Sugar Surfactant Based Microemulsions at Solid ... - ACS Publications

Article pubs.acs.org/Langmuir

Sugar Surfactant Based Microemulsions at Solid Surfaces: Influence of the Oil Type and Surface Polarity Salomé Vargas-Ruiz,† Olaf Soltwedel,‡,§ Samantha Micciulla,† Ramsia Sreij,∥ Artem Feoktystov,⊥ Regine von Klitzing,† Thomas Hellweg,∥ and Stefan Wellert*,† †

Stranski-Laboratorium für Physikalische und Theoretische Chemie, Technische Universität Berlin, Straße des 17 Juni 124, D-10623 Berlin, Germany ‡ Max-Planck-Institute for Solid State Research, Outstation at MLZ, Lichtenbergstr. 1, 85748 Garching, Germany § Physik-Department, Technische Universität München, James-Franck-Str. 1, 85748 Garching, Germany ∥ Physikalische und Biophysikalische Chemie (PC III), Universität Bielefeld, Universitätsstrasse 25, 33615 Bielefeld, Germany ⊥ Jülich Centre for Neutron Science (JCNS) at Heinz Maier-Leibnitz Zentrum (MLZ), Forschungszentrum Jülich GmbH, Lichtenbergstr. 1, 85748 Garching, Germany S Supporting Information *

ABSTRACT: The structure of sugar-surfactant-based bicontinuous microemulsions in the bulk and at hydrophilic and hydrophobic solid planar surfaces was studied by means of neutron scattering techniques (SANS, NR, and GISANS). In particular, the influence of the type of oil (tetradecane and methyl oleate) on the structural properties in the vicinity of surfaces was investigated at different oil-to-water ratios. In the case of hydrophilic surfaces, the analysis of the scattering length density profiles reveals an induced ordering of the oil and water domains perpendicular to the solid−liquid interface in both sets of microemulsions. At hydrophobic surfaces, differences in the near-surface ordering between microemulsions containing polar and nonpolar oils are observed.

1. INTRODUCTION Biodegradable, nontoxic, and environmentally compatible microemulsions are ideal media for use in multiple applications such as drug delivery,1 decontamination and cleaning of sensitive surfaces,2−5 cosmetics and personal care,6 food science,7 and oil recovery,8 among others. The performance of the microemulsions in these applications depends mainly on their ability to wet and adhere to a variety of solid surfaces (e.g., skin, fabrics, metals, polymers, construction materials, etc.) as well as on their capacity to solubilize, extract, or transport substances through a solid−liquid interface. Here, the presence of the solid surfaces can change the physical and chemical properties of the microemulsion and thereby influence its performance in applications. In this respect, a better understanding of the interaction of microemulsions with solid surfaces is essential, as the control and design of these interactions could allow the optimization of the desired processes. The contact between a solid surface and a fluid depends on their interaction potential and can result either in surface melting or ordering of the bulk phase of the fluid in the vicinity of the surface. For example, these effects have been studied for metals, molecular crystals, and also colloidal systems such as thermotropic liquid crystals,9 particle suspensions,10 micellar solutions,11−13 and microemulsions.14 In such studies, surfacesensitive scattering techniques (GISAXS, GISANS, and neutron © XXXX American Chemical Society

and X-ray reflectometry), atomic force microscopy, or surface force measurements are frequently applied.15−18 The influence of the solid surface on the structure of the microemulsions is of particular interest because the surface will support the ordering of complex fluids at the interface.19,20 Previous work studied the near-surface structure of bicontinuous microemulsions on planar surfaces via neutron reflectometry, grazing incidence scattering, and computer simulations.21−23 These studies have shown that the presence of the solid surface induces a structural transition in the microemulsion from bicontinuous to well-defined alternating channels of oil and water (lamellar structure) in the vicinity of the solid−liquid interface. Furthermore, it has been observed by X-ray reflectometry that the presence of the air−liquid interface induces an ordering of microemulsions based on oil droplets dispersed in water (O/W). The structure of the oil droplets is preserved, and the degree of ordering depends on their volume fraction in the bulk.24 These investigations demonstrated an ordering of microemulsion nanodomains due to confinement. Most experiments addressing confinement-induced structures of microemulsions were carried out on nonionic CiEj-type or ionic surfactants and alkane-oil-based microemulsions. Studies related Received: September 19, 2016

A

DOI: 10.1021/acs.langmuir.6b03441 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir to the structuring of microemulsions based on “green” surfactants and biocompatible oils have not yet been reported to the best of our knowledge. Such investigations are crucial because of increasing demands to develop biocompatible and biodegradable media for surface treatments, and hence new model systems are required to understand the influence of microemulsion composition on its structural properties at solid− liquid interfaces. In this study, the influence of planar surfaces of different polarity on the structure of two sugar-surfactant-based microemulsion systems is investigated. The first system contains tetradecane as nonpolar oil phase, and the second system is formulated with biocompatible, slightly polar oil methyl oleate. In both systems, the oil-to-water ratio, and hence the domain size in the bicontinuous region, was varied. The bulk structure of the microemulsions is analyzed via small-angle neutron scattering (SANS). The structure of the microemulsions at the solid−liquid interface is investigated by neutron reflectometry (NR) and grazing incidence small-angle neutron scattering (GISANS). The comparison of the measured bulk and near-surface structure allows us to discuss changes due to the (I) oil composition in the microemulsion, (II) variation of the oil-to-water ratio, and (III) surface polarity.

