User login


You are here

Journal Club Theme of August 2007: Nanoindentation with focus on soft matter

VirginiaLFerguson's picture

Nanoindentation testing of time-dependent materials is becoming increasingly relevant with advances in areas such as polymer science and the study of materials in biological systems. However, mechanical characterization at micrometer or nanometer length scales is not trivial in such materials that may be highly compliant, heterogeneous, or possess unique morphological characteristics. Although many researchers have made progress in overcoming key challenges that are involved in testing non-traditional materials, significant advancements must still be made to optimize testing methodologies and analytical techniques.

There is no one “right” way to perform mechanical testing of viscoelastic materials. The information that is desired drives selection of the testing methodology. The varied testing methodologies and analytical approaches are nicely summarized by Oyen (2006) and in a related discussion on IMechanica in a forum on viscoelastic contact. The exact testing and analytical methodologies are not the intended focus of this Journal Club discussion, rather the intent is to highlight specific challenges that face our community in furthering our ability to study increasingly complex materials using nanoindentation. Plus an additional topic for discussion is proposed with the goal of improving the ability of the larger scientific community to obtain high quality indentation data on time-dependent materials.

Polymers and hydrogels can be engineered in layers or with a gradient in composition, porosity, and/or modulus. The inherent hierarchical nature of biological materials enables substantial variations in composition and structural organization at length scales ranging from nanometers to millimeters. The interface of engineered and biological materials, such as a metallic or ceramic implant that contacts both hard and soft tissues, demonstrates a common and interesting problem where modulus varies significantly at small length scales, say between a titanium implant material and the adjacent bone or soft tissue. Functionally graded materials can be constructed of conventional hard materials, polymers, or hydrogels and possess graded porosity, composition, or modulus permit matching or ingrowth of native tissue in applications such as bone replacement (Kromova et al., 2001). A key issue arises in that a functionally graded material, a sample containing an interface of bone to metal to soft tissue, or even diseased hypermineralized tissue within a more compliant and less viscoelastic region of bone may be examined in a single indent array (shown here in figure 4) without any alteration in the testing parameters between compliant (and time-dependent) versus stiff (and less time-dependent) regions.

Heterogeneous and composite materials. The use of conventional analytical techniques in nanoindentation requires assumption of material homogeneity. In truth, most materials are heterogeneous and can be considered as composites. Nanoindentation measurements in bone, for example, are commonly analyzed using the Oliver-Pharr method. However bone is made up of many materials: collagen, mineral, and water. It contains pores that range in size from nanometers to millimeters. Bone mineral is most often assumed to consist of hydroxylapatite when it is instead a conglomeration of many forms of tricalcium phosphate, octacalcium phosphate, and other minerals that may exist at length scales that are large enough to influence nanoindentation measurements. The complications posed by indentation of nanocomposites are examined by Constantinides et al. (2006) in titanium-titanium boride (Ti-TiB). Arrays of uniformly spaced indents that randomly ‘selected’ Ti, TiB, and TiB2 in four different samples provided a reasonable indication of material composition and an approach for testing and analyzing indentation data from composite materials is presented.  This paper opens the door for discussion of nanoindentation of composites containing a single or multiple time-dependent phases such as presented by Ko et al. (2006) for a biomimetic bone replacement material of hydroxylapatite in a gelatin matrix.

Is a ‘primer’ on indenting soft materials needed? Papers published as recently as this year continue to present indentation data on time-dependent materials that demonstrate creep during the unloading period and analyzed using the Oliver-Pharr method. While disappointing, this common scenario presents an opportunity to discuss how to effectively educate non-experts to perform high quality indentation testing despite a lack of detailed contact mechanics knowledge. It appears that many ‘casual’ nanoindentation users exist in the general scientific community who are want to quickly obtain modulus or hardness data on their specific materials. This sounds trivial until one considers the wide range of both candidate materials for nanoindentation testing as well as the range of potential users’ areas of expertise. Nanoindentation systems are marketed to and used by research physicians, biologists, and others who may not have a strong grasp on mechanics as well as physicists, materials scientists, and engineers!

