AmericanChemicalSociety.com

In the second period of this project we have made strong progress towards completing the final main aim of our proposal (aim (c) of section 2.3 of our proposal), and this has been published both as a Report in Science, as well as being highlighted in several general publications, including a Research Highlight in Nature. In addition we have extended the work to examine several related issues which shed further light on the proposal goals. The publications resulting from these in the second period are detailed separately, and further publications are in preparation.

In the second year of the project we have measured the frictional forces between polymer brushes, with zwitterionic monomers, grown covalently from initiator groups attached to a mica surface, utilizing atom transfer radical polymerization (ATRP) methods. This grafted-from brush growth of poly(2-Methacryloyloxyethyl phosphorylcholine), or pMPC brushes, directly from a macroinitiator attached at the surface was described in our First Year report, and the frictional measurements described below constitute the major part of the third aim of the project. Each of the monomers on these chains has a phosphorylcholine (PC)-like structure, and is consequently highly hydrated. This is a crucial issue for their lubrication properties, and it is a hypothesis of the proposal that such hydration layers would strongly reduce frictional forces. We have now demonstrated that the lubrication afforded by such brushes is indeed remarkable.

Using a surface force balance (SFB), a uniquely sensitive device for measuring friction between molecularly smooth surfaces with a sub-nanometer resolution in surface separation, we measured directly the frictional forces between two pMPC brush-bearing surfaces over a range of pressures and sliding velocities. In particular, because of the robust nature of the brush attachment to the substrate arising from its grafted-from we were able attain very much larger pressures than previously accessible with the SFB without causing the brushes to shear-off the surfaces. In this way we managed to attain pressures up to ca. 7.5 MPa, or some 75 atmospheres, across the region of contact between the brushes. These values of the pressure are of particular interest since they are comparable with or higher than the pressure in the major human joints such as hips or knees. We found that even at these high pressures the lubrication afforded by the pMPC brushes resulted in sliding friction coefficients μ that had values μ ≈ 0.001 or lower. These values are of particular significance since they are similar to those between healthy human articular cartilage in synovial joints, and may point the way to the meachanism that is responsible for such low friction. We were especially interested in the molecular mechanism that resulted in such low friction, and particularly why this was attained at high pressures where charged brushes were known to fail as lubricants.

The origin of the low friction observed in our studies even at the highest pressures may be attributed to the following. Polymer brushes – whether charged, neutral or zwitterionic - are known to reduce friction at moderate loads by virtue of their brush-like structure. This is due to the entropic cost of the brush interpenetration, so that as two brushes press against each other the interface between them is only weakly interpenetrated and so remains unentangled when the brush-bearing surfaces slide past each other. At higher pressures this entropic mechanism becomes less important, and it appears that in the case of the pMPC brushes studied here a different effect takes over. This is related to the highly hydrated nature of the monomers on the chains. The PC groups are known from a variety of different studies  to be particularly well lubricated, having of the order of 15 – 20 or so water molecules in the primary hydration shell. These hydration layers then act as molecular ball-bearings and provide the final efficient lubrication at the highest pressures, in particular at pressures where the less well-hydrated charged brushes in earlier studies experienced much higher friction, which caused them to be removed by the sliding motion.

We also examined the dependence of the friction on sliding velocity over 3 or more orders of magnitude variation in the latter, and showed that it was only weakly velocity dependent. This weak variation is characteristic of solid-like friction (even though the friction coefficients are distinctly liquid-like). It may be due to a self-regulating effect whereby at the higher velocity the pMPC chains are pulled out of the interpenetration zone and so reduce the extent of interfacial dissipation at the same time as it is increasing by viscous effects at the higher velocities. The two opposing effects thus to some extent cancel each other, leading to the weak overall velocity dependence.

To summarize this friction work, we have shown for the first time that physiologically-low values of friction at the pressures pertaining in major mammalian joints may be achieved using macromolecular surface phases consisting of zwitterionic brushes. This provides a clue to better clinical treatment of widespread ailments such as osteoarthritis which result from the breakdown of lubrication in previously healthy mammalian joints. It may also, as emphasized in a recent Science Perspective, point the way to more efficient tissue engineering approaches as well as to prosthetic implants with low friction, low wear properties.

While the friction studies with pMPC brushes, published in Science, was the main thrust of the second year of research in this project, fulfilling much of its aims, auxilliary investigations supported by the grant were also instrumental in helping to understand the low friction mechanism. These include studies of the mechanism of polymer attachment to surfaces, additional studies with charged brushes, and a detailed study of the role of hydration layers in modulating shear forces, and were published (acknowledging PRF support) as detailed separately.