Table 1. Composition of the Investigated Microemulsion Samples Based on Tetradecane and Methyl Oleate Oil microemulsions based on tetradecane α

γ

δ

0.24 3.00 × 10−2 0.35 6.00 × 10−2 0.40 8.00 × 10−2 0.35 9.00 × 10−1 microemulsions based on methyl oleate

0.1 0.3 0.5 0.7 α

γ

δ

0.1 0.3 0.5 0.7

0.22 0.22 0.34 0.34

6.80 × 10−3 4.00 × 10−2 7.00 × 10−2 8.50 × 10−2

Germany).28,29 The measurements were carried out at sample-todetector distances of 2, 8, and 20 m using a nonpolarized, monochromatic incident beam with a neutron wavelength of λ = 5 Å. In this configuration, the accessible q ranges from 0.003 to 0.3 Å−1. The microemulsion samples were measured in bulk contrast in quartz cells with a 1 mm neutron pathway at 25 °C. Raw scattering data were radially averaged and brought to absolute scale by the procedures provided by the JCNS using the QtiKWS10 software package. The scattering data were further analyzed according to the Teubner− Strey model using the Sasfit software (by J. Kohlbrecher from the Paul Scherrer Institut, Villigen, Switzerland). The Teubner−Strey model is a semiempirical model used to describe the typical broad scattering peaks arising from bicontinuous microemulsion structures.30,31 It yields the domain size dTS and the correlation length ξTS of the microemulsion by

2. EXPERIMENTAL SECTION 2.1. Materials. The oil methyl oleate (Synative ESMETI 05) was provided by BASF (Germany). This oil is a synthetic equivalent of the main component of rapeseed methyl ester. The sugar surfactant Simulsol SL55 (C12/13G1.3) was provided by Seppic (Germany). Pentanol (99%), tetradecane (≥99%), dichlorodimethylsilane (≥99%), anhydrous toluene (99.8%), and ammonium hydroxide solution (28.0−30.0% NH4OH) were purchased from Sigma-Aldrich (Germany). Hydrogen peroxide (30%) was purchased from ChemSolute, D2O (heavy water) was purchased from Eurisotop (France), and polished silicon blocks of (8 × 5 × 1) cm3 with a surface roughness of 0 are the fitting parameters, and q is the magnitude of the scattering vector. The renormalized and the bare bending elasticity constants κSANS and κbare of the surfactant membrane in the microemulsions32,33 can be calculated from the structural parameters (domain size dTS and correlation length ξTS) obtained from the Teubner−Strey analysis, where lc represents the thickness of the surfactant membrane. 2

2

κSANS 10 3π ξTS = kBT 64 d TS

(4)

κbare κ 3 ⎛ d TS ⎞ = SANS + ln⎜ ⎟ kBT kBT 4π ⎝ 2lc ⎠

(5)

2.5. Neutron Reflectometry. Neutron reflectometry (NR) measurements were carried out on the NREX reflectometer at the Heinz Maier-Leibnitz Zentrum (Garching, Germany). The neutron reflectometry measurements are based on the determination of the ratio between the intensity of the specularly reflected beam from the solid− liquid interface and the intensity of the incident neutron beam (λ = 4.3 Å). Such a ratio is defined as the reflectivity, R = I/I0, and R is evaluated B


Article

Langmuir

fields in the free-energy functional.35−37 Similar theoretical concepts are used for the description of planar amphiphilic membranes adhered to a planar solid substrate.38 The scattering length density profile of the nearsurface structure of the bicontinuous microemulsion contains an oscillatory contribution describing the alternating scattering length densities of the layered oil and water domains. The exponential term describes the transition of this layered structure into the randomly oriented bicontinuous bulk structure, represented by a mean scattering length density of the bulk medium. Parameters A and φ are the amplitude and phase of the oscillation, and dz and ξz are the mean repeat distances of oil and water domains and the correlation length of this layered structure. As a consequence of the theoretical approach, both parameters should be identical to corresponding parameters dTS and ξTS in the bulk phase. In the fitting process, A, φ, dz and ξz are taken as free parameters, and the SLD of the pure silicon, the silicon oxide layer (20 Å), and the bulk of the microemulsions were kept constant (Table 2). This approach leads

as a function of the momentum-transfer vector Qz. Figure 1 sketches the scattering geometry in the neutron reflectometry- and GISANS-type measurements.

Table 2. Scattering Length Density of the Pure Components of the Microemulsion and Solid Surfaces

Figure 1. Scheme of the scattering geometry in the neutron reflectometry and GISANS measurements. Importantly, below the critical edge of total external reflection, only mirrorlike reflectivity occurs, whereas above it, the intensity decreases as the fourth power of the incident angle, as described by the Fresnel reflectivity. This restricts the accessible Qz range from the critical edge up to the angle where the signal-to-noise ratio has the same order of reflectivity. Thus, a background correction is mandatory. In fact, the specular reflected signal-to-noise ratio depends not only on Qindependent factors such as the intrinsic background level in the surroundings and incoherent scattering generated by the sample and sample holder but also on Q-dependent factors such as the footprint of the direct beam and diffuse scattering at the sample and its holder.34 A general prediction of the background is therefore hopeless and hence has to be done for each reflectivity curve individually. As shown by the GISANS scans, the investigated microemulsions generate individually strong off-specular scattering. In that case, the most promising strategy estimating the background is using the offspecular reflected intensity in the vicinity of the specular condition. This method has two major advantages: (i) the background is a Qz-dependent estimate and (ii) if a second detector is present in the vicinity of the specular reflected beam, then the required data can be collected while the specular signal is measured. In this experiment, we made use of a position-sensitive detector, which allows shaping the sensitive detector area arbitrarily. Practically two rectangular ranges of interests (ROIs) with identical areas were applied to mime the detectors. Although the ROI for the specular reflectivity matches the condition θf = θi, the offspecular signal is recorded below and above this region (θf = θi ± 0.12°). The geometry is in general described by 2⎞ ⎛ 2 ⎞ ⎛ ⎛Q x ⎞ 2π ⎜ cos(θf ) − cos(θi)⎟ π ⎜ θi − θf ⎟ ≅ ⎜ Q (θi , θf ) = ⎜⎜ ⎟⎟ = ⎟ ⎜ λ ⎝ sin(θf ) + sin(θi) ⎠ λ ⎝ 2θf + 2θi ⎟⎠ ⎝Q z ⎠

SLD (Å−2)