A key issue is thus raised for discussion in that the mechanics community serves as the educators for the general scientific community. Do we have a responsibility to establish standard approaches for indentation of soft materials to enable general educated user to readily select an appropriate testing approach and method of analysis? Should we establish standards designed to improve data quality and critical evaluation of another’s work? We stand to gain tremendously from the potential knowledge that can be gained by indenting unprecedented and unique materials that may come from scientists in varied disciplines. However our gain is limited by the knowledge that we disseminate, how effectively it is conveyed to the general scientist, and their ability to readily produce high quality data.

While it may be argued as to how well such an effort might catch on, there are many long-standing and effective cases that set precedence for key publications aimed to guide a field of non-experts. One highly successful example lies in the bone research community where histomorphometry, a technique to study bone tissue growth in 2-D, was relatively inaccessible to non-experts prior to the publication of a standard for nomenclature, symbols, and units (Parfitt et al., Journal of Bone and Mineral Research, 1987). While this paper is not suggested for specific review in this Journal Club, it serves as an example where a group of experts enabled the growth of a highly valuable technique by making it more accessible through the sharing of information and standardization of key factors. To date, this article has been cited 2069 times. A similar approach may positively benefit those interested in indentation of time-dependent and complicated materials by creating a set of standards that, say, call for inclusion of a representative force-contact depth plot. Such simple additions would readily reveal critical characteristics, such as the presence of creeping on unloading, and thus aid in the interpretation of the quality of the work presented.


The papers included here for discussion are linked to below: 

(1)  Microhardness studies on functionally graded polymer composites. M Krumova, C Klingshirn, F Haupert and K Friedrich. Composites Science and Technology: 61 (4): 557-563, 2001. 

(2) Grid indentation analysis of composite microstructure and mechanics: Principles and validation. G Constantinides, KS Ravi Chandran, FJ Ulm and KJ Van Vliet. Materials Science and Engineering: A 430 (1-2): 189-202, 2006. 

(3) Mechanical properties and cytocompatibility of biomimetic hydroxyapatite-gelatin nanocomposites. CC Ko, ML Oyen, AM Fallgatter, JH Kim, J Fricton, WS Hu Journal of Materials Research: 21 (12): 3090-3098, 2006. 




Initial points to consider include:

-        Current gaps in our knowledge and ability to adequately test materials with complicated functional properties or structures that also possess time-dependence

-        How are we currently limited by experimental capabilities (i.e., instrumentation and methods) and analysis techniques?

-        Is it reasonable or even feasible to consider the development of standard approaches? Or is this a gross oversimplification that would not be applicable to the broad range of materials that are candidates for analysis using nanoindentation?



Links to additional references are included above. Please feel free to request pdf versions of files from me via email.


PDF icon Parfitt.pdf1.41 MB


Xiaodong Li's picture

Thanks for kicking off this month's J-club discussion. The topic is VERY timely. We know that the current nanoindentation theories were initially established on solid, rigid materials. How do we extend the nanoindentation application to soft matter? There are challenges and opportunities for us. Can the existing nanoindentation instruments, methodologies and theories offer us "solid" capabilities to obtain viscoelastic/plastic properties? Are the storage and loss modulus values obtained from nanoindentation comparable with those from the conventional tensile/bending DMA? The papers you listed are excellent for discussion. I have been working on some soft matter (I feel very challenging...), below are some of the papers:

Xiaodong Li, Hongsheng Gao, Wally A. Scrivens, Dongling Fei, Xiaoyou Xu, Michael A. Sutton, Anthony P. Reynolds and Michael L. Myrick, "Nanomechanical Characterization of Single-Walled Carbon Nanotube-Reinforced Epoxy Composites," Nanotechnology,15 (2004) 1416-1423.

