a subcell structure. In lipid, surfactant, or fatty alcohol systems, a β form (2D triclinic lattice) gives a single peak for a spacing of 0.46 nm.33 Thus, the WAXS curves also confirm the coexistence of the lamellar phase and the 12HOA fibers. 3.4. FFEM. We studied different gelled and nongelled samples (bold in Table 3) using the FFEM technique as described in section 2.2.

Figure 5. FFEM pictures of the binary gel n-decane−12-HOA at η = 0.05 (left) and of the nongelled Lα phase in the system H2O− 2C12DAB at γ = 0.10, η = 0 (right). Black and white arrows show the twisted gelator fibers (left) and the lamellar layers in the MLVs (right), respectively. A schematic single fiber is also shown. All samples were prepared at room temperature.

single fiber consisting of the twisted layers is shown as well. Note that in the most recent preparation of the binary gel we observed also some fibers of about 150−200 nm in width whose pitch is much larger compared to the pitch estimated in ref 25. As already mentioned, FFEM pictures of the second parent binary system, namely of the nongelled lamellar phase H2O− 2C12DAB were shown in ref 37 for three different surfactant concentrations (γ = 0.002, 0.02, and 0.20). Small unilamellar vesicles with an average diameter of about 200 nm were observed at γ = 0.002, while large multilayered vesicles (MLV) with a diameter of up to 5 μm and a lamellar repeat distance of about 10 nm could be seen at γ = 0.20. The coexistence of both structures was shown for the sample at γ = 0.02. It is reported in ref 39 that large multilayer vesicles occur up to 30 wt % of 2C12DAB in the swollen Lα1 region, and optical micrographs of the MLVs at 25 °C are also shown. At high surfactant concentrations (and at low surfactant concentrations but at high temperatures) a collapsed lamellar phase Lα2 (regular lamellae) exists. Zemb et al.38 carried out a detailed SAXS and SANS study of the coexistence of two lamellar phases of the binary system H2O−2C12DAB and determined the periodicity of these two lamellar phases (11 and 3.2 nm for swollen Lα1 and collapsed Lα2, respectively). In our case, i.e., at γ = 0.10 and 0.20 at room temperature, only the swollen Lα1 region exists. As expected, we observed large MLVs with diameters from ∼0.8 μm up to ∼9 μm in the nongelled lamellar phase H2O− 2C12DAB at γ = 0.10 (see Figure 5, right). The lamellar repeat distance in the MLVs estimated from the FFEM pictures is ∼20−25 nm at γ = 0.10 and ∼10−13 nm at γ = 0.20 (the FFEM picture is not shown), respectively. These values are in good agreement with the SAXS data (24.1 nm at γ = 0.10 and 12.1 nm at γ = 0.20 from Table 3 and with the d-spacing calculated with d ≈ 2δ/ϕc,i,46 which leads to 24 and 12 nm, respectively. Here 2δ ≈ 2.4 nm39 corresponds to the thickness of the surfactant bilayer, while ϕc,i ≈ 0.10 and 0.20 correspond to the volume fraction of surfactant in the lamellar layers. To visualize the structure of the gelled Lα phase (H2O−2C12DAB− 12-HOA) and to find out if the structures of the both parent systems (namely, gelator fibers and multilayer vesicles) exist in the ternary system, we have studied two samples with different

Table 3. Gelled and Nongelled Samples where γ is the Mass Fraction of 2C12DAB and η the Mass Fraction of 12-HOAa System H2O−2C12DAB−12-HOA γ

η

0.10 0.10 0.20 0.20 0.10 0.20

0.03 0.05 0.03 0.05 0 0

fiber width (FFEM), nm

dLa (FFEM), nm

d-spacing (SAXS), nm

20−40 25−35

∼20−25 20−25

25.3 26.1 12.5 12.3 24.1 12.2

n.a. 20−25 n.a. 10−13 System n-Decane−12-HOA

γ

η

fiber width (FFEM), nm

d12HOA (FFEM), nm

0 0

0.03 0.05

20−50

∼5

d-spacing (SAXS), nm 5.2

a The width of most fibers in the gelled samples and the lamellar repeat distance d was deduced from FFEM pictures. d spacing obtained from SAXS are also shown.