Si SiO2 D2 O SL55 surfactant pentanol methyl oleate tetradecane

2.07 × 10−6 3.47 × 10−6 6.36 × 10−6 −1.16 × 10−6 −3.29 × 10−7 2.39 × 10−7 −4.38 × 10−7

to the best fit results, especially in the peak region and at higher Q values. The SLD of the bulk microemulsions was calculated according to their composition. Before the measurements, the treated hydrophilic and hydrophobic silicon blocks were placed in a liquid cell trough made of poly(tetrafluoroethylene) with stainless steel inlet and outlet tubes. The contact between the silicon blocks and the trough was sealed by using a Viton O-ring. Approximately 10 mL of microemulsion samples was injected into the cell. The inlet and outlet lines were closed, and the sample inside the cell was equilibrated for 1 h. Subsequently, the neutron reflectometry measurements of the solid−liquid interface were carried out at 25 °C. 2.6. Grazing Incidence Small-Angle Neutron Scattering. Small-angle neutron scattering under grazing incidence was measured at NREX (MLZ, Garching, Germany). Compared to the setup described for measuring the neutron reflectivity, the resolution in the in-plane direction Qy had to be increased. Therefore, the incoming beam was collimated by two slits (separated 2 m from each other) down to 1 × 1 mm2. Because the scattering length density of the subphase differs with the oil/water ratio and the incoming beam strikes the interface through the silicon substrate, a critical edge (θc) occurs only for high D2O content in the microemulsions, namely, for α = 0.1 and 0.3 of the tetradecane-based microemulsions. Thus, an evanescent wave is present only for the two most dilute systems. To maximize the neutrons in the vicinity of the solid−liquid interface, the incident angle θi was fixed to 0.18° for α = 0.1 and to 0.15° for all other oil/water ratios investigated. The resulting penetration depths Λ of the evanescent wave are calculated according to

(6)

Data analysis was done by applying the Parratt recursion algorithm (Parratt 32 software, by J. C. Braun from Helmholtz-Center Berlin, Germany) and using a scattering length density profile (SLD) of the form

⎛⎛ 2πz ⎞ ⎞ SLD = Ae(−z / ξz) cos⎜⎜⎜ ⎟ + ϕ⎟⎟ ⎝⎝ dz ⎠ ⎠

component

Λ=

λ 4π θc 2 − θi 2

(8)

Table 3 summarizes the parameters of the grazing incidence measurements. The 20 × 20 cm2 position-sensitive detector placed 2.5 m behind the sample counts vertical and horizontal scattered neutrons within a 4.5° solid angle. To enlarge the covered Q space during a single measurement, we focused on the first quadrant (only positive Q vectors). Therefore, the detector was moved out of the scattering plane

(7)

which results from an oscillatory one-dimensional order parameter profile in the microemulsion. On the basis of Ginzburg−Landau theory, this order parameter profile was obtained by including additional surface C


Article

Langmuir

results are affected by the intrinsic approximations of the model. The results and the calculated renormalized and bare bending moduli (eqs 4 and 5) are summarized in Table 4.

Table 3. Critical Angle (θc), Incident Angle (θi), and Penetration Depth (Λ) of the Neutron Beam at the Solid/ Liquid Interface for Tetradecane-Based Microemulsions at Different Oil Contents (α) α

θc (deg)

θi (deg)

Λ (Å)

0.1 0.3 >0.3

0.19 0.16

0.18 0.15 0.15

320 190 ∞

Table 4. Structural Parametersa of the Tetradecane- and Methyl Oleate-Based Microemulsions tetradecane-based microemulsion α 0.1 0.3 0.5 0.7

by 2θhor = 1.6°, and in the scattering plane to θf = 1.6°. As a result, direct and reflected beams are located in a detector corner, and the maximum Q space is covered. The highly collimated beam decreased the incoming neutron flux dramatically. To account for this, each GISANS pattern was summed over 12 h of counting time. The position-sensitive detected events were then transformed into a lateral component Qxy Q xy =

Q x2 + Q y2

(9)

where

Qx =

2π (cos(θf ) cos(2θ hor) − cos(θi)) λ

(10)

Qy =

2π (cos(θf ) sin(2θ hor)) λ

(11)

2π (sin(θf ) + sin(θi)) λ

κSANS (kBT)

127 ± 7 54 ± 1 0.36 ± 0.04 97 ± 8 58 ± 1 0.51 ± 0.04 87 ± 5 44 ± 1 0.42 ± 0.02 87 ± 2 30 ± 1 0.29 ± 0.01 methyl oleate-based microemulsion

κbare(kBT) 0.70 ± 0.04 0.78 ± 0.08 0.68 ± 0.02 0.54 ± 0.01

α

d (Å)

ξ (Å)

κSANS (kBT)

κbare (kBT)

0.1 0.3 0.5 0.7

128 ± 3 172 ± 6 107 ± 3 108 ± 2

47 ± 1 61 ± 1 41 ± 1 26 ± 1

0.31 ± 0.02 0.30 ± 0.03 0.32 ± 0.02 0.20 ± 0.02

0.67 ± 0.02 0.73 ± 0.03 0.64 ± 0.02 0.53 ± 0.02

The domain sizes are smaller for TD-microemulsions in comparison to MO-microemulsions for α larger than 0.1. This difference becomes more significant when the oil content in both series of microemulsions is α = 0.3 with differences of about 70 Å. At this oil content, the MO-microemulsion reaches a maximum in the domain size, while the domain size decreases continuously by increasing the oil content for the TD-microemulsions. The correlation length of both series of microemulsions follows the same trend, reaching a maximum at α = 0.3. Elastic constants κSANS and κbare have values of between 0.3kBT and 0.8kBT, indicating soft interfaces. The data suggest that the chemical composition of the oil has an influence on the elasticity of the surfactant membrane in the microemulsion. For instance, TD microemulsions show renormalized bending moduli of around 20 up to 60% larger than renormalized bending moduli of MOmicroemulsions. 3.2. Microemulsion Structure at the Solid−Liquid Interfaces. 3.2.1. Structure in the Vertical Direction. Figure 3 shows the reflectivity curves of TD-microemulsions with four different oil-to-water ratios α in contact with a hydrophilic (Figure 3a) and a hydrophobic (Figure 3b) surface. Because D2O is the dominant component in both investigated micro-

(12)

Because the diffuse scattering is Debye−Scherrer−Ring-like distributed, the data were finally radially (Q =

ξ (Å)

a Domain size (d), correlation length (ξ), and bending modulus (renormalized (κSANS) and bare (κbare).

and a vertical component

Qz =

d (Å)

Q xy 2 + Q z 2 ) sorted and

−1

binned (ΔQ = 0.001 Å ).