Xiaodong Li and Bharat Bhushan, "A Review of Nanoindentation Continuous Stiffness Measurement Technique and Its Applications," Materials Characterization, 48 (2002) 11-36.

 I believe that this month's J-club discussion will stimulate an integrated effort from both experimental and modeling communities to team up on this challenging area.

VirginiaLFerguson's picture

Thank you for your positive comments, Xiaodong. I agree that this discussion is needed to push the envelope. The community as a whole seems to have generally focused on conventional methods of analysis such as Oliver-Pharr to extract data from soft and/or time-dependent materials. I myself have studied bone and other mineralized, viscoelastic tissues using a prolonged creep hold such that I could obtain values for modulus and hardness upon unloading. This method, while useful, negects perhaps the most interesting aspects of time-dependent materials - that is the time dependent behaviour itself!

The key factors in better exploring softer and time-dependent materials may thus lie in development of improved instrumentation or novel analytical approaches using currently available instrumentation and testing methods. One example lies in a powerful but simple analytical approach of deconvolution of conventionally obtained load-contact depth data in a recent (2006) paper by Michelle Oyen - linked to here. Her simple approach provides a useful insight into how elastic, plastic, and visco behaviour contributes to overall nanoindentation response. This approach is not limited to bone and teeth and may be useful in the study of engineered nanocomposites or functionally varied/graded materials.

Also, as Xiaodong suggests, the advent of continuous stiffness measurements (CSM) recently enabled a large advancement in nanoindentation. However, advances in instrumentation may be necessary for extension of CSM to time-dependent materials. One example that comes to mind is in a composite that contains two materials of varying time-dependence - see Ko et al. paper above where hydroxyapatite was embedded in a gelatin matrix. Use of CSM to test both constituents in such a material is highly problematic due to their different stiffness and time-dependent responses. Is it therefore appropriate to view CSM as a reasonable method for analysis of time-dependent materials, or should we resign it to a fate of being used primarily to find the surface of soft materials (via significant shifts in harmonic contact stiffness - which it does very well!). In any case, progress in using CSM and other techniques for soft materials lags that of stiff, non-viscoelastic materials and presents a great opportunity for many. Works described in this J-club (see multiple posts below) illustrate that good progress is being made and that advancements in our understanding will only add to our understanding of these fantastic soft, time-dependent, and complicated materials.

Rohit Khanna's picture

Hi, Ferguson... i read your opinion about the current state of the research about mechanics of soft materials. I agree with you that there are also non-experts also in this field which may be biologist, material scientists etc.... May be because of lack of exposure in this field, the research has not progressed much in this direction. It is practically impossible for just one or two groups to be active in this area, the contribution has to come from a large group of researchers and then, only we can expect advancement in science and technology. I personaly believe that experts in this field should take the initiative to educate the researchers who are non-experts in this field. It may be good to organize educational workshops. I had been to such a workshop, quite recently in UIUC, on Cell mechanosenstivity, which dealt with various topics of interest like cell adhesion, cell mechanics, cell-biomaterial interactions etc. Experts in the field like Prof. Dennis Discher (UPENN), Prof. Michael Sheetz and many others, presented a review of the literature in these topics, and we also had hands-on experience on some experiments.

I hope to know your opinion on this.

Best wishes,

Rohit Khanna 

Yong-Wei Zhang's picture

Some of our previous work on nanoindentation on polymer film/substrate system is closely related to the Journal Club Theme.  It is believed that two issues are important in determining mechanical properties of a viscoelastic polymeric thin film on a substrate by indentation. The first issue is due to the influence of the substrate on the mechanical behaviors of the thin film, causing difficulties to extract “film-only” parameters. The second issue is due to the influence of the viscoelasticity of the polymeric film itself, causing the load-displacement curves to depend on the loading or unloading rate.  