The parent binary systems of this study, namely the lamellar phase of the system H2O−2C12DAB and the binary gel ndecane−12-HOA, were already investigated in detail in the past.23−25,34−39 FFEM pictures of the nongelled lamellar phase are shown in ref 37, while the structure of the binary gel is visualized in refs 23, 25, and 34. Let us at first summarize what is known from FFEM pictures about the structure of the binary gel (n-decane−12-HOA). It has long been recognized that 12HOA forms twisted fibers in different organic solvents.40−42 For the system n-decane−12-HOA at η = 0.05 it was shown in ref 25 that (1) the width of the individual gelator fibers in ndecane is on average 39 ± 13 nm; (2) the single fibers consist of several twisted layers with a spacing of approximately 5 nm; (3) the average pitch of the gelator fibers is 155 ± 35 nm; (4) the gelator fibers are left-handed helices if the sample predominantly consists of (R)-12-HOA, while they are righthanded helices for samples where (S)-12-HOA prevails;42 (5) 12-HOA fibers have the same structure as twisted nanofilaments formed by bent core molecules in the LC phase B4,43 F

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

active at all there should be no change of the interlayer spacing (which, indeed, will be the focus of our future work). Thus, we can summarize that the self-assembly of the surfactant and of the gelator molecules clearly occur in an orthogonal way.

gelator concentrations. FFEM pictures of the gelled Lα phase at γ = 0.10 are shown in Figure 6 (top) for η = 0.05 and in Figure 6 (bottom) for η = 0.03.

4. CONCLUSIONS The present study is about the structure and properties of a gelled lamellar liquid crystal and its “parent systems”, i.e., the lamellar phase and the gel. The system under question consists of water, the cationic surfactant didodecyldimethylammonium bromide (2C12DAB), and the organogelator 12-hydroxyoctadecanoic acid (12-HOA). Optical and electron microscopy, differential scanning calorimetry (DSC), rheometry, as well as small and wide-angle X-ray scattering (SWAXS) proved the coexistence of an Lα phase and a 12-HOA gel network in the gelled Lα phase. However, a small influence of the Lα phase on the gel properties and of the gel network on the structure of the Lα phase was seen. 4.1. Influence of Lα on the Gel Network. The sol−gel transition temperatures and the viscoelastic moduli of the gelled Lα phase are slightly lower compared to those of the binary gel. The presence of the surfactant solution obviously weakens the gel. As discussed in this study, there are two possible reasons for this weakening. (1) The organogelator 12-HOA is partly used as cosurfactant and thus not available for the formation of the gel network. (2) The lamellar phase consisting of water and 2C12DAB is a “better” solvent (different intermolecular interactions) for 12-HOA than n-decane. As regards the structure of the gel fibers we could not detect a difference between the fibers formed in n-decane and in the lamellar phase. 4.2. Influence of the Gel Network on Lα. The evaluation of the SAXS data revealed that the interlayer spacing of the surfactant bilayers in the gelled Lα phases is slightly larger compared to the nongelled Lα phases. Again, it is the incorporation of some 12-HOA molecules in the Lα bilayers which explains this observation. We did not observe any other change than this. 4.3. Orthogonal Self-Assembly: Yes or No? Due to the slight surface-activity of the organogelator a small number of 12-HOA molecules are incorporated in the surfactant bilayer. However, since this incorporation has only a very small effect on the structure of both the gel and the Lα phase, it is justified to call the studied gelled lamellar phases orthogonal selfassembled systems. Nevertheless, our future work will focus on different gelators, which are clearly not surface-active and are thus not incorporated in the surfactant layer. The first one to be tested will be dibenzylidene-D-sorbitol (DBS).47

Figure 6. FFEM pictures of the gelled Lα phase in the system H2O− 2C12DAB−12-HOA at γ = 0.10 and η = 0.05 (top) and at γ = 0.10 and η = 0.03 (bottom). Black and white arrows show the twisted gelator fibers (left) and the lamellar layers in the MLVs (right), respectively. All samples were prepared at room temperature.

