The power of the NR to deal with complex interfacial structures has so widened the range of application that it continues to reveal phenomena whose existence has not been suspected. Furthermore, this is happening against a background of strong expansion of the study of liquid surfaces. Not only is there considerable progress still to be made in the examples identified in the previous section but significant qualitative jumps in our understanding of the physicochemical properties of liquid interfaces can be expected as NR is able to probe smaller samples and shorter timescales.
The new reflectometer SHOULD:
(a) extend the range of accessible experiments by an improvement in sensitivity over existing instruments
(b) create much needed extra beam-time for reflectometry from liquid surfaces.
The ILL reactor should make it possible to build a reflectometer with higher incident flux than existing instruments and therefore open up the possibilities of doing new science. Below is a list of research areas that will be covered by the new reflectometer if flux levels are similar to those of D17, and therefore higher than what available at the moment of writing elsewhere in the world:
(i) ADSORPTION AT FREE LIQUID INTERFACES OF COMPLEX SYSTEMS FROM “SOFT-MATTER” (I.E. SURFACTANTS, POLYMERS, ETC.) AND “BIOLOGY” (LIPIDS, PROTEINS, DNA, ETC.) AT Å RESOLUTION : As the instrument flux and hence the sensitivity is increased, a greater extent of complexity can be explored. In this category belong all systems where one component is anchored at the surface, e.g. as a spread monolayer, and the second is adsorbed from solution (or vapour). The general issues of how or why a solute penetrates an insoluble monolayer are almost totally unexplored. The mode of interaction of biopolyelectrolytes (e.g. actin, spectrin, biotin) with phospholipids, either as supported bilayers or spread monolayers, is not well understood at the molecular level. Similarly, our knowledge of the interaction and conformation of small peptides and membrane proteins with phospholipid layers (monolayers or bilayers) is rudimentary. Although the structure of many enzymes that operate at the membrane surface is known with atomic resolution and even their mode of operation at the level of the catalytic site is understood (e.g. phospholipases), we do not have a clear picture of the infrastructure of their mode of action in the assembly (how or why they attach to the membrane surface, how their substrates or inhibitors are transported to them, what happens to the products, etc.). An understanding of this infrastructure is a vital part of understanding their biological function, which, for example, is still incompletely understood for the ubiquitous phospholipases. Even though a phospholipid monolayer may be somewhat different from an entire membrane, it is sufficiently similar that NR should be able to make really important contributions in this area. Similarly, although gene therapy promises exciting opportunities to tackle disease, the problem of the delivery of the gene is not yet solved. Recent focus has been on complexes involving plasmid DNA and cationic liposomes but studies on these by a variety of techniques have not yet led to any significant new knowledge. NR is able to probe interactions of this type (at a flat surface) at a higher resolution (in space and composition) than almost any other technique. The behaviour of polymer/surfactant mixtures at the liquid surface, which tends to create insoluble monolayer complexes and which is of immense industrial importance (nearly all commercial formulations of surfactants also include polymers which are thought to act synergistically with the surfactant at interfaces), is not understood even qualitatively. Similar arguments apply to protein/surfactant interactions, which are important to understand if only because of the widely used method of using surfactants to purify and separate proteins. This has already been explored at a preliminary level by NR but there is clearly an enormous amount of further work required to gain further insights. The build up of composite layers from either mixtures of oppositely charged polyelectrolytes or from mixtures of oppositely charged polyelectrolytes and nanoparticles, all important in industrial coatings and nano-composites, is ideally followed using NR, especially when off-specular reflectometry is a realistic option.
High resolution is attained by measuring reflectivity in a wide the qz range combined to the use of contrast variation technique. Maximum qz measurable in practice depends on the minimum reflectivity which can be measured in a practical time-scale. This implies maximum flux on the sample with a minimum background. Background from sources other than the sample itself should be <10^-7 of the main beam maximum flux.
To fulfill the requirements for performing the science cited above the instrument must have:
- VERTICAL REFLECTION PLANE
- MAXIMUM Q-RANGE
- MINIMUM BACKGROUND (Reflectivity<10-7)
(ii) KINETICS (IN THE FRACTION OF MINUTE TIMESCALE): The reduction in measurement time of between a half and a fifth compared with existing instruments will extend the range of kinetic experiments to hitherto inaccessible systems. The need for kinetic studies on conformational changes in the supramolecular complexes associated with complex formation in templating has been identified. Other accessible kinetics where deuterium labelling creates exciting opportunities are the actions of phospholipases on phospholipid monolayers. Here it is interesting to note that XR has been used for this type of study, but contrast variation gives a decisive advantage to NR. Thus use of D2O as solvent allows NR to be effective for systems closer in composition to those encountered in real biological systems and perdeuteration of, for example, the acid leaving the phospholipid opens up the exciting possibility of following the hydrolysis process in close and immediate detail. For all biological systems the timescales associated with transport and removal of materials to and from membranes, not just substrates and products, but associated proteins, fall in a range that could be well explored by the faster data acquisition (<1min). The formation of interfacial films using initial physical accumulation at the interface followed by in situ chemical processing, e.g. in situ polymerisation, is another area where the new machine would open up a range of interesting experiments.