Two papers have been published to deal with these two issues: The first one (Zhang et al, 2004) is based on creep indentation test using a flat-end punch; the second one (Zhang, et al, 2006) is based on the relaxation test using a sharp indenter. In both papers, theoretic analysis and experiment were performed and contrasted. In addition, reverse analysis was carried out to extract elastic and viscoelastic material parameters.

Comments and criticisms on these papers are welcome. 


C.Y. Zhang, Y.W. Zhang and K.Y. Zeng, “Extracting the mechanical properties of a viscoelastic polymeric film on a hard elastic substrate”, J. Mater. Res. 19(10), 3053-3061 (2004). 

C.Y. Zhang, Y.W. Zhang, K.Y. Zeng, L. Shen, Y.Y. Wang, “Extracting the elastic and viscoelastic properties of a polymeric film using a sharp indentation relaxation test”, J. Mater. Res. 21(12): 2991-3000 (2006).  

Xiaodong Li's picture

Thanks Yong-Wei for listing two excellent JMR papers for discussion. I agree that there are still a lot of fundamental questions and experimental issues. For instance, what loading/unloading rate is needed to avoid the time-dependent effect for obtaining hardness and elastic modulus? In general, a holding segment is needed to avoid creep effect on unloading for accurate measurement of unloading stiffness. I think that modeling work may help provide some insightful guidelines for nanoindenter manufacturers and experimental validation.

MichelleLOyen's picture

I have to object to the question here, "what loading/unloading rate is needed to avoid the time-dependent effect for obtaining hardness and elastic modulus."  For certain classes of materials, the time-dependence in response is a fundamental aspect of the material's mechanical behaviour.  People do not seem to have any difficulty regardling the elastic and plastic behavior of a material as two fundamental and complementary aspects of the response, both worth measuring, and yet many persist in regarding the time-dependent response as an annoyance that must be avoided.  There are plenty of techniques for exploring the viscoelastic or viscoelastic-plastic behavior of a material by indentation, as have been discussed elsewhere on this site.  In many instances, these techniques involve examination of the loading response and/or a load-and-hold response similar to a traditional creep test.  These techniques are as such fundamentally different from an Oliver-and-Pharr technique emphasizing the unloading response.   Trying to "correct" Oliver-Pharr data for pesky viscoelastic effects fundamentally misses the point in that the viscous deformation in an appropriately-chosen experimental time-frame relevant to the applications being considered (i.e. the reasons for characterizing the material in the first place)  is a fundamental aspect of the material's behavior and deserves attention in its own right.

Xiaodong Li's picture

Thanks. Very good point. I agree that we need to find out more from the holding and unloading. I think that modeling work may provide more insightful information that helps the experiment. Tons of papers have been published on nanoindentation of solid, rigid materials while very less on soft matter. I believe that there is still a lot we can do in this area.

MichelleLOyen's picture

I hate to be the persistent contrarian on this topic, but again I beg to differ.  There have been a large number of papers published on indentation of soft matter.  There are a few points to make about this: (1) soft matter studies frequently involve the use of custom instruments or AFM-type instruments and not traditional "nanoindenters" (e.g. MTS Nanoinstruments, Hysitron, Micromaterials, etc.)  (2) soft matter studies are typically published in other venues than the venues/journals in which the hard materials indentation studies are published. 

The subject of soft matter indentation is extremely well covered in biomechanics/biophysics journals--indentation is the number one most popular technique for characterization of the soft, hydrated tissue articular cartilage and there is a wealth of literature going back 30 years.  Yes, not all of it is "nanoscale" but the basic contact mechanics and experimental issues associated with soft, hydrated matter is not completely scale dependent.  And certainly in the recent literature, in the last 5-10 years, much of the work that is going on IS getting down to sub-micron scales including things like multiscale cartilage characterization and probing cartilage in vivo with an AFM

Key to the challenges with using classic hard materials "nanoindenters" for soft materials include:

