Nanotribology Solutions Provided by Atomic Force Microscopy

The cost of everyday wear and tear caused by friction is high, costing almost one-fourth of the total energy consumption of the world for one year.1 Thus, enormous economic and environmental benefits are to be reaped by developing better methods to decrease frictional forces. Some markers along this pathway include the use of material films which are only micro- to nanometers thick, such as nanoparticles, ultrathin boundary lubricants and surfaces with nanotextures.

At this scale of length, however, tribology needs to be studied in much more depth and in greater detail. For instance, the ratio of surface to volume increases at this scale, which means a greater adhesion force, viscosity, and atomic effects, as a result of which the material may behave quite differently.

AFMs have become preferred tools for nanotribology (tribology carried out at meso to atomic scales of length) because of their very high force sensitivity as well as the excellent spatial resolution.2, 3 This article deals with the advantages of AFMs in this field as well as their capabilities. The emphasis is on how modern AFMs can be used for this purpose in biomedical devices and nano- or microelectromechanical systems (NEMS or MEMS).

Figure 1. Imaging compositional contrast with LFM The sample was an epitaxial film of LMSO (La0.7Sr0.3MnO3, ~18 nm thick) on an atomically flat, TiO2-terminated STO (SrTiO3) substrate. Images show topography (upper left), LFM friction (upper right), and current (lower left) obtained with conductive AFM, with sections corresponding to the dashed lines. Terraces with irregular edges and alternating high and low current are observed, suggesting alternating La1−xSrxO- and MnO2-terminated surfaces. This is confirmed by the LFM image, where low and high friction indicates differences in chemical termination from one terrace to the next. Scan size 3 μm; acquired on the MFP-3D AFM. Data courtesy of L. López, L. Balcells, B. Martínez, and C. Ocal, ICMAB-CSIC. Adapted from Ref. 5.

AFM Measurement of Friction

The primary mode of nanotribology experimentation is lateral force microscopy (LFM), which is also termed friction force microscopy (FFM). 4 Here the probe tip is put in contact with the sample to which a normal load is applied. The cantilever arm scans at right angles to its long axis, producing a lateral deflection signal which is monitored using a photodiode detector. The frictional force between the tip and the sample is shown by the hysteresis between the trace and retrace signals.

If LFM is used for raster scanning, the spatial variation in the sideways forces is imaged, with contrast being provided by the materials composing the surface, as Figure 1 shows. Another method is to acquire friction loops of the lateral force vs. sliding distance using a single line scan, as shown in Figure 2. The slope of frictional force vs. applied normal load gives the local coefficients of friction.

If the measurement is quantitative, the signal voltages from the LFM are first converted to absolute force values. This means the cantilever must be calibrated first for its lateral and flexural spring constants, as well as for its sensitivity to both horizontal and vertical deflection. A number of routines have been published for LFM calibration 6-8 but commercial AFMs do not usually have integrated calibration protocols.

Engineering Biomedical Interfaces

The tribological behavior is an important variable in biomedical device performance in a host of ways. One instance is the presence of corrosion in aqueous environments, which causes metal implants to wear away, or produces vascular graft breakdown as a result of friction caused by the flow of blood. Currently work is going on to produce novel surface treatments and coatings which can customize the tribological properties of biomedical device surfaces.

One promising area is soft polymers, such as hydrogels and brush structures. These can be tuned for as much adhesion and wettability as is optimal for a particular application. As a result they are widely considered for boundary-layer lubricants, targeted drug release, and so on.

As may be seen in Figure 2, LFM can be of great use in characterizing the micro- and nano-tribology of biomedical coatings, and the benefits of using colloidal probes to carry out tribological analysis of soft materials. Colloidal probes comprise cantilevers without a tip, where sample surface contact is made by a bead, usually of silica, which is attached to it. This allows for a larger area of contact and so reduces sample damage by decreased contact stress and instead permitting good definition of the sample-tip contact.

The effect of the surface chemistry with tribological properties may be examined using varying materials for the bead, or by changing the bead or tip properties for added functionality, through the use of coatings or chemical treatments with gold deposition or silanization.

