Measuring Contact Lens Friction

With multiple measurement variables and testing methods, is standardization possible?


Measuring Contact Lens Friction

With multiple measurement variables and testing methods, is standardization possible?

Lakshman N. Subbaraman, PhD, BSOptom, MSc, FAAO; John Pruitt, PhD; & Lyndon Jones, PhD, FCOptom, FAAO

Contact lens-related discomfort (CLD) continues to be a major problem associated with contact lens wear, with up to 50% of patients reporting CLD (Dumbleton et al, 2013). The majority of these dissatisfied contact lens wearers report that discomfort and dryness, particularly at the end of the day, are their major reasons for discontinuation of contact lens wear (Fonn, 2007).

CLD is influenced by many factors, including those related to the patients and to contact lens material (Nichols JJ et al, 2013). When a contact lens is placed on eye, there are several interactions in effect—this includes the interaction of the posterior surface of the contact lens with the corneal surface and the anterior surface of the contact lens with the posterior surface of the eyelid during the blink. Recent studies suggest that there is a strong relationship between the coefficient of friction (CoF) measurements of hydrogel contact lenses and subjective comfort (Brennan, 2009; Coles and Brennan, 2012; Kern et al, 2013; Jones et al, 2013).

Measuring the friction properties of soft lenses has gained a significant amount of attention over the past few years, and researchers have employed a number of differing methods. The most common method is through the use of an instrument called a “microtribometer” (Rennie et al, 2005; Roba et al, 2011). Some have used a method based on atomic force microscopy (Kim et al, 2001; Kim et al, 2002), while others have developed “in-house” techniques such as that based on an inclined plane method (Tucker et al, 2012).

Qualitatively, lubricity ratings have also been determined by manually rubbing contact lenses between the fingers (Tucker et al, 2012); however, these qualitative measurements can only differentiate relatively large differences in lubricity between contact lenses. The majority of these methods are described in our previous review paper (Subbaraman and Jones, 2013).

Determining the friction properties of soft contact lenses is not trivial, and various papers have reported widely differing results for identical contact lens materials (Subbaraman and Jones, 2013; Sawyer et al, 2012). This review will describe the factors that can influence the CoF values determined and will attempt to explain why such differing data exists.

CoF is not an intrinsic material property (such as modulus or Dk), but is a “system property” that depends upon many factors of the measurement system, including the counter-surface substrate, contact mode, pressure, movement speed, and lubricating fluid, among other factors. All of these terms have been described in detail in a previous publication (Subbaraman and Jones, 2013).

Issues Related to Friction Measurements

The ocular surface is substantially “softer” compared to the hard surfaces that are typically employed by the microtribometers to determine the CoF values of contact lens materials. Most CoF measurements are conducted using a hard substrate (typically glass or stainless steel). To carry out friction measurements, the substrate (a plate or probe of some sort) and test material (contact lens) must be moved relative to each other. To achieve this, either the contact lens can be held in position and the probe moved across the surface of the lens, or the plate or probe remains in place and the lens is moved across it. These methods are fundamentally different, as in one method only a fixed point on the lens is being examined, whereas in the other method differing sections of the contact lens are being sampled, and thus differing results may occur.

In addition, many other variables can impact the results obtained, and a review of the literature clearly demonstrates that CoF values are highly dependent upon the entire measurement system (Jones et al, 2013; Subbaraman and Jones, 2013). Thus, it is only possible to compare CoF values within a single measurement method, and we should be extremely cautious in comparing the CoF values of various contact lens types that are measured using different testing methods. Furthermore, when evaluating any reported CoF values, we must consider the limitations inherent to each method. For instance, if the speed range is limited compared to the blink speeds (in the case of lid/lens friction) or to the relative contact lens/cornea speeds, then the relevance of the results must take such a limitation into account.

Factors Impacting CoF Measurements

This section provides an overview of various factors that play a role in determining the measured CoF values and also highlights the parameters that can potentially impact the measured CoF values. A few suggestions have also been made that will help to improve some of the existing methods, which will ideally help in moving toward friction measurements that better mimic what happens on eye. None of the existing methods can determine the actual frictional forces that occur between a contact lens and the surrounding ocular surface in a living eye, but they can provide valuable insights into differences in contact lens materials. It will remain important that any CoF measurement technique be ultimately demonstrated to correlate with actual clinical outcomes, such as comfort.

Substrate Choice The choice of the substrate against which the contact lens is moved applies to all methods of measuring friction. As described above, all friction measurements involve the relative movement of two surfaces, and the substrate employed will markedly impact the reported results. Substrates utilized in previous reports include glass (Rennie et al, 2005; Sawyer et al, 2012), stainless steel (Zhou et al, 2011), polyethylene terephthalate (Ross et al, 2005), or mucin-coated silanized glass (Roba et al, 2011).