  • there is a current lack of appropriate built-in software algorithms to analyze viscoelastic responses.  If the indenter manufacturers would just add these modules--based on well-established contact mechanics--to the software then people could stop using the unloading method/Oliver-Pharr-after-a-hold approximation.
  • the stiffness scales in soft materials are not comparable to those in hard/stiff materials.  In hard materials, we work with nm-mN scales.  In sub-MPa materials, we work all the way down to mm-mN stiffnesses.  Again there are optimizations that could be done by instrument manufacturers to facilitate these changes in contact stiffness scale.  In the absence of such changes, AFM type instruments are sometimes easier to work with.
  • sharp tips such as a Berkovich cone or a classic AFM tip are not ideal for probing soft materials.  The AFM community has tended towards spherical micron-scale radius tips for soft materials and I personally have done the same.  The biomechanics community still tends to emphasize flat punches although alignment and corner issues become increasingly important at fine length-scales.
  • in soft matter, the bonding is fundamentally different than in metals and ceramics.  Due to the relative importance of non-covalent bonds, including hydrogen bonds and van der Waals interactions, there are probably also multiple lengh-scales of importance in soft matter.  Indentation is unlikely to catch all aspects of the nanomechanical behavior of these types of materials, which frequently exhibit quite different responses in tension and compression when covalent vs. non-covalent (respectively) interactions are emphasized.
  • finally, there is an open debate about the best ways to characterize time-dependence.  Depending on the application, frequency-based methods (using a lock-in amplifier and approximating DMA type data) and time-based methods (creep or relaxation tests) have advantages and disadvantages.  Far more work needs to be done on the comparison between these methods in contact--especially at nanoscale--and these methods in bulk testing such as confined and unconfined compression (again, especially at small scales in microcompression).
Rohit Khanna's picture

Apart from the review of biomechanics literature presented by Michelle, Li and others, there are another important issues which need to be addressed. I agree that bonding is quite different in soft materials. Another research field which is not majorly touched by many of the researchers is the adhesion effects seen during unloading from a soft substrate. They do affect the elastic modulus  measurement. There have been recent studies on such effects on soft material like polydimethysiloxane (PDMS). We know that there are limitations on the part of instrumentation unavailability, due to which accurate determination of time-dependent properties of soft materials has not been succeded to much extent. But i am surprised, why research has not progressed in looking at adhesion effects in nanoindentation. There has been model development to take into account such effects for last 35 years. First paper had come from Prof. K L Johnson who did a pioneered in establishing the theory of adhesion effects.

One can not ignore adhesion effects without which accurate determination of elastic modulus is not possible as far as my understanding is concerned.

I have done some review on this topic with the hope that it does not go un-noticed. 

Adhesion is observed as a region of negative load during unloading in a load-displacement curve. Adhesion between the diamond tip and the sample can interfere with the measurement of indentation modulus using the compliance method for polymers and tissues [1-4]. Recent research reports on nanoindentation of soft polymers [1,4] have mentioned that the compliance method overestimates the modulus when there is significant tip-sample adhesion. For indentation with spherical tip, Johnson-Kendall-Roberts (JKR) [5] adhesion model has been shown to be give more accurate measurement of sample modulus [1,4,6]. The JKR method can be described as, first starting the indent out of contact and then capturing a full force curve as the tip approaches, indents, and retracts from the sample [6-8]. These force curves have been commonly used in AFM to measure the adhesive forces [9,10], but not yet applied to many nanoindentation studies of biomaterials.

