S. Bare, A. R. Rennie (Kings College London), E. Bellet-Amalric, G. Fragneto (ILL).
Amphiphilic molecules such as surfactants are used in many applications that range from emulsion stabilisers, detergents, cosmetics, food and mineral processing. Surfactants at high concentrations in water often form lamellar phases with bilayers packed together to give a regular interlayer spacing (Fig. 1). In some circumstances other concentrated phases (cubic, hexagonal, etc.) may form. The properties of concentrated phases of surfactants can be of importance in many areas: such as the formulation of concentrated detergents, properties of foods and behaviour of cosmetic preparations. Flow of soft materials has attracted considerable interest and the rheological properties of surfactants are no exception.
|
|
| Figure 1: Schematic diagram of lamellar phase structure of surfactants showing the inter-lamellar spacing used to determine the orientation. |
For some years small-angle neutron scattering (SANS) has been used to produce detailed
pictures of structures under flow [1,2]. In the presence of shear field two different orientations of the lamellar
layers have been observed, respectively with the normal to the layers parallel to
the velocity gradient direction (orientation c) and perpendicular to it (orientation
a). Orientational transitions have been observed depending on the shear rate and
the system studied. In particular, Penfold et al. [1] observed a transition from orientation a at low shear rate to orientation
c at high shear rate for the C16EO6 (hexaethylene glycol monohexadecyl
ether)/water system. In a recent work Richtering at al. [2] observed the same transition for a system
of SDS/decanol/water.
Additionally, they observed two more orientational transitions at very low and intermediate
shear rates and the formation of vesicles. Both the transitions and the formation
of vesicles were influenced by the content of decanol. Also in block copolymer melts,
a transition from the a to the c alignment was observed, but there is no evidence
of vesicle formation. Recently, neutron diffraction has been used to determine orientational
order in dispersions of crystalline, anisotropic colloidal particles [3]. Measurement of the distribution
of intensity of a particular Bragg reflection directly determines the orientation
distribution of the plate-like particles. In many respects lamellar surfactant phases
resemble the structure of dispersions of plate-like particles. The distribution of
scattered intensity arising from the lamellar spacing can be used to determine the
orientation distribution. The flexibility and dynamic equilibrium of lamellae can
give rise to new and interesting phenomena.
The principal directions in a shear flow field are shown in Fig. 2 and a full description
of the alignment will require determination of the full three-dimensional structure,
or distribution of orientation, in relation to these axes. In order to achieve a
better understanding of the flow behaviour of lamellae under shear, a rotating disc
shear-cell has been built. This has been used on the diffractometer D16 to investigate
the orientational alignment, using a sample geometry that is not readily accessible
with a Couette shear cell and conventional SANS measurements. Data were collected
for the non-ionic surfactant tetraoxyethylene glycol monododecyl ether (C12EO4)
in D2O, which has a wide range of lamellar phases at room temperature.
The cell geometry allowed scans to determine the orientation distribution of the
lamellar phases in the plane of shear velocity and shear gradient.
With the cell normal to the incident beam the measured intensity corresponds to scattering
from the lamellae with normals almost in the flow direction. As 
increases to 90° this angle approaches the condition of lamellae normal to gradient
direction. Two concentrations corresponding to 40 and 60%wt surfactant were studied.
|
|
 |
|
| Figure 2: Shear-cell orientation on the diffractometer. The principal axes of the shear field can be aligned with respect to the scattering vector to allow full determination of the orientation distribution. It is useful to identify not just the axes of flow, shear gradient and vorticity but also to distinguish in measurements between the extensional and compressional quadrants associated with the shear flow. | Figure 3: Plots of the normalised intensity versus
the angle w for a) C12E4 at 40 wt% and b) 60 wt% at two different
shear rates (  0.5 s—1 and  30 s—1). For the 60% sample the maximum is
at w = 50 for both shear rates, while for the 40% sample the maximum is at
= 0 for 30 s—1 and at 10 < <
20 for 0.5 s—1. |
First results from these studies are shown in Fig. 3. Data were collected over
a range of angles and corrected for detector efficiency, sample thickness and transmission.
Spectra were acquired varying the angle w from —75° to +75° (see Fig. 2)
at different values of 
and 
(ranging between 0° and 24°
for and 0° and 20° for ). All experiments were repeated
at different shear rates. A strong Bragg reflection at Q = 0.1 Å—1
for the 60% sample and at Q = 0.07 Å—1 for the 40% sample was observed.
This corresponds to a d-spacing of 63 Å and 90 Å respectively corresponding
to the distance between adjacent lamellar planes. Two less intense peaks were observed
in the diffraction pattern of the 60% sample, at Q = 0.075 Å—1 (d
= 82 Å) and at Q = 0.2 Å—1. Peaks have been fitted with a
Gaussian and their intensities have been normalised making allowance for the variation
of the scattering volume, which depends on the cell orientation. The plot of the
intensity of the Bragg reflection at Q = 0.1 Å—1 versus the rotation
angle w is shown in Fig. 3b. A maximum is observed for
= 50°. A similar pattern is observed at higher shear rates. The experimental
arrangement allowed to investigate only small a range of
and ;
small variations are observed in the peak intensities for experiment recorded after
rotating the cell through and . For the less concentrated sample
the plot of the intensity of the Bragg reflection at Q = 0.07 Å—1
versus the rotation angle w showed a different pattern with a maximum corresponding
to 0 < < 20 depending on the shear rate (Fig.
3a). These preliminary data clearly show alignment that changes with concentration
and in directions that have not been reported previously. Suprisingly for the sample
at 60%wt fraction, the maximum is seen well away from either the flow or gradient
directions. The positive direction corresponds to the
extensional quadrant that lies between the flow and gradient directions (see Fig.
2). The normals to the lamellae are seen to lie in this quadrant.
Under the flow conditions studied, orientation is observed in a rather different
manner to that reported in previous work on concentrated surfactants. The lamellae
are oriented with the normal to the plane in the flow gradient-plane and forming
an angle with the vorticity-flow plane. The tilt depends both on the concentration
of the surfactant and the applied shear rate. The orientation of the particles is
only observable when scanning a large part of reciprocal space. The observed orientation
of the layers with the peak in the compressional direction of the flow field shows
certain similarities to the behaviour of plate-like colloidal particles
[4]. Further work is required to determine the exact orientation of the layers
as a function of surfactant concentration and shear rate.
[1] J. Penfold, E. Staples, A. Klan Lodhi and G. J. T. Tiddy, J. Phys. Chem B 101 (1997) 66. [2] J. Zipfel, J. Berghausen, P. Lindner and W. Richtering, J. Phys. Chem B 103 (1999) 2841. [3] S.M. Clarke, A.R. Rennie and P. Convert Europhysics Letters 35 (1996) 233. [4] A.B.D. Brown 'Order in Concentrated Colloidal Dispersions of Anisotropic Particles Under Shear' Ph.D. Thesis, University of Cambridge (1998). A.B.D. Brown and A.R. Rennie, submitted to Phys. Rev. E.
