User login


You are here

Journal Club Theme of May 2013: Mechanics of biological and bio-inspired supramolecular assemblies

Luis Ruiz Pestana's picture

Supramolecular assemblies are dynamic structures composed of simple molecular building blocks that self-assemble into hierarchically organized complexes driven by reversible (weak) intermolecular interactions (e.g. hydrogen bonds). This bottom-up approach to materials synthesis is utilized in nature to create flaw tolerant, highly functional materials that exhibit extraordinary mechanical properties arising from parallel clusters of weak interactions such as hydrogen bonds that work collectively. Such geometric arrangements are observed in stable biomolecular assemblies such as spider’s silk and amyloid plaques (see for instance an earlier journal club entry on protein materials node/2653 ).

In this journal club, we highlight some of the recent advances in our understanding of supramolecular assemblies in biological, biomolecular and bio-inspired systems. The mechanical behavior of weak bonds lies at the heart of this discussion. Weak bonds are reversible, follow non-deterministic dynamics, and their lifetime depends on the magnitude, direction and rate of the applied load, with higher loads leading to stiffer responses. Force dependent single barrier kinetic models such as Bell’s model (G.I. Bell, Science, 1978) provide a convenient starting point on the discussion of the breaking mechanisms of weak reversible bonds. This simple model is usually used in conjunction with Monte Carlo (MC) techniques to study the collaborative response of bond clusters with specific bond network architecture. It has also been instrumental in interpreting single molecule atomic force microscopy experiments and molecular dynamics simulations on protein unfolding, as well as cell adhesion and motility experiments. We refer the reader to the work of Prof. Ben Freund, Prof. Huajian Gao and colleagues for the theoretical predictions of the optimum size of a molecular bond clusters under various boundary conditions. These studies illustrated that larger clusters are favored by bond reversibility while the nonuniform load distribution between the individual bonds favors small ones [1]. Additionally, it was found that the optimal size can arise from a transition between surface- and bulk-dominated interaction regimes at the nanoscale [2].

The counterintuitive mechanical behavior of supramolecular materials (high strength from weak interactions) often arises from size and confinement effects at different length scales or hierarchies (Buehler & Keten, Reviews of Modern Physics, 2010). Several examples can be found in the literature of materials that through size restrictions or geometrical constraints achieve optimum performance [3]. Keten et al. showed, using MD simulations that beta-sheet crystals, responsible for the high strength of spider silk, use geometric nanoconfinement to achieve optimum stiffness and toughness. By limiting their size, the failure mechanisms of the beta-sheet become restricted to the one that optimizes the mechanical response (stick-slip mechanism under uniform shear), emphasizing the importance of size and confinement effects in supramolecular systems [4]. More recently, designer peptides and organic polymers have been developed to leverage hydrogen bonds and other non-covalent interactions in new bio-inspired high performance functional materials that are capable of exceptional stimuli-responsiveness, self-healing, and mechanical properties. However, numerous challenges lie ahead in modeling the mechanical response of these novel materials systems. In particular, the dynamic character of supramolecular assemblies, arising from the stochastic nature of the non-covalent bonds under thermal fluctuations, poses an important challenge on the assessment of their mechanical response using classical engineering analysis tools. Meso-scale models based on statistical mechanics and coarse-grained approaches are critically needed to explain the interplay between assembly, structure and fracture. An additional consideration is that in self-assembling systems, coupling between these three factors are observed at all stages of the materials development continuum, including processing.

In our group, we are currently studying cyclic peptide nanotubes (CPNs), a type of supramolecular organic nanotubes that exhibit outstanding mechanical stiffness and transport properties that match those of biological transmembrane proteins, having an alluring potential as synthetic nanopores in membranes for applications such as carbon capture or low energy water desalination. Using Bell’s model and MC simulations on a dimer, we characterized the load dependence strength of a CPN and were able to predict the probability distribution of the failure location along the nanotube under bending deformation [5]. These findings are important for understanding the fragmentation and ultimate size-distribution of these self-assembling systems, which may involve secondary growth mechanisms arising from tube fracture similar to self-replicating systems in biology (e.g. amyloids).

Our more recent investigations have focused on understanding how attachment of polymers to the nanotube building blocks can be used to control the mechanical properties and stacking order in binary mixtures of functional peptides. The meso-scale mechanics of these polymer-peptide conjugates is enigmatic in many ways, involving competing effects such as enthalpic gain through hydrogen bonding versus entropy loss upon binding due to confinement, as well as kinetic effects characteristic of non-equilibrium self-assembly processes.  We are currently establishing a mechanics approach to quantify the forces arising from the nanoconfinement of entropic polymer springs and weak interactions. We anticipate that if successful, this approach can be exploited to direct the stacking order of mixed CPs within block-copolymer membranes to generate artificial ion channels with well-defined interior properties. More broadly, new theories and mechanics models that lead to a fundamental understanding of the physics of supramolecular materials will be crucial for discovering strong, tough and multifunctional systems inspired from biology.

Luis Ruiz (PhD student, Theoretical and Applied Mechanics, Northwestern University)

Computational Nanodynamics Laboratory (PI: Prof. Sinan Keten)


Suggested Reading:

 1.         Lin, Y. and L.B. Freund, Optimum size of a molecular bond cluster in adhesion.  Physical Review E, 2008. 78(2): p. 021909.

2.         Yao, H., P. Guduru, and H. Gao, Maximum strength for intermolecular adhesion of nanospheres at an optimal size . Journal of The Royal Society Interface, 2008. 5(28): p. 1363-1370.

3.         Gao, H., et al., Materials become insensitive to flaws at nanoscale: Lessons from nature . Proceedings of the National Academy of Sciences, 2003. 100(10): p. 5597-5600.

4.         Keten, S., et al., Nanoconfinement controls stiffness, strength and mechanical toughness of [beta]-sheet crystals in silk.  Nature materials, 2010. 9(4): p. 359-367.

5.         Ruiz, L., et al., Persistence length and stochastic fragmentation of supramolecular nanotubes under mechanical force . Nanotechnology, 2013. 24(19): p. 195103.


swcranford's picture

Great post and some interesting points Luis.

Exploiting robust self-assembly is a necessary step in both the development of new materials as well as understanding the assembly processes of Nature's materials (or mistakes, in the case of  amyloids). One need only turn to the recent advancement of “DNA/protein origami” [1] as an exciting example.

Bond confinement and cooperativity have shown to be important, or “parallel clusters of weak interactions... that work collectively”, but it seems to always in consideration of hydrogen bonding. I am curious (and pose the question) that if hydrogen bonds have some sort of advantage in clusters, can other bond types and interactions follow suit? For example, pi-pi stacking has been successfully used for supramolecular assembly [2], but to the best of my knowledge, no studies have explored the cooperativity of such bonding (if it even exists). Other candidates may include disulfide bonds, or electrostatic interactions (to name just a few).

The other interesting behavior, of course, is the potential for “high strength from weak interactions”, exploiting reversible, non-covalent systems. From an energetic perspective, the “low cost” of such bonds is ideal for self-assembled soft matter system (as illustrated by Nature, most likely by biological necessity). However, if high strength can be attained from weak interactions, one could pose that even higher strength materials can be designed from strong interactions. There is obviously a trade-off with cost of assembly, but why must the “cooperative bonds” be weak? Perhaps the stronger the bonds get, the more “narcissistic” they become Cool.

Simply put, if we could (hypothetically) replace the H-bonds in spider silk with covalent bonds, how much stronger would the silk become?


[1] Peplow, Protein gets in on DNA's origami act, Nature News & Comments, 2013 (and citations therein). Link:

[2] Surin et al., Supramolecular Organization of ssDNA-Templated π-Conjugated Oligomers via Hydrogen Bonding,
Advanced Materials, 21(10-11), 1126-1130, 2009. Link:


Steve Cranford

Assistant