Looking at the pictures in Figure 6 one clearly sees the two different structures of the binary systems. Since we did not find both structures close to each other at the same position on the replica, two FFEM pictures are shown for each sample, where either the twisted fibers (left) or the multilayer vesicles (right) are visible. The large multilayer vesicles are 0.2−1.7 μm in diameter. The lamellar repeat distance in the MLVs evaluated from the FFEM pictures of the sample at γ = 0.10 and η = 0.05 is d ≈ 20−25 nm (estimated from seven different pictures), which is in line with the SAXS data (26.1 nm, Table 3) and with calculations for the gelator free system. Unfortunately, it was not possible to determine precisely the lamellar repeat distance for the sample at γ = 0.10 and η = 0.03 from the FFEM pictures, because the fracture was quite unfavorable (and only one MLV was pictured). Nevertheless, the d value is in the expected range and similar to the repeat distance of the sample with γ = 0.10 and η = 0.05 and to the d spacing from SAXS (25.3 nm, Table 3). Small unilamellar vesicles of 20−100 nm in diameter were also observed for the samples at γ = 0.10 at both gelator concentrations. Comparing the fibers formed by 12HOA, one can conclude that the width of the twisted fibers in the gelled lamellar phase is approximately the same as the width of the fibers in the binary gel. The FFEM images provide further experimental evidence that the structures of both the Lα phase and the gel network are hardly disturbed in the gelled Lα phase. On the one hand, the change of the rheological properties, i.e., of the gel properties, is not reflected in the microstructure. On the other hand, the slight change of the interlayer spacing, i.e., of the Lα properties, detected via SAXS can clearly be attributed to 12-HOA acting as cosurfactant. In other words, if 12-HOA were not surface

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.7b02101. Rheology temperature scan and DSC curve, additional rheology data, and calculation of d spacings. (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: +49711-685 64470. G

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir ORCID

(19) Magno, M.; Tessendorf, R.; Medronho, B.; Miguel, M. G.; Stubenrauch, C. Gelled polymerizable microemulsions. 3. Rheology. Soft Matter 2009, 5, 4763−4772. (20) Laupheimer, M.; Jovic, K.; Antunes, F. E.; da Graça Martins Miguel, M.; Stubenrauch, C. Studying orthogonal self-assembled systems: phase behaviour and rheology of gelled microemulsions. Soft Matter 2013, 9, 3661−3670. (21) Laupheimer, M.; Sottmann, T.; Schweins, R.; Stubenrauch, C. Studying orthogonal self-assembled systems: microstructure of gelled bicontinuous microemulsions. Soft Matter 2014, 10, 8744−8757. (22) Kelly, C. V.