 1. Carrillo, F., et al., J. Mater. Res.(2005) 20, 2820
 2. Klapperich, C., et al., J. Tribol. – Trans. ASME(2001) 123, 624.
 3. Grunlan, J. C., et al., Rev. Sci. Instrum.(2001) 72, 2804
 4. Carrillo, F., et al., J. Mater. Res.(2006) 21, 535
 5.  Johnson, K. L., et al., Proc. R. Soc. London, Ser. A(1971) 324, 301.
 6. Ebenstein, D. M., and Wahl, K. J., J. Colloid Interface Sci.(2006) 298, 652
 7. Wahl, K. J., et al., J. Colloid Interface Sci.(2005) 296, 178
 8. Giri, M., et al., Langmuir (2001) 17, 2973.
 9. Butt, H. J., et al., Surf. Sci. Rep.(2005) 59, 1
10. Cappella, B., and Dietler, G., Surf. Sci. Rep.(1999) 34, 1


Best regards,

 Rohit Khanna

MichelleLOyen's picture

Very nice summary, Rohit. 

Just one comment on perhaps why this has been a bit neglected as a topic... many of the biological materials of interest to researchers are fundamentally hydrated.   For example, tissues in the body are in wet environments, as are plant materials.  The hydration state has a fundamental effect in these materials' mechanical behavior in general, but also seems to limit the adhesion with a hard, stiff indenter.  In fact, I had a colleague once tell me to put a thin water film on PDMS and that would limit adhesion effects in AFM-based indentation analysis. 

Rohit Khanna's picture

  • Thank you Prof Michelle for the response. I did not get much responses from imehanica community. may be this subject is not of great interest to many of imechanica members. I hope by coming MRS meeting, i will have a good reason to tell you, why research in this direction is important. You are absolutely right in saying that hydration state limits adhesion. It may be good to study on fundamental science aspect.

I have a poster in your symposium on Fundamentals of nanoindentation and nanotribology. It will be great to have useful discussion with you. This time, i don't see a big crowd of researchers for this symposium. 

Best wishes,



Hello mechanicians,

I am acquiring a Quanta FEG SEM (an environmental SEM) for other types of studies, but was curious about its potential use for indentation, including nanoindentation?

Allows operation/imaging up to 20 Torr pressure of H2O.

Thanks, Rod

Xiaodong Li's picture

Thanks! I am very happy to know that you are going to have a Quanta FEG ESEM. That is great! I do see the potential use for indentation. I remember that Prof. Kato's group at Tohoku University and a group in Cavendish Lab at Cambridge University have done some micro/nanoindentations in-situ in SEM. I think that in-situ imaging and indentation load-displacement curve together offer more than conventional nanoindentation tests. I look forward to your exciting results.


I am doing research about the bone testing (compression and fracture toughness) using impact instrument (SHPB). Experimentally, it was difficult to obtain the ideal specimen preparation. Specimens are prepared with two different groups. one is longitudinal direction and another is transverse direction. Preliminary testing results showed the rate dependence (from 0.001/s to 750/s). Later, I will update the experimental phenomena and I appreciate the papers.

Xiaodong Li's picture

Thanks for the post. I think that you can cut your samples using microtome method if you are working at the small scale. I cut seashells and polymer composite samples before and it worked very well.

Hello Professor

Thank you for your suggestion.

We used the water jet to prepare bone specimens. The microtome method will be worked for ideal specimens.

MichelleLOyen's picture

A slow diamond saw (like Buehler Isomet) is also very common and quite useful for bone and tooth specimen preparation, and allows for samples with more variety in sizes than microtoming. 

Gregory FAVARO's picture

Dear all,

We are working a since 5 years in Bone and dentine application with Geneva Hospital
(Prof. Patrick Amman).
Actually in Geneva
they cut the bone with a microtome and polish its but the biggest difficulty is
to measure the Nanoindentation with the good hydratation! Actually we re-hydrated
the bones but we have develop a liquid cell to measure the bone in

We really need to compare the result in wet atmosphere if we want simulate
the reality!

Please see this very interesting article about intrinsic Nanoindentation for

Do not hesitate to contact me for furteher information 

Subscribe to Comments for "Journal Club Theme of August 2007: Nanoindentation with focus on soft matter"

Recent comments

More comments


Subscribe to Syndicate