3. RESULTS 3.1. Microemulsion Structure in the Bulk. Figure 2 depicts the scattering curves of the tetradecane (TD)-based microemulsions and methyl oleate (MO)-based microemulsions. In all data sets of the two series of microemulsions, a well-defined correlation peak can be observed, which shifts to larger q values and vanishes with increasing oil content (increasing α) in the microemulsions. The structural parameters of both series of microemulsions have been estimated from the scattering curves by means of the Teubner−Strey model. It is worth mentioning that this model was derived for symmetric microemulsions. It is not strictly valid for asymmetric oil-to-water ratios. However, we apply the model also in this case, taking into account that these

Figure 2. Scattering curves of tetradecane- (a) and methyl oleate-based (b) microemulsions prepared at different oil-to-water ratios α as a function of q. Solid lines are fits to the scattering data according to the Teubner−Strey model (eq 1). D


Article

Langmuir

Figure 3. Reflectivity curve of tetradecane-based microemulsions at (a) a hydrophilic surface and (b) a hydrophobic surface. Solid lines correspond to the fit of the scattering data according to the Parratt model, and the corresponding SLD profiles for the hydrophilic and hydrophobic surfaces are presented in panels c and d.

microemulsions show well-defined oscillations for all α at the hydrophilic surface, whereas the oscillations are strongly damped at the hydrophobic surface at increasing α. Figure 5 shows the SLD values in the first SLD minimum after the SiO2 layer for both series of microemulsions as a function of α. These values qualitatively indicate the preferentially adsorbed and accumulated molecules at the solid−liquid interface. In general, the SLD values of both series of microemulsions exhibit the same behavior at the hydrophilic solid−liquid interface, where the SLD value decreases at increasing α and reaches a plateau value at higher oil content. The change in the surface polarity from hydrophilic to hydrophobic results in a change in the SLD values near the interface. The SLD values of the TDmicroemulsions at the hydrophobic surface slightly decrease, while the values of the MO-microemulsions increase with increasing α. Figures 6 and 7 plot domain size dz and correlation length ξz near the surface as determined from the SLD profiles (eq 7) for both series of microemulsions at the hydrophilic and hydrophobic surfaces, which are compared to the structural parameters determined in the bulk (Table 4). The structural parameters of the microemulsions at the hydrophilic and hydrophobic surfaces are in the same range as their corresponding structural parameters in the bulk. Generally, the domain size and correlation length decrease with increasing α for the TDmicroemulsions. For the MO-microemulsions, a maximum in both structural parameters was found when the content of oil in this microemulsion was equal to α = 0.3. The structural parameters for the MO-microemulsions with higher oil content at the hydrophobic surface could not be computed because of the

emulsions, a critical angle of total external reflection (critical wave vector transfer Qz,c) is detectable only at the lowest oil concentrations (α = 0.1) At higher oil contents (α ≥ 0.3), Qz,c was not detectable in the studied Qz regime as a result of the given contrast. Importantly, the reflectograms of the TD-microemulsions reveal a distinct Bragg peak that broadens and shifts to higher Qz at increasing oil content in the microemulsion. The Bragg peak originates from the structuring of the microemulsion at the solid/liquid interface. The broadening of the Bragg peak is interpreted as the deorientation of the microemulsion structure in the direction perpendicular to the surface. The reflectivity curves can be described by eq 7 because of the oscillating variation of the scattering length densities related to alternating oil- and waterrich regions of the TD-microemulsions in the vicinity of the hydrophilic (Figure 3c) and hydrophobic (Figure 3d) surfaces, respectively. First, the profiles show a constant value due to the silicon substrate (situated at negative z values), followed by a layer of SiO2 with a thickness of 15 Å and an SLD value of 3.47 × 10−6 Å−2 at z = 0. Similar oscillation behavior exists in both systems where the SLD oscillations decay over one to four oscillations to the bulk value at z → ∞. Figure 4 depicts the reflectivity curves and the corresponding SLD profiles of MO-microemulsions at a hydrophilic (Figure 4a,c) and a hydrophobic (Figure 3b,d) surface. The reflectivity curves of the MO-microemulsions show the same features as the reflectivity curves of the TD-microemulsions. However, the SLD profile of the MO-microemulsions at the hydrophilic solid surface indicates a difference in the near-surface structure when it is compared to the hydrophobic case. In particular, the MOE


Article

Langmuir

Figure 4. Reflectivity curves of methyl oleate-based microemulsions at (a) a hydrophilic surface and (b) a hydrophobic surface. Solid lines correspond to the fit of the scattering data according to the Parratt model, and the corresponding SLD profiles for the hydrophilic and hydrophobic surfaces are presented in panels c and d.

This enhanced scattering has a ringlike distribution. This feature of the scattering intensity distributions becomes clearly visible in the transformation of the scattering intensity distribution shown in the lower row of Figure 8. Here, the scattering intensity is mapped as a function of Q and polar angle ϕ, with ϕ = a tan(Qxy/ Qz). In this representation, the enhanced ringlike scattering transforms into a smeared but nearly horizontal distribution. This illustrates that the same structure is observable in all lateral directions. Figure 9 plots the enhanced scattering as a function of Q, with Q = (Qxy2 + Qz2)1/2 and binning (ΔQ = 0.001 Å−1). The resulting intensity distributions show broad scattering peaks in the Q range between 0.05 and 0.08 Å−1. Qualitatively, they correspond to the features observed in the small-angle scattering in the bulk. The solid lines in Figure 9 are fits of the Teubner−Strey model to the averaged diffuse scattering intensity profiles. The analysis focuses on the peak region, which is well described by the fits. At larger Q, the intensity decrease and the background level are not described by the model as a result of the limited precision of the background correction. However, domain size d and correlation length ξ were extracted from the fits. The resulting structural parameters are summarized in Table 5. The quality of the fits limits the precision of the estimation to about 20%.