-D. Perturbations of cellular membranes with synthetic polymers and ultrafast lasers; Ph.D. Thesis; University of Michigan: Ann Arbor, MI, 2009; and ref 12 therein. (23) Xu, Y.; Laupheimer, M.; Preisig, N.; Sottmann, T.; Schmidt, C.; Stubenrauch, C. Gelled Lyotropic Liquid Crystals. Langmuir 2015, 31, 8589−8598. (24) Warr, G. G.; Sen, R.; Evans, D. F.; Trend, J. E. Microemulsion Formation and Phase Behavior of Dialkyldimethylammonium Bromide Surfactants. J. Phys. Chem. 1988, 92, 774−783. (25) Laupheimer, M.; Preisig, N.; Stubenrauch, C. The molecular organogel n-decane/12-hydroxyoctadecanoic acid: sol-gel transition, rheology, and microstructure. Colloids Surf., A 2015, 469, 315−325. (26) Ross-Murphy, S. B.; Shatwell, K. P. Polysaccharide strong and weak gels. Biorheology 2017, 30, 217−227. (27) Tachibana, T.; Mori, T.; Hori, K. Chiral Mesophase of 12Hydroxyoctadecanoic Acid in Jelly and in the Solid State. I. A New Type of Lyotropic Mesophase in Jelly with Organic Solvent. Bull. Chem. Soc. Jpn. 1980, 53, 1714−1719. (28) Kunieda, H.; Shigeta, K.; Ozawa, K.; Suzuki, M. Self-Organizing Structures in Poly(oxyethylene) Oleyl Ether-Water System. J. Phys. Chem. B 1997, 101, 7952−7957. (29) Aramaki, K.; Kunieda, H. Solubilization of Oil in Mixed Cationic Surfactant Liquid Crystal. Colloid Polym. Sci. 1999, 277, 34−40. (30) Keum, J. K.; Burger, C.; Hsiao, B. S.; Somani, R.; Yang, L.; Chu, B.; Kolb, R.; Chen, H.; Lue, C.-T. Synchrotron X-ray scattering studies of the nature of shear-induced shish-kebab structure in polyethylene melt. Prog. Colloid Polym. Sci. 2005, 130, 114−126. (31) Aramaki, K.; Hossain, Md. K.; Rodriguez, C.; Uddin, Md. H.; Kunieda, H. Miscibility of Block Copolymers and Surfactants in Lamellar Liquid Crystals. Macromolecules 2003, 36, 9443−9450. (32) Ishizuka, C.; Arima, S.; Aramaki, K. Head group effects on molecular packing in lamellar liquid crystals. J. Colloid Interface Sci. 2011, 361, 148−153. (33) Marangoni, A. G. Crystallography. In Fat Crystal Network; Marangoni, A. G., Ed.; Marcel Dekker: New York, 2005; p 18. (34) Laupheimer, M. Gelled bicontinuous microemulsions: a new type of orthogonal self-assembled systems; Series Springer Theses, Springer: Heidelberg, Germany, 2014. (35) Fontell, K.; Ceglie, A.; Lindman, B.; Ninham, B.; et al. Some observations on phase diagrams and structure in binary and ternary systems of didodecyldimethylammonium bromide. Acta Chem. Scand. 1986, 40, 247−256. (36) Dubois, M.; Zemb, T. Phase Behavior and Scattering of DoubleChain Surfactants in Diluted Aqueous Solutions. Langmuir 1991, 7, 1352−1360. (37) Dubois, M.; Gulik-Krzywicki, T.; Cabane, B. Growth of Silica Polymers in a Lamellar Mesophase. Langmuir 1993, 9, 673−680. (38) Zemb, T.; Gazeau, D.; Dubois, M.; Gulik-Krzywicki, T. Critical Behaviour of Lyotropic Liquid Crystals. Europhys. Lett. 1993, 21 (7), 759−766. (39) Caboi, F.; Monduzzi, M. Didodecyl-dimethyl-ammonium bromide vesicles and lamellar liquid crystals. A multinuclear NMR and optical microscopy study. Langmuir 1996, 12, 3548−3556. (40) Tachibana, T.; Kambara, H. Studies of helical aggregates of molecules. I. Enantiomorphism in the helical aggregates of optically active 12-hydroxystearic acid and its lithium salt. Bull. Chem. Soc. Jpn. 1969, 42, 3422−3424. (41) Tachibana, T.; Kitazawa, S.; Takeno, H. Studies of Helical Aggregates of Molecules. II. The Sense of Twist in the Fibrous