To fulfill the requirements for performing the science cited above the instrument must have:
- ABILITY TO MEASURE A RANGE OF Qz SIMULTANEOUSLY
- MAXIMUM FLUX
(iii) REDUCED SAMPLE SIZE AND INCREASED SENSITIVITY; LIQUID/LIQUID INTERFACES: The combination of the additional incident flux and an intrinsic reduction in illuminated sample length (higher reflection angle) offers the possibility of a substantial reduction in sample size. This has the obvious benefit of giving NR access to small regions of surfaces inhomogeneous in either space or time. No experiments on inhomogeneous surfaces have yet been performed, although there are a number of such phenomena of interest in spread monolayers. The greatest gain from the smaller sample size will be in the study of the highly important but still somewhat inaccessible liquid/liquid interface. To do experiments at this interface the neutrons must penetrate one of the liquids. All liquids scatter neutrons too strongly for NR to be at all generally applicable unless the pathlength of transmission can be reduced substantially. X-ray reflectivity is also applied to structural studies at liquid-liquid interfaces and in fact offers an advantage in that the transmission of the X-rays through the sample is not a concern with modern-day laboratory and synchrotron sources. However, this transmission advantage is offset by the lack of intrinsic refractive index contrast at the interface in many hydrogen-containing systems; a feature of the neutron reflection technique is the opportunity to use isotopic substitution to tune the refractive index contrast. The proposed instrument should introduce gains in the physical stability of the interface and gains in the accessible range of composition that should give NR the same lead in studies of the liquid/liquid interface that it already enjoys for the air/liquid interface. Examples of studies at liquid/liquid interfaces include the preparation of ultra-thin polymer films at a liquid interface by having monomers in one liquid and oxidising agents in the other liquid or gas phase (or monomers in the gas phase and oxidant in the liquid). Since the two liquids are immiscible and the polymer does not dissolve in either, this is a way of making very thin coatings that can then be extracted and laid on surfaces. Another system involves looking at the localisation of species at the solution interface when there is heterogeneous electron transfer or some other chemical processes going on. For these and similar examples, it is important that the incoming beam can approach the interface from above or below the horizon as one liquid phase may be far more easily penetratable than the other. In addition if both measurements can be done this can eliminate the need for an extra contrast. Reduced sample size is also a possible key to doing useful experiments on lubrication. Although the actual interfacial contact will always be too small for exploration by NR, the mode of action and the fate of lubricant additives at a surface in use, where the effects would be averaged over a wider area, would be accessible and this is something about which little is known at present.
To fulfill the requirements for performing the science cited above the instrument must have:
- MAXIMUM FLUX OVER SMALLEST WIDTH BEAM
- REFLECTION FROM BOTH ABOVE AND BELOW THE HORIZONTAL INTERFACE
(iv) LATERAL ORDER AND SURFACE INDUCED STRUCTURE (RESOLVING LATERAL SIZES BETWEEN 50 AND 500 Å): There have been so far relatively few studies of organized structures at interfaces, e.g. either adsorption from structured phases, bicontinuous, lamellar, rod, vesicular phases of aqueous solutions of amphiphiles, or even micelles or microemulsions, or the incipient formation of these phases at an interface. The existence of incipient phases has always been suspected and, although non-contact AFM has proved valuable for identifying a number of related structures at the solid/liquid interface, and the surface balance is sometimes able to identify them at the mica/water interface, the complementary and more detailed information available from NR on the simpler air/water interface will be able to map out the general phenomena occurring in a much more comprehensive way. Preliminary studies indicate that this will be an area where off-specular scattering may provide valuable information on the lateral correlations. A particularly exciting area is in the formation of templated structures (aluminosilicate or silicate structures forming on templates of surfactant aggregates at the air/water interface) where reflection, off-specular reflection and the ability to follow the kinetic process of growth will all play important roles. The gain here is that currently the time required to gather off-specular data is too long for it to compete for beam time with the high demand for specular reflection. Also many of the systems of interest change on a time scale that is suitable for gathering specular reflection data, but is too slow for off-specular reflection. Biology offers a great number of examples where the study of lateral structure features would help in the un derstanding of how cell membranes work. These range from the study of domains in membranes (“rafts” which sizes range from 50 to 200Å) and their specific interactions with proteins; to the study of pores formed by peptides and proteins; to clustering of proteins at cell surfaces; etc.