Figure 5. SLD values near the solid−liquid interface for tetradecanebased microemulsions (squares) and methyl oleate-based microemulsions (circles) at the hydrophilic (closed symbols) and hydrophobic (open symbols) surfaces.

extremely thin surface layer and the subsequent fast transition of the SLD to the corresponding bulk-phase properties (Figure 4d). 3.2.2. Structure in the Lateral Direction. The upper row of Figure 8 shows the scattering intensity distributions measured with the TD-microemulsions at the hydrophilic functionalized surface at different oil-to-water ratios α. The scattering intensity is mapped as a function of Qz and Qxy. At low Qz and Qxy, the transmitted and specular reflected beams are visible. Additionally, a broad Bragg peak at Qz = 0.03−0.04 Å−1 is slightly visible.

4. DISCUSSION 4.1. Microemulsion Structure in the Bulk. The results of the bulk structure characterization by SANS summarized in Table 4 indicate that the oil influences their structural properties. The domain size is smaller for the TD-microemulsions compared to the domain size of the MO-microemulsions at the same α. F


Article

Langmuir

Figure 6. Comparison of the domain size d and correlation length ξ structural parameters of tetradecane-based micreomulsions in the bulk phase and at hydrophilic and hydrophobic solid surfaces.

Figure 7. Comparison of the domain size d and correlation length ξ structural parameters of methyl oleate-based micreomulsions in the bulk phase and at hydrophilic and hydrophobic solid surfaces.

Figure 8. (Upper row) GISANS pattern of tetradecane-based microemulsions at the hydrophilic functionalized surface at different oil-to-water ratios α. The scattering intensity is mapped as a function of Qz and Qxy. (Lower row) Polar coordinate representation of the same scattering intensity distributions as a function of Q and ϕ (ϕ = a tan(Qxy/Qz)).

preferentially adopted because the trans-conformation of the alkane chain has the lowest energy.39 In contrast, the methyl oleate molecule has a bent structure due to the presence of a cis

This difference could be attributed to the packing of the molecules in the oil domains. Tetradecane molecules can densely pack in the domains because of their linear structure, which is G


Article

Langmuir

(sugar surfactant and pentanol as cosurfactants), the differences could be related to the interaction between the oil molecules and the amphiphiles. In this respect, it has been probed via diffusion NMR spectroscopy that the unsaturated fatty acid ethyl oleate can penetrate the surfactant membrane of nonionic microemulsions.46 In a similar fashion, the methyl oleate with a chemical structure similar to that of ethyl oleate could penetrate the surfactant membrane. This could enhance the fluidity of the membrane, which in turn would result in lower elastic constants. Furthermore, a more flexible surfactant membrane could reduce the near-surface ordering. 4.2. Microemulsion Structure at the Solid−Liquid Interfaces. 4.2.1. Hydrophilic Interface. Every neutron reflectivity curve of both systems at hydrophilic and hydrophobic interfaces was fitted satisfactorily using the SLD profile given by eq 7. The oscillations in the derived SLD are damped over a distance to the solid substrate of up to 400 Å. Both systems exhibit the same behavior. This is also in agreement with previous experimental and numerical results described for bicontinuous microemulsions containing nonionic surfactant C10E4.21−23 The observed behavior is attributed to a wetting of the solid substrate by a lamellar phase and its subsequent transition into the bicontinuous structure in the bulk. Most likely, the same transition occurs for the TD and MO microemulsions at the hydrophilic interface. However, the comparison of the structural parameters shown in Figures 6 and 7 indicates that the domain size and correlation length are the same in the bulk and at the hydrophilic surface. This suggests only a weak influence of the lamellar ordering and a fast transition into the bulk structure or, as theoretically predicted, a nonwetting of the surface by the lamellar phase.47 Speculatively, the structuring of the sugar surfactant-based microemulsions at the hydrophilic solid−liquid surfaces results from the patchlike arrangement of oil domains in the water-rich region or vice versa, as the maximum and minimum SLD values deviate from the values of the pure components (oil or water). The only exceptions are the lowest oil content microemulsions (α = 0.1), where the first SLD maximum value indicates a region rich in pure D2O, followed by an oil-rich region containing D2O domains. Moreover, Figure 5 suggests similar SLD values close to the hydrophilic interface for both microemulsion systems. Possibly, the surfactant molecules could be preferentially adsorbed at the hydrophilic interface, and the surface coverage depends on the oil content. At low oil contents, the hydrophilic solid−liquid interface is assumed to be covered by patches of surfactant and water, and the increasing oil content results in an enrichment of the surfactants at the interface. In this case, perforated rather than perfect lamellae would be in contact with the substrate, although a clear experimental distinction between both cases is not given by the measurements. 4.2.2. Hydrophobic Interface. Although the TD- and the MO-microemulsions show the same behavior at the hydrophilic surface, the hydrophobic surface supports the near-surface ordering of the TD-microemulsions. In contrast to this, the ordering of the MO-microemulsion is suppressed at the hydrophobic surface at higher oil content. This is reflected by the resulting SLD-profiles in Figures 4 and 3. Nevertheless, domain sizes and correlation lengths also correspond to the results found in the bulk. Because there is almost no difference in these results (Figures 6 and 7), again a weak influence of a lamellar structure exists.

Figure 9. Radially averaged diffuse scattering intensity for selected samples as a function of Q. The inset enlarges the peak region of the scattering curves. The solid black lines are fits of the Teubner−Strey model to the data.

Table 5. Summary of the Resulting Structural Parameters from Fitting of the Teubner−Strey Model to the Data in Figure 9 sample

surface polarity

d (Å)

ξ (Å)

T1 T3 T5 T5

hydrophilic hydrophilic hydrophilic hydrophobic

131 108 90 90

20 21 37

double bond. This structural conformation reduces the degree of packing of the methyl oleate and thus induces an increment in the domain size. The elastic properties of the surfactant membrane were characterized by the determination of the renormalized bending modulus and the bare bending modulus. The renormalized bending modulus is the effective bending modulus of the membrane on the scale of the domain size. At this scale, the bending of the membrane is superimposed by thermal fluctuations at smaller scales such that the larger the thermal fluctuations, the easier it is to deform the membrane.32,40 Conversely, the bare bending modulus is associated with the high momentum cut off, given by the effective thickness of the amphiphilic films, lc.33,41 The values of the renormalized and bare bending modulus for both series of microemulsions are between 0.3kBT and 0.8kBT. This is slightly lower than bending moduli commonly reported in the literature for nonionic microemulsion systems, for example, microemulsions stabilized with CiEj or sugar surfactants.42−44 This suggests that the amphiphilic interface is highly flexible at all α.45 Nevertheless, it is important to point out that the renormalized bending moduli are comparatively larger for TD-microemulsions than for MOmicroemulsions. Taking into account that the composition of the surfactant membrane is the same for both sets of microemulsions H