Cosima Stubenrauch: 0000-0002-1247-4006 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Funding of the Fonds der Chemischen Industrie (FCI) is gratefully acknowledged.

■

REFERENCES

(1) Stubenrauch, C.; Gießelmann, F. Gelled Complex Fluids: Combining Unique Structures with Mechanical Stability. Angew. Chem., Int. Ed. 2016, 55, 3268−3275. (2) Kang, L.; Cheong, H. H.; Chan, S. Y.; Lim, P. F. C. Applications of small-molecule gels − drug delivery. In Soft Fibrillar Materials; Liu, X. Y., Li, J.-L., Eds.; Wiley-VCH: Weinheim, Germany, 2013; pp 115− 128. (3) Kim, D.-H.; Jahn, A.; Cho, S.-J.; Kim, J. S.; Ki, M.-H.; Kim, D.-D. Lyotropic liquid crystal systems in drug delivery: a review. J. Pharm. Invest. 2015, 45, 1−45. (4) Warriner, H. E.; Idziak, S. H. J.; Slack, N. L.; Davidson, P.; Safinya, C. R. Lamellar biogels: fluid-membrane-based hydrogels containing polymer lipids. Science 1996, 271, 969−973. (5) Wang, C.; Chen, D.; Jiao, X. Lyotropic liquid crystal directed synthesis of nanostructured materials. Sci. Technol. Adv. Mater. 2009, 10, 023001. (6) Tolbert, S. H.; Firouzi, A.; Stucky, G. D.; Chmelka, B. F. Magnetic field alignment of ordered silicate surfactant composites and mesoporous silica. Science 1997, 278, 264−268. (7) Kato, T. From Nanostructured Liquid Crystals to Polymer-Based Electrolytes. Angew. Chem., Int. Ed. 2010, 49, 7847−7848. (8) Molecular Gels. Materials with Self-Assembled Fibrillar Networks; Weiss, R. G., Terech, P., Eds.; Springer: Dordrecht, The Netherlands, 2006. (9) Hu, X.-Y.; Xiao, T.; Lin, C.; Huang, F.; Wang, L. Dynamic Supramolecular Complexes Constructed by Orthogonal Self-Assembly. Acc. Chem. Res. 2014, 47, 2041−2051. (10) Kumar, D. K.; Steed, J. W. Supramolecular gel phase crystallization: orthogonal self-assembly under non-equilibrium conditions. Chem. Soc. Rev. 2014, 43, 2080−2088. (11) Saha, M. L.; De, S.; Pramanik, S.; Schmittel, M. Orthogonality in discrete self-assembly − survey of current concepts. Chem. Soc. Rev. 2013, 42, 6860−6909. (12) Yilmaz, M. D.; Huskens, J. Orthogonal supramolecular interaction motifs for functional monolayer architectures. Soft Matter 2012, 8, 11768−11780. (13) Li, S.-L.; Xiao, T.; Lin, C.; Wang, L. Advanced supramolecular polymers constructed by orthogonal self-assembly. Chem. Soc. Rev. 2012, 41, 5950−5968. (14) Atkinson, P. J.; Grimson, M. J.; Heenan, R. K.; Howe, A. M.; Robinson, B. H. Structure of microemulsion-based organo-gels. J. Chem. Soc., Chem. Commun. 1989, 23, 1807−1809. (15) Heeres, A.; van der Pol, C.; Stuart, M.; Friggeri, A.; Feringa, B. L.; van Esch, J. Orthogonal Self-Assembly of Low Molecular Weight Hydrogelators and Surfactants. J. Am. Chem. Soc. 2003, 125, 14252− 14253. (16) Brizard, A. M.; Stuart, M.; van Bommel, K.; Friggeri, A.; de Jong, M.; van Esch, J. H. Preparation of nanostructures by orthogonal selfassembly of hydrogelators and surfactants. Angew. Chem., Int. Ed. 2008, 47, 2063−2066. (17) Stubenrauch, C.; Tessendorf, R.; Strey, R.; Lynch, I.; Dawson, K. Gelled Polymerizable Microemulsions. 1. Phase Behavior. Langmuir 2007, 23, 7730−7737. (18) Stubenrauch, C.; Tessendorf, R.; Salvati, A.; Topgaard, D.; Sottmann, T.; Strey, R.; Lynch, I. Gelled polymerizable microemulsions. 2. Microstructure. Langmuir 2008, 24, 8473−8482. H

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir Aggregates from the Alkali Metal Soaps of Optically Active 12Hydroxystearic Acid. Bull. Chem. Soc. Jpn. 1970, 43, 2418−2421. (42) Uzu, Y.; Sugiura, T. Electron Microscopic and Thermal Studies of the Optical Active 12-Hydroxysearic Acids in Soap Formation. J. Colloid Interface Sci. 1975, 51 (2), 346−349. (43) Hough, L. E.; Jung, H. T.; Krüerke, D.; Heberling, M. S.; Nakata, M.; Jones, C. D.; Chen, D.; Link, D. R.; Zasadzinski, J.; Heppke, G.; Rabe, J. P.; Stocker, W.; Körblova, E.; Walba, D. M.; Glaser, M. A.; Clark, N. A. Helical Nanofilament Phases. Science 2009, 325, 456−460. (44) Sato, H.; Hori, K.; Sakurai, T.; Yamagishi, A. Long distance chiral transfer in a gel: Experimental and ab initio analyses of vibrational circular dichroism spectra of R- and S-12-hydroxyoctadecanoic acid gels. Chem. Phys. Lett. 2008, 467, 140−143. (45) Sato, H.; Sakurai, T.; Yamagishi, A. Comparison of Vibrational Circular Dichroism between the Langmuir-Blodgett Films and Gels of 12-Hydroxyoctadecanoic Acid. Chem. Lett. 2011, 40, 25−27. (46) Sottmann, T.; Strey, R.; Chen, S.-H. A small-angle neutron scattering study of nonionic surfactant molecules at the water-oil interface: area per molecule, microemulsion domain size, and rigidity. J. Chem. Phys. 1997, 106, 6483−6491. (47) Okesola, B. O.; Vieira, V. M. P.; Cornwell, D. J.; Whitelaw, N. K.; Smith, D. K. 1,3:2,4-Dibenzylidene-D-sorbitol (DBS) and its derivatives − efficient, versatile and industrially relevant lowmolecular-weight gelators with over 100 years of history and a bright future. Soft Matter 2015, 11, 4768−4787.

I

DOI: 10.1021/acs.langmuir.7b02101 Langmuir XXXX, XXX, XXX−XXX