Plastics or other hard substrates can also be utilized, but the disadvantage of using a substrate other than glass or steel is that other substrates may not be as smooth or as reproducible. Moreover, a substrate that has an increased roughness can have a large effect upon the measured friction. Substrate adsorption of lubricating fluid components and specific chemical interactions between the substrate and the test contact lens must also be considered. Another major disadvantage of all currently used substrates is that they are much harder compared to the ocular tissues. In-vitro corneal friction measurements using harvested corneal tissue have been reported (Wilson et al, 2013); however, the actual in-vivo friction of the cornea against eyelid tissue remains unknown.

Ideally, the substrate that is used to measure CoF should mimic the chemistry, surface roughness, and softness of the ocular tissues. Further, it should be available in a highly reproducible form that can be used to replicate the results obtained in various testing laboratories. However, there are currently no commercially available substrate options that meet all of these criteria.

In a recent publication (Dunn et al, 2014), “twin” or “Gemini” surfaces were developed, and the results from that study confirm that friction values obtained when two soft, porous hydrogel surfaces are moved relative to each other can be very different compared to friction values obtained by measuring a very soft, porous hydrogel surface against a hard, impermeable counter-surface. It is significant to note that in the real-eye situation, the glycocalyx of the eyelid rubs against the corneal glycocalyx, and the use of novel “Gemini” surfaces is a major advancement in this research arena in terms of simulating what truly happens in the eye.

Dunn et al also demonstrated that the use of these Gemini surfaces results in friction measurements that are essentially independent of both the movement speed and the contact duration over a wide range of speeds and contact times. It appears that a soft, permeable counter-surface may better predict the frictional forces occurring on eye. However, further work is needed to develop a reproducible substrate that uses a soft counter-surface with properties similar to those of the glycocalyx.

Type of the Substrate Counter-Surface and Movement Most of the current friction testing processes involve the application of a constant, single point of contact on the contact lens surface in which the lens is moved relative to the substrate. This “static” contact mode using a hard substrate is very different from the in-vivo condition.

The drawback of using this kind of continuous point of contact method is that it can result in deformation of the contact lens surface, and the substrate can actually sink into the lens material. This is especially the case with contact lens materials that possess an extremely soft surface, but it is less of a risk for lenses with a surface modulus that is much higher than the test pressure utilized in the measurement.

Moreover, in dynamic contact friction testing, a moving contact point results in new sites of contact and compression at the lens surface. We see a similar scenario on the ocular surface, wherein the lid wiper comes in contact with the lens surface and creates a migrating point of compression on the contact lens surface. For very soft surfaces, this may result in a phenomenon called “weeping lubrication,” as liquid is temporarily squeezed out of the soft lens surface (or the ocular glycocalyx, in the case of the eye). This weeping lubrication mechanism would be lost if a constant point of contact is maintained between the contact lens and the substrate.

Therefore, future friction testing to predict lid-lens interactions on very soft materials should consider the use of a “dynamic contact” that moves across the contact lens surface to avoid a single continuous pressure point on the lens (Dunn et al, 2014). If a static contact mode is utilized, then care should be taken to consider both the surface modulus of the material being tested and the persistence time of the static contact mode. As previously noted, the use of a soft, porous counter-surface appears to minimize the issues observed with “static” versus “migrating” contact modes.

Normal Force Pressure The contact pressure used in a microtribometer tries to mimic the force applied by the lid wiper on the ocular surface. In-vivo eyelid pressures are typically estimated at 1kPa to 7kPa. The average pressure of a contact lens against the cornea will be much lower than 1kPa due to the much larger surface area between the contact lens and the cornea compared to that between the lid wiper and the contact lens.

Achieving accurate force measurements at these low pressures can be challenging. A recent study has determined frictional values at very low contact pressures (approaching 1kPa) and showed that at the pressures such as those seen in the eye (3kPa to 5kPa), the newly developed gradient gel surfaces of 2 to 3 microns thick can support smooth sliding and provide a lubricous surface (Sawyer et al, 2012).

Contact pressure is a very important parameter when testing lens materials that have a very soft surface. This is due to the fact that if the surface modulus is of a similar order of magnitude as the applied test pressure, the lens surface may be altered, and this could result in high friction measurements not representative of the on-eye situation, especially in the case of a static contact mode.

In-vitro friction measurements of contact lenses are typically designed to mimic the pressures at the lens/eyelid interface instead of the lens/corneal interface due to the difficulty in replicating the extremely low pressures and dynamics of the contact lens/corneal interface. The contact lens moves against the cornea with much lower velocity compared to the blink speed and, for most of these corneal movements, the average contact pressure across the contact lens is extremely low.

During a blink, as the lid wiper moves across the front surface of a contact lens, the force of the lid wiper is transferred to the lens/corneal interface. However, these localized, intermittently higher forces and the resulting “pressure wave” that they induce across the lens/corneal interface are extremely difficult to replicate in a tribometer.

Sliding Speed This parameter is representative of ocular movements including the speed at which the eyelid glides over the contact lens surface. Previous studies have used a wide range of sliding speeds—from 63 to 6,280µm/sec (Rennie et al, 2005), 0.01 to 0.5cm/sec (Zhou et al, 2011), or 10 to 600µm/sec (Sawyer et al, 2012). Sliding speeds as low as 0.1mm/sec (Roba et al, 2011) have been used to measure “boundary lubrication” (where there is a solid-to-solid contact of the sliding surfaces), which occurs when the lens is in direct contact with the ocular tissues.

To maximize differentiation among materials, it is generally preferable to test at low speeds (the CoF is usually highest at low speeds) to ensure that boundary lubrication exists; at higher speeds, “hydrodynamic lubrication” may occur (where a thin layer of fluid is present between two sliding surfaces), and the individual surface properties become much less important in determining the CoF. A wide range of movement speeds occurs in vivo; testing at multiple speeds would be required to more fully characterize materials, and this would be the preferred method of characterization.

Contact Lens Sample Preparation Many of the currently available commercial contact lenses are packaged in their blister packs, with added surfactants or surface active polymers (Menzies et al, 2010). This packaging can have an impact on the measured CoF value.

Although the CoF value in the presence of packaging solution may be representative of friction upon lens application, it is extremely likely that these additives will disappear with time. Therefore, it may be important to remove the packaging solution from the contact lenses by soaking the lenses in a neutral solution, such as a phosphate buffered saline (PBS) solution, prior to testing the CoF measurements, especially if the goal is to measure the lens material without the influence of any additives in the packing solution.

The ideal method would be to determine CoF values on eye. Alternatively, it may be worthwhile to test the CoF measurements of patient-worn contact lenses to determine any change in surface properties after the lens material is soiled with a variety of proteins, lipids, and mucins found in the tear film.

Interfacial or Lubricating Fluid CoF is an interfacial property, and the measured CoF values can change dramatically depending on the interfacial fluid between the substrate and the contact lens being tested. For example, it has been observed that higher CoF values can be reported when measuring nelfilcon/polyvinyl alcohol (PVA) lenses in a borate buffer due to the cross-linking of borate-PVA.

Historically, a complex artificial tear solution and/or solutions containing a variety of tear proteins that mimic human tears have been used as lubricating fluids (Roba et al, 2011; Ngai et al, 2005). However, there is no standard for the composition of the artificial tear solution, and it may not be possible to develop such a standard. The difficulty lies in the layered arrangement of the tear film, including the fact that mucins found on the ocular surface are likely “gelled” and directly interact with the glycocalyx. This phenomenon will be extremely difficult to replicate using an in-vitro model.

The composition of the artificial tear solution gains significant importance when using hard surfaces, since dissolved proteins, lipids, mucins, and other components may provide some boundary lubrication. In a previous study by Nairn and Jiang (1995), various commercial ophthalmic solutions were used. The study showed that by varying the lubricants, the reported CoF values changed significantly.


In summary, the reported CoF values can vary dramatically depending upon the technique employed. In addition, the measured values can change significantly depending upon the parameters employed during the testing process, including the substrate, contact time, contact pressure, sliding speed, and the interfacial fluid. Therefore, we should be very cautious in the interpretation of results that are presented in the literature, and we should closely look into the conditions under which the friction measurements were performed.

While it is not currently possible to develop a standardized measurement method that captures all of the relevant test parameters, researchers have made significant strides in trying to develop models that mimic the on-eye condition. It will remain important that published CoF results fully disclose all details of the test methodology utilized, and that individual test methodologies are demonstrated to relate to actual clinical comfort outcomes. CLS

Acknowledgements: The authors would like to acknowledge Alcon Laboratories Ltd, USA for its support in developing this article through an unrestricted grant. The authors would also like to thank Dr. Gregory Sawyer and Dr. Thomas Angelini from the University of Florida for reviewing this manuscript.

For references, please visit and click on document #242.

Dr. Subbaraman is the head of Biological Sciences and a senior clinical scientist at the Centre for Contact Lens Research (CCLR), School of Optometry and Vision Science, University of Waterloo, Canada. He serves as a member of the American Academy of Optometry’s Research Committee. He is a member of the International Society for Contact Lens Research, Tear Film & Ocular Surface Society, and ARVO. He is a consultant or advisor to Johnson & Johnson Vision Care Inc. (JJVCI) and has received research funding or honoraria from Alcon and JJVCI.

Dr. Pruitt is currently a project head at Alcon Vision Care R&D. He is a chemical engineer with a background in contact lens surface chemistry.

Dr. Jones is a professor, University Research chair, and director of the CCLR at the University of Waterloo. He is a consultant or advisor to Alcon and JJVCI and has received research funding or honoraria from Advanced Vision Research, Alcon, AlgiPharma, Allergan, CooperVision, Essilor, JJVCI, Ocular Dynamics LLC, Oculus, Ocusense, TearScience, and Visioneering Technologies.