Article

Langmuir

and its tendency to form comparatively small domains with low ordering, as indicated by the short correlation lengths. These findings complement previously reported results from bicontinuous microemulsions based on the C10E4 surfactant where a pronounced lamellar ordering was induced by the solid substrate.23

The different ordering of the two microemulsion systems at the hydrophobic surface is likely due to the wetting properties and the orientation of the oil molecules at the hydrophobic solid−liquid interface (Figure 5). From contact angle measurements, it was found that tetradecane (TD) has a lower static contact angle ((23 ± 3)°) at the hydrophobic surface than methyl oleate (MO) ((33 ± 4)°). Additionally, the arrangement of the TD and MO molecules at the hydrophobic surface might have a significant influence. It was shown via force−distance measurements that the symmetry of liquid molecules determines their ability to pack and orient at solid surfaces.48 For example, nalkane molecules can easily pack and orient parallel to bare solid surfaces, whereas asymmetric molecules (e.g., the introduction of the methyl groups in the backbone) cannot. Possibly, the adsorption of tetradecane at the surface favors the near-surface ordering of these microemulsions. In contrast, the different arrangement of the methyl oleate molecules at the hydrophobic solid interface could increase the roughness of the interface, which in turn hinders the effective formation of lamellae at the hydrophobic solid−liquid interface. This in turn corresponds to the nonwetting of the surface by a lamellar phase, and an oil-rich and strongly perforated structure of the microemulsion close to the surface might result. The conservation of characteristic bulk structural lengths under confinement at solid substrates is in good agreement with observations on other colloidal systems, such as polyelectrolytes and colloidal particles. A transient network of polyelectrolyte chains shows similar values for interchain distance, correlation length, and interaction strength, suggesting no compression of the network in the confined geometry.49 Colloidal particles, dominated by electrostatic repulsion and in a low-pressure reservoir under confinement, are still randomly distributed in a fluidlike manner as they are in the bulk.50 4.3. Lateral Structure. The measurements in grazing incidence scattering geometry have been used to detect offspecular scattering at small and also virtually infinite penetration depths Λ. In the first case, the lateral structure was probed at a penetration depth corresponding to a few domain sizes of the bicontinuous structure in the bulk. In the second configuration, the penetration is determined by the damping of the penetrating neutron wave by absorption processes. However, the experimental results shown in Figures 8 and 9 show no scattering feature corresponding to a lamellar structure. This indicates either no or only very few lamellae in the vicinity of the surfaces. Moreover, the observed enhanced off-specular scattering intensity distribution shows no spotlike features as in the case of crystalline order but rather a ringlike distribution, which qualitatively corresponds to the bulk scattering of bicontinuous microemulsions. The structural parameters shown in Table 5 obtained from fitting the Teubner−Strey model to the data from the TD system in Figure 9 compare very well to the results given in Table 4 obtained from SANS in the bulk. These results also suggest a conservation of the bicontinuous structure parallel to the substrate. Instead of extended oil or D2O layers, a rather patchy arrangement of oil and D2O with domain sizes comparable to the bulk values seems to occur on both types of surfaces at the penetration depths probed in the experiment. The fast decay of the scattering profiles shown in Figures 3 and 4 is in good agreement with the results of the GISANS measurements. Because most of the effects have been found in both microemulsion systems, we assume that the surfactant properties are of importance for the near-surface structuring. We attribute the observations to the amphiphilicity of the sugar surfactants

5. CONCLUSIONS Knowledge of the near-surface ordering of complex liquids is of utmost importance when applications require the contact between these liquids and solid surfaces. Moreover, the investigation of surface ordering and melting processes are of fundamental interest. Here, we studied the near-surface ordering of sugar-surfactant-based bicontinuous microemulsions with different oil-to-water ratios at hydrophilic and hydrophobic surfaces. The structure of these microemulsions at solid surfaces was investigated using neutron scattering techniques. Two sets of microemulsions were prepared by using tetradecane (nonpolar) and methyl oleate (polar) as the oil phase. The bulk structure of the two series of microemulsions was characterized by SANS at different oil-to-water ratios (α). At all α, the tetradecane-based microemulsions have a smaller domain size and larger renormalized bending elasticities in comparison to the methylbased microemulsions. Compared to the bending elasticities of CiEj surfactant-based microemulsions, our results are smaller. Also, the structural lengths of the two microemulsion systems are smaller.42 The reason is a rather disordered structure resulting from the use of the amphiphilic components. This observation raised the question of how the observed bulk structure changes when the microemulsion is in close contact with a hydrophilic and hydrophobic solid surface. Neutron scattering in reflectometry and the GISANS mode have been applied to investigate this aspect. The neutron reflectometry data indicates that the bulk structure continuously deforms into a more layered structure in the vicinity of the substrates. The presence of one to four oscillations in the vertical scattering length density profile supports the assumption of a transition over small vertical distances of less than 103 Å. This corresponds to the observation of small domain sizes and correlation lengths in the bulk. Moreover, a quantitative comparison of the structural parameters showed no difference between the bulk and nearsurface region. This was predicted by theory for the case of the nonwetting of a solid substrate by a lamellar phase.47 Hence, we assume a weak contribution of the lamellar phase to the nearsurface structure. Additionally, GISANS data gave no indication of a lateral ordering in addition to the propagation of the bicontinuous structure. Also in the lateral direction, within the precision of the experiment, the same domain sizes were observed. These findings suggest that a lamellar structure exists only very close to the substrate. Perforations in the layers or regions with patchy oil and water domains seem to be responsible for the fast transition found for the systems. Furthermore, a difference in the ordering was found between the apolar tetradecane and the polar methyl oleate at the hydrophobic surface. The different oil components in the two series of microemulsions also influence the near-surface structure of the solid−liquid interface. Here, the scattering length density profiles reveal that both tetradecane- and methyl oleate-based microemulsions develop a well-defined structure in the direction perpendicular to the hydrophilic solid−liquid interface. This result is in good agreement with previous experimental findings. On the contrary, the structuring of the methyl oleate microemulsions is hindered at the hydrophobic solid−liquid interface, I