Figure 3 depicts the greater understanding achieved by measuring other nanomechanical properties. Adhesion, especially, is closely correlated with friction and leads to significant changes in the area of contact between the tip and the sample. The modulus influences the amount of normal and lateral deformation. Using AFM, measurements of local adhesion and of modulus can be acquired in the form of force-curves, either one by one or in arrays. Even newer modes are now available for very rapid but gentle mapping of viscoelastic and elastic modulus.

Figure 3 also shows the influence of solvent, temperature and relative humidity, all environmental conditions, on tribology. This may be the solution to tuning the performance of the material in many applications, whether this is drug delivery or contact lenses. Thus it is important to perform experiments in conditions as close to real life as possible.

For this reason the ability to regulate the surroundings has been built into AFMs, such as sealed cells to regulate the gas or liquid conditions around the sample and the tip. There are also optional, specially made sample stages to allow accurate regulation of temperature in a stable manner, from below ambient temperatures to several hundred degrees.

Figure 2. Characterizing friction of layered polymer films  Two-layer films of polyacrylamide (PAAm) on silicon were created with either a brush supporting a crosslinked brush-hydrogel (“gel”) or a gel supporting a brush. LFM experiments were performed with a colloidal probe in water with different film configurations: pure brush (PAAm- 0), pure gel (PAAm-1), brush-gel (PAAm-0-1), and gel-brush (PAAm- 1-0). The top graph shows friction force versus applied load and the bottom graph contains friction loops, with curves color-coded to film type. Although the films displayed different tribological behavior, the response was always determined primarily by the structure of the film’s outer layer where the sliding occurred. Acquired on the MFP-3D AFM with the Closed Fluid Cell. Adapted from Ref. 9.

Reducing Friction in MEMS and NEMS

MEMS/NEMS has become big news in nanotechnology, and serves as a very useful illustration of the effects of a reducing length scale on the tribology of a material. As already stated, these machines are extremely promising if only the drastic surge in friction, stiction and adhesion of sliding surfaces at nanoscale are overcome.

This may entail the use of novel lubricants which are only nanometers thin, such as the 2D material called graphene. These are similar to their 3D counterparts in owing their reduced frictional force to poor coupling forces between the layers. An additional contribution is made by the superlubricity of surfaces whose lattices are not aligned properly or are not proportional to each other. This property refers to vanishingly small friction and stiction. Such phenomena can be investigated at nanoscale in 2D materials with a few or only one layer, as Figure 4 shows.

Being atomically flat, 2D materials also allow even higher spatial resolution using LFM, to atomic level. These studies have been performed almost from the inception of AFM12 and the resulting findings are quite often different from those obtained at macroscale studies.

One instance of note is the stick slip phenomenon in which a sawtooth pattern is obtained when scanning is done across a crystal lattice, due to lateral forces. This is seen in Figure 5. This helps to understand the basic processes going on during energy storage and release in such machines, while depicting lattice-scale images at unparalleled clarity. The actual area of contact between the tip and the sample is larger, at about 100 nm2 on average.

Figure 3. Evaluating adhesion in thermoresponsive polymers Layers of end-grafted poly(di(ethylene glycol)methyl ether methacrylate) (PDEGMA) homopolymer brushes were prepared with different dry thickness d. (left) Examples of force-displacement curves acquired with a colloidal probe in water at different temperatures (d = 27 nm), indicating the adhesion force Fadh and work of adhesion wadh. (middle) Fadh increased linearly with d and monotonically with temperature. (right) wadh showed a sigmoidal temperature dependence, as did elastic modulus (not shown). The results indicate a thickness-dependent characteristic temperature at which layer properties change significantly. This tunable, thermally-induced switching could be used for nonfouling coatings and other biomedical applications. Acquired on the MFP-3D AFM with the BioHeater sample stage. Adapted from Ref. 10.