To fulfill the requirements for performing the science cited above the instrument must have:
- MAXIMUM FLUX
- DETECT SIMULTANEOUSLY THE SPECULAR AND THE SIGNAL AWAY FROM THE SPECULAR DIRECTION IN THE REFLECTION PLANE
- COLLIMATE BEAM IN HORIZONTAL PLANE (GISANS) AND RESOLVE THE SCATTERED BEAM ON THE DETECTOR OUT OF THE REFLECTION PLANE
- DETECTOR WHICH CAN MOVE ALONG THE BEAM AND VERTICALLY
INTRINSIC ROUGHNESS OF FREE LIQUID SURFACES: Liquid samples are very sensitive to external vibrations which can induce surface waves and roughness of the surface rapidly reducing reflectivity. In addition to this there will always be thermally induced capillary waves having an amplitude which can be a large fraction, f, of the thickness of the surface layer of interest (roughness induced by capillary waves is ~3Å). We can gain a large factor in flux if we have the ability to measure with a a fractional resolution in q of the order of f. However for thick samples this fraction will be much smaller so we also need high resolution [16,17].
This requires:
- FLEXIBLE RESOLUTION IN Q (1%<DQ/Q<10%)
- MINIMUM VIBRATIONS AT SAMPLE AREA
SAMPLE SIZES:
Because of the limited neutron flux available, the size of samples needed in neutron reflectivity is of the order of several square centimeters. While this imposes some practical limitations, it has the advantage of allowing the inspection of macroscopic features of surfaces not possible with x-rays. Practical limitations need to be taken into account though. For solids sizes currently used range from 20 to 60 mm wide samples.
Liquid samples do not have any intrinsic limitation although restriction might be imposed by space available in the sample area. Normally either Langmuir throughs are used, whose surface is bound to be around 5 times bigger than the illuminated area for reasons of homogeneity, or a series of simpler reservoirs put on a translation table that allows computer controlled scans on several samples. In either case100mm wide samples would be an upper limit set by practical issues of sample handling.
For liquid-liquid samples overcoming the low transmission through a liquid remains the major obstacle. The most promising technique today for preparing liquid/liquid interfaces suitable for neutron studies consists of spin coating and freezing a very thin layer (~10microns) of liquid on a solid substrate and put this in contact with another immiscible liquid [18]. By warming up the sample a liquid/liquid interface is formed. The neutrons approach the liquid/liquid interface via the solid substrate and cross the thin film. Footprints of 140mm are being currently used from 60mm wide beams.
In conclusion, an upper limitation of the neutron beam width of about 100mm is dictated only by practical reasons related to sample quantity, sample handling and space available in the sample area. For these reasons it is certainly desirable to focus the neutron beam to maximum flux. Nevertheless, measurement of very thin layers and kinetics would benefit from the highest possible intensity over the whole sample: the bigger the illuminated area, the higher the total intensity, the faster the acquisition time.
USE OF IN-SITU COMPLEMENTARY TECHNIQUES : Brewster Angle Microscopy
The imaging of thin films at the gas/liquid interface is often used for studying in-plane features of such systems. It would be of great importance to be able to record images of the gas/liquid interface while collecting neutron data in order to check the quality of the layers and inspect in-plane features and their possible modifications during the acquisition (in kinteics scans, for example).
On a neutron reflectometer one of the most interesting techniques that could be applied for in-situ imaging of the layers is Brewster angle microscopy (BAM). This is a relatively new technique that allows direct observation of the films onto the water/air surface without modifying the layer (as it happens when techniques as fluorescence microscopy or AFM are used).
BAM is sensitive to the surface density and to the anisotropy of phase domains in monolayers, where the reflectivity of a planar interface between two media depends on the polarization of the incident light and on the angle of incidence. For a Fresnel interface (an interface where the refractive index changes steeply) and for a polarization where the electric field is in the plane of incidence, the reflectivity vanishes at the Brewster angle. For a real interface, the reflected light intensity has a minimum at the Brewster angle, but does not vanish. However, the reflected intensity at the Brewster angle is strongly dependent on the interfacial properties and is particularly sensitive to monolayers at the interface. The reflectivity of a real interface at the Brewster angle for the mentioned polarization has three origins: (a) the thickness of the interface, (b) the roughness of real interfaces, and (c) the anisotropy of monolayers.
The figure shows an example of Brewster angle microscope in a schematic way. The interface is illuminated at the Brewster incidence (~ 53°) with a polarized laser beam from a He-Ne laser. The reflected beam is received by a microscope. The beam is analyzed by a polarization analyzer and received by a CCD video camera to develop an image of the monolayer. It’s geometry is sch that it could be adapted easily in the reflecometer sample area.
Input by Bob Thomas and Adrian Rennie and useful discussion with Andrew Glidle, Ali Zarbakhsh and Jean Daillant are gratefully acknowledged.
Legal notice