Article

Langmuir

(8) Bera, A.; Mandal, A. Microemulsions: a novel approach to enhanced oil recovery: a review. J. Pet. Explor. Prod. Technol. 2015, 5, 255−268. (9) Lang, P.; Steitz, R.; Braun, C. Surface effects of lyotropic liquid crystalline phases of nonionic surfactants. Colloids Surf., A 2000, 163, 91−101. (10) Klapp, S. H. L.; Zeng, Y.; von KLitzing, R. Surviving Structure in Colloidal Suspensions Squeezed from 3D to 2D. Phys. Rev. Lett. 2008, 100, 118303. (11) Gerstenberg, M. C.; Pedersen, J. S.; Smith, G. S. Surface induced ordering of micelles at the solid-liquid interface. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1998, 58, 8028−8031. (12) Gerstenberg, M. C.; Pedersen, J. S.; J, M.; Smith, G. S. Surface Induced Ordering of Triblock Copolymer Micelles at the Solid-Liquid Interface. 1. Experimental Results. Langmuir 2002, 18, 4933−4943. (13) Gerstenberg, M. C.; Pedersen, J. S. Surface Induced Ordering of Triblock Copolymer Micelles at the Solid-Liquid Interface. 2. Modeling. Langmuir 2001, 17, 7040−7046. (14) Lang, P. Surface induced ordering effects in soft condensed matter systems. J. Phys.: Condens. Matter 2004, 16, R699−R720. (15) Fragneto-Cusani, G. Neutron reflectivity at the solid/liquid interface: examples of applications in biophysics. J. Phys.: Condens. Matter 2001, 13, 4973. (16) Majewski, J.; Kuhl, T. L.; Wong, J. Y.; Smith, G. S. X-ray and neutron surface scattering for studying lipid/polymer assemblies at the air-liquid and solid-liquid interfaces. Rev. Mol. Biotechnol. 2000, 74, 207− 231. (17) Müller-Buschbaum, P. Grazing incidence small-angle neutron scattering: challenges and possibilities. Polym. J. 2013, 45, 34−42. (18) Petrov, P.; Olsson, U.; Wennerström, H. Surface forces in bicontinuous microemulsions: Water capillary condensation and lamellae formation. Langmuir 1997, 13, 3331−3337. (19) Bowers, J.; Vergara-Gutierrez, M. C.; Webster, J. R. P. Surface Ordering of Amphiphilic Ionic Liquids. Langmuir 2004, 20, 309−312. (20) Hamilton, W. A.; Porcar, L.; Butler, P. D.; Warr, G. G. Local membrane ordering of sponge phases at a solid-solution interface. J. Chem. Phys. 2002, 116, 8533−8546. (21) Zhou, X.-L.; Lee, L.-T.; Chen, S.-H.; Strey, R. Observation of surface-induced layering in bicontinuous microemulsions. Phys. Rev. A: At., Mol., Opt. Phys. 1992, 46, 6479−6489. (22) Lee, D. D.; Chen, S. H.; Majkrzak, C. F.; Satija, S. K. Bulk and surface correlations in a microemulsion. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1995, 52, R29−R32. (23) Kerscher, M.; Busch, P.; Mattauch, S.; Frielinghaus, H.; Richter, D.; Belushkin, M.; Gompper, G. Near-surface structure of a bicontinuous microemulsion with a transition region. Phys. Rev. E 2011, 83, 030401. (24) Kraska, M.; Domschke, M.; Stühn, B. Concentration induced ordering of microemulsion droplets in bulk and near the liquid−air interface. Soft Matter 2013, 9, 3488−3496. (25) Langevin, D. Micelles and microemulsions. Annu. Rev. Phys. Chem. 1992, 43, 341−369. (26) Strey, R. Microemulsion microstructure and interfacial curvature. Colloid Polym. Sci. 1994, 272, 1005−1019. (27) Hellweg, T.; von Klitzing, R. Evidence for polymer-like structures in the single phase region of a dodecane/C12E5/water microemulsion: a dynamic light scattering study. Phys. A 2000, 283, 349−358. (28) Feoktystov, A. V.; Frielinghaus, H.; Di, Z.; Jaksch, S.; Pipich, V.; Appavou, M.-S.; Babcock, E.; Hanslik, R.; Engels, R.; Kemmerling, G.; et al. KWS-1 high-resolution small-angle neutron scattering instrument at JCNS: current state. J. Appl. Crystallogr. 2015, 48, 61−70. (29) Frielinghaus, H.; Feoktystov, A.; Berts, I.; Mangiapia, G. KWS-1: Small-angle scattering diffractometer. Journal of large-scale research facilities JLSRF 2015, 1, A28. (30) Teubner, M. Scattering from 2-phase random media. J. Chem. Phys. 1990, 92, 4501−4507. (31) Strey, R.; Winkler, J.; Magid, L. Small-angle neutron-scattering from diffuse interfaces 0.1. monolayers and bilayers in the water octane C12E5 system. J. Phys. Chem. 1991, 95, 7502−7507.

and the structuring of the tetradecane-based microemulsion is preserved. This effect at the hydrophobic surfaces is related to the polarity and adsorption behavior of the used oils. The results are of practical importance for the use of bicontinuous microemulsions for extraction or decontamination media. Here, the extraction of mostly lipophilic molecules (e.g., contaminants) out of the surface is an important task. Modeling experimental extraction data requires a knowledge of potential barriers and pathways out of the surface region. Speculatively, differences in the extraction may occur between systems tending to lamellar ordering, perforated lamellar structures, or less distorted bicontinuous structures in the vicinity of a solid substrate, and knowledge of the ordering can support such studies.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.6b03441. Phase diagrams of tetradecane-based microemulsions at different oil/water ratios (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS Methyl oleate and surfactant solution Simusol SL55 were kindly provided by BASF and Seppic GmbH. We gratefully acknowledge funding by the Deutsche Forschungsgemeinschaft (DFG) (grants HE2995/3-1 and WE5066/1-1). This work is based upon experiments performed at the KWS-1 instrument operated by JCNS and at the NREX instrument at the MLZ. The authors gratefully acknowledge the financial support provided by the MLZ.