Figure 4. Demonstrating microscale superlubricity Coefficients of friction were measured for microspheres of bare silica (SiO2) and multilayer-graphene-coated SiO2 (MLG) sliding on different surfaces (see inset in graph on right): (left) SiO2 and CVD-grown transferred graphene (Gsub), (middle) highly oriented pyrolytic graphite (HOPG), and (right) MLG sphere on hexagonal boron nitride (h-BN). Note the large differences in magnitude between different sphere-surface combinations. The ultra-low coefficients for MLG on HOPG and h-BN indicate superlubricious behavior, suggesting graphene coating as a strategy for reducing friction in MEMS/NEMS and other shaped parts. Acquired on the MFP-3D AFM. Adapted from Ref. 11.

When 2D materials are used for lubrication of NEMS/MEMS, it is necessary to gain insight into the relationship between the sliding velocity, humidity and other conditions operating in the experiment, as illustrated by Figure 6. At present, AFMs offer fast scanning which allows LFM readings to be acquired at almost any required rate, from 1 nm/s to 100 µ/s. The newest video-rate AFMs allow for scanning at higher speeds like 10-100 times faster, which brings it near the actual operating speed of an MEMS/NEMS (about 10-1000 cm/s) and to the hydrodynamic regime. Moreover, as already noted, the ability to regulate the environment allows the tests to be done in realistic operating conditions, as shown later.

Figure 5. Investigating stick slip of 2D materials Lateral forces on a molybdenum disulfide (MoS2) nanosheet were measured with two tips: (left) Si-OH tip created by treating a silicon tip with oxygen plasma for 1 min and (right) Si-OH tip coated with a self-assembled monolayer (SAM) of FDTS (1H,1H,2H,2H-perfluorodecyltrichlorosilane). Results are shown at (top) ambient humidity (30-40% relative humidity, RH) and (bottom) 73% RH. Adding the hydrophobic SAM decreases the already low forces observed with the bare tip and means the forces do not increase with humidity. The stick-slip behavior was understood by modeling the competing contributions of load and adhesion in each system. Acquired on the MFP-3D AFM. Adapted from Ref. 13.

Figure 6. Exploring velocity and humidity effects LFM experiments were performed with a silicon tip on a single layer of tungsten disulfide (WS2, thickness 5 nm) as the scan velocity n and relative humidity RH were varied. (left) The friction force increased logarithmically with n (to ~3.1 μm/s at 45% RH) and then became approximately constant. The behavior can be explained with a model that includes energy dissipation via stick slip at high velocities. (right) Both friction coefficient and adhesion force increased linearly with RH, consistent with the presence of an adsorbed water monolayer whose thickness increases with RH. Acquired on the Cypher AFM. Adapted from Ref. 14.

When considered in the light of the instrumentation required, AFM experiments need absolutely accurate and repeatable positioning in the XY axis. The newer AFMs boast closed-loop controls which have greater precision than the older open-loop controls. These decouple the actuation in the X, Y and Z axes and reduce the occurrence of artifacts due to measurement errors. The Z noise floor also affects the measurement of friction, and is dependent upon the mechanical vibration and electronic noise. Keeping the noise floors down to 5-10 pm in many laboratories has made it possible for advanced AFMs to achieve the resolution of applied force differences in the ultra-small piconewton to micronewton range.


This article describes only two of many applications that could change a lot as our understanding of nanoscale frictional forces and of wear is improved. This can happen using the advanced tribology tools presented in the form of current AFMs, which have exquisite sensitivity to force, excellent spatial resolution, and rapid scanning. Versatile controls for the environment, added to easier setup and other sophisticated features add to the punch.

Nanotribology using Asylum AFMs

  • The Cypher ES AFM is a superb device for investigating nanotribology in multiple environments. It comes with a sealed cell for liquid operation, and offers relative humidity adjustments, along with a heater to set the ambient temperature to 250 oC and a CoolerHeater for temperatures between 0 oC to 120 oC, helping to achieve stable and accurate operating temperatures.
  • The Cypher VRS AFM allows tip velocities of 1-10 mm/s, at line rates up to 625 Hz, in gas as well as liquid.
  • The Cypher AFMs have optional Interferometric Displacement Sensors to quantitatively assess cantilever movement, so that the torsional and flexural spring constants of the cantilever can be accurately calibrated compared to conventional beam displacement techniques as discussed in Reference 8.
  • Asylum AFMs have closed-loop scanners based on flexures to make sure their movement is reliable, independent and repeatable, as well as ensuring accurate control of sliding velocity. This leads to the absence of image distortions and produces very precise offsets, as well as focusing on specific areas for scanning. All Cypher and MFP-3D Infinity TM AFMs have the most advanced position sensors and boast unusually low noise down to 35 pm in the Z and 60 pm in the X and Y, or lateral axes.
  • GetRealTM Software in Asylum AFMS makes it easy to obtain absolute force measurements, performs automatic calibration of the flexural cantilever spring constant, as well as of the sensitivity of vertical deflection, at one click, without having to touch the sample.
  • The Fast Force Mapping Mode on the Cypher and Infinity AFMs allows rapid acquisition of force curve arrays, because the complete images are obtained in minutes rather than hours, with one image of 256x256 pixels being acquired in under 10 minutes.
  • All MFP-3D AFMs have controls for environmental settings such as the Closed Fluid Cell, the Humidity Sensing Cell, and the BioHeater which allows fluid exchange from ambient temperature to 80 oC. They also offer optional high temperature heaters for samples from 275 oC to 400 oC, and except for Origin, they also allow fitting of the CoolerHeater temperature control stage which regulates temperatures between -30 oC and 120 oC.


  1. K.  Holmberg and A.  Erdemir, Friction 5, 263 (2017).
  2. R.  Carpick and M.  Salmeron, Chem.  Rev.97, 1163 (1997).
  3. Nanotribology and  Nanomechanics  (4th  edition),  ed.  B.  Bhushan (Springer-Verlag, Berlin, 2017).
  4. R.  Bennewitz,  “Friction  Force  Microscopy,”  in  Fundamentals of Friction  and Wear  on the  Nanoscale,  eds.  E.  Gnecco and E.  Meyer (Springer, Berlin, 2015), Ch.  1.
  5. L.  López-Mir, L.  Balcells, S.  Valencia, F.  Kronast, B.  Martínez, J.  J. de  Miguel,  and  C.  Ocal, Cryst.  Growth Des.16, 5479 (2016).
  6. D.  Ogletree, R.  Carpick, and M.  Salmeron, Rev.  Sci.  Instrum.67, 3298 (1996).
  7. M.  Munz, J.  Phys.  D:  Appl.  Phys.43, 063001 (2010).
  8. A.  Labuda, C.  Cao, T.  Walsh, J.  Meinhold, R.  Proksch, Y.  Sun, and T.  Filleter,  “Static and dynamic calibration of torsional spring constants of cantilevers,”  Rev.  Sci.  Instrum.89, 093701 (2018).
  9. A.  Li, S.  N.  Ramakrishna, P.  C.  Nalam, E.  M.  Benetti, and N.  D.  Spencer, Adv.  Mater.  Interfaces 1, 1300007 (2014).
  10. E.  Wassel, D.  Wesner and H.  Schönherr, Eur.  Polym.  J.89, 440 (2017).
  11. S.-W. Liu, H.-P.  Wang, Q.  Xu, T.-B.  Ma, G.  Yu, C.  Zhang, D.  Geng, Z.  Yu, S.  Zhang, W.  Wang, Y.-Z.  Hu, H.  Wang, and J.  Luo, Nature Commun.8, 14029 (2017).
  12. C.  Mate, G.  McClelland, R.  Erlandsson, and S.  Chiang, Phys.  Rev.  Lett.59, 1942 (1986).
  13. X.  Cao, X.  Gan, Y.  Peng, Y.  Wang, X.  Zeng, H.  Lang, J.  Deng and K.  Zou, Nanoscale10, 378 (2018).
  14. D.  Feng, J.  Peng, S.  Liu, X.  Zheng, X.  Yan, and W.  He, Mater.  Res.  Express5, 015026 (2018).

This information has been sourced, reviewed and adapted from materials provided by Asylum Research - An Oxford Instruments Company.

For more information on this source, please visit Asylum Research - An Oxford Instruments Company.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback