X-ray and neutron reflectivity studies of plasma polymer coatings

Andrew Nelson and Michael James (ANSTO), Keith McLean, Patrick Hartley and Ben Muir (CSIRO)

Surface functionalisation is a key step in the passivation of biocompatible devices and objects. In addition, production of well-defined biologically active surfaces is essential in the production of biosensors for medical and other diagnostic applications.

In recent years plasma polymerisation has been shown to be an ideal method for preparing thin, uniform, defect free polymer films with excellent thickness control using monomers not typically used in conventional polymer synthesis.

While it has been recognised that these polymer are highly cross-linked and that the film properties depend strongly on the monomer, the actual composition of films generated in this manner is generally uncertain. The polymerisation process generally destroys the structure of the monomer, leading to a jumble of surface active sites within the film.

Following removal of the polymer from the deposition chamber films may rapidly oxidise to incorporate stoichiometric quantities of oxygen within the polymer network.

The composition of the plasma film can be determined using x-ray Photoelectron Spectroscopy (XPS), however this technique is only able to provide information relating to heavier atoms such as C, N, O, F, S, P etc. The quantity of light atoms in the polymer (such as H) cannot be determined using XPS. Using a combination of x-ray and neutron reflectometry, one can determine the composition of such organic surfaces.

Where the composition of a surface is previously well known, reflectometry can be used to determine the mass density of the film. We demonstrate in the following study that by utilising XPS, x-ray and neutron reflectometry, one can uniquely decouple the density, composition and hydrogen content in these plasma polymer films.

Figure 1 shows the x-ray and neutron reflectivity data for an allylamine plasma polymer film deposited on a 100 mm diameter Si wafer. The excellent film quality is evident from the well-defined Kiessig fringes and low Rmin in the x-ray data. The relatively weak looking fringes in the neutron data are due to the poor neutron contrast between the film and underlying Si substrate.

A structural model was refined for each set of data, giving a film thickness of 278 ? and a surface roughness of 3 ?. The scattering length density (SLD) from the x-ray data was 1.33x10-5 ?-2, and from the neutron data 2.03x10-6 ?-2. XPS measurements gave the C:O:N composition as C162:O27:N11. Using this data and the two refined values for SLD we simultaneously solved for H content (C162O27N11H190) and mass density (1.47 gcm-3).

These results support the notion of a high-degree of cross-linking and unsaturation in the polymer, as the observed H content is well below that possible for aliphatic hydrocarbon chains.

Figure 1. x-ray and neutron reflectivity data for an allylamine plasma polymer thin film on Si. The solid line is the calculated reflectivity based on a refined structural model. The neutron data are offset by a factor of 100 for clarity.

The nature of these plasma polymer films in contact with an aqueous environment is a central aspect of their performance in relation to bio-functionalisation or bio-passivation. Our previous work revealed that allylamine plasma films in contact with water (D2O) showed both swelling as well as a substantial increase in scattering length density. On the surface, these results suggested a substantial amount of water penetrating into the film (~40%), however this is at odds with the known lack of porosity in these highly cross-linked polymers.

Another possibility for the dramatic increase in SLD may be that labile H atoms on the functional groups within the polymer were exchanging with D atoms in solution. The neutron scattering length of H (b = -3.739 fm) is different in magnitude and opposite in sign to that for D (b = 6.671 fm).

Further neutron reflectometry measurements were conducted with the allylamine plasma polymer film in contact with water of different SLD (i.e. different amounts of H2O mixed with D2O) ? see Figure 2 for a schematic of the solid-liquid cell. By using the above data collected versus air, along with neutron reflectivity collected at 3 different water contrasts: D2O (6.37x10-6 ?-2); 42% H2O:58% D2O (3.47x10-6 ?-2) and H2O (-0.56x10-6 ?-2) we were able to determine both the amount of water penetrating the film as well as the number of labile protons present.

Figure 3 shows the neutron reflectivity data collected at the different water contrasts. The allylamine film in contact with water was found to have a thickness of 293 ? (swollen by ~5%), contain only ~3% water and have 60 labile protons.

Figure 2. The cell used for solid-liquid neutron reflectometry experiments. The lower surface of the (100 mm diameter) Si wafer is coated with plasma polymer. Neutrons enter and leave the cell via the 10 mm thick Si wafer.

Figure 3. Neutron reflectivity data from the silicon-allylamine-water system. The lower curves are offset for clarity and the solid lines represent fits to the data.