■

REFERENCES

(1) Lawrence, M.; Rees, G. D. Microemulsion-based media as novel drug delivery systems. Adv. Drug Delivery Rev. 2000, 45, 89−121. (2) Mulligan, C.; Yong, R.; Gibbs, B. Surfactant-enhanced remediation of contaminated soil: a review. Eng. Geol. 2001, 60, 371−380. (3) Giorgi, R.; Baglioni, M.; Berti, D.; Baglioni, P. New Methodologies for the Conservation of Cultural Heritage: Micellar Solutions, Microemulsions, and Hydroxide Nanoparticles. Acc. Chem. Res. 2010, 43, 695−704. (4) Wellert, S.; Imhof, H.; Dolle, M.; Altmann, H.-J.; Richardt, A.; Hellweg, T. Decontamination of chemical warfare agents using perchloroethylene−Marlowet IHF−H2O-based microemulsions: wetting and extraction properties on realistic surfaces. Colloid Polym. Sci. 2008, 286, 417−426. (5) Vargas-Ruiz, S.; Schulreich, C.; Kostevic, A.; Tiersch, B.; Koetz, J.; Kakorin, S.; von Klitzing, R.; Jung, M.; Hellweg, T.; Wellert, S. Extraction of model contaminants from solid surfaces by environmentally compatible microemulsions. J. Colloid Interface Sci. 2016, 471, 118−126. (6) Boonme, P. Applications of microemulsions in cosmetics. J. Cosmet. Dermatol. 2007, 6, 223−228. (7) Kalaitzaki, A.; Emo, M.; Stébé, M. J.; Xenakis, A.; Papadimitriou, V. Biocompatible nanodispersions as delivery systems of food additives: A structural study. Food Res. Int. 2013, 54, 1448−1454. J


Article

Langmuir (32) Gompper, G.; Endo, H.; Mihailescu, M.; Allgaier, J.; Monkenbusch, M.; Richter, D.; Jakobs, B.; Sottmann, T.; Strey, R. Measuring bending rigidity and spatial renormalization in bicontinuous microemulsions. EPL 2001, 56, 683. (33) Gompper, G.; Kroll, D. Membranes with fluctuating topology: Monte Carlo simulations. Phys. Rev. Lett. 1998, 81, 2284. (34) Sinha, S. K.; Sirota, E. B.; Garoff, S.; Stanley, H. B. X-ray and neutron scattering from rough surfaces. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 38, 2297−2311. (35) Lee, D. D.; Chen, S. H.; Majkrzak, C. F.; Satija, S. K. Bulk and surface correlations in a microemulsion. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1995, 52, R29−R32. (36) Gompper, G.; Schick, M. Correlation between Structural and Interfacial Properties of Amphiphilic Systems. Phys. Rev. Lett. 1990, 65, 1116−1119. (37) Chernov, A. A.; Mikheev, L. V. Wetting of Solid Surfaces by a Structured Siple Liquid: Effect of Fluctuations. Phys. Rev. Lett. 1988, 60, 2488−2491. (38) Charitat, T.; Lecuyer, S.; Fragneto, G. Fluctuations and destabilization of single phospholipid bilayers. Biointerphases 2008, 3, FB3−FB15. (39) Thomas, L. L.; Christakis, T. J.; Jorgensen, W. L. Conformation of Alkanes in the Gas Phase and Pure Liquids. J. Phys. Chem. B 2006, 110, 21198−21204. (40) Pieruschka, P.; Safran, S. Random interface model of sponge phases. EPL 1995, 31, 207. (41) Morse, D. C. Topological instabilities and phase behavior of fluid membranes. Phys. Rev. E: Stat. Phys., Plasmas, Fluids, Relat. Interdiscip. Top. 1994, 50, R2423. (42) Sottmann, T.; Strey, R.; Chen, S. 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. (43) Schulreich, C.; Angermann, C.; Höhn, S.; Neubauer, R.; Seibt, S.; Stehle, R.; Lapp, A.; Richardt, A.; Diekmann, A.; Hellweg, T. Bicontinuous microemulsions with extremely high temperature stability based on skin friendly oil and sugar surfactant. Colloids Surf., A 2013, 418, 39−46. (44) Stehle, R.; Schulreich, C.; Wellert, S.; Gäb, J.; Blum, M.-M.; Kehe, K.; Richardt, A.; Lapp, A.; Hellweg, T. An enzyme containing microemulsion based on skin friendly oil and surfactant as decontamination medium for organo phosphates: Phase behavior, structure, and enzyme activity. J. Colloid Interface Sci. 2014, 413, 127− 132. (45) Arleth, L.; Marcelja, S.; Zemb, T. Gaussian random fields with two level-cutsModel for asymmetric microemulsions with nonzero spontaneous curvature? J. Chem. Phys. 2001, 115, 3923−3936. (46) Kaur, G.; Chiappisi, L.; Prévost, S.; Schweins, R.; Gradzielski, M.; Mehta, S. K. Probing the microstructure of nonionic microemulsions with ethyl oleate by viscosity, ROESY, DLS, SANS, and cyclic voltammetry. Langmuir 2012, 28, 10640−10652. (47) Gompper, G.; Zschocke, S. Ginzburg-Landau theory of oil-watersurfactant mixtures. Phys. Rev. A: At., Mol., Opt. Phys. 1992, 46, 4836− 4851. (48) Gee, M. L.; Israelachvili, J. N. Interactions of surfactant monolayers across hydrocarbon liquids. J. Chem. Soc., Faraday Trans. 1990, 86, 4049−4058. (49) Uzum, C.; Makuska, R.; von Klitzing, R. Effect of Molecular Architecture on the Polyelectrolyte Structuring under Confinement. Macromolecules 2012, 45, 3168−3176. (50) Zeng, Y.; von Klitzing, R. Scaling of layer spacing of charged particles under slit-pore confinement: an effect of concentration or of effective particle diameter? J. Phys.: Condens. Matter 2012, 24, 464125.

K


Sugar Surfactant Based Microemulsions at Solid ... - ACS Publications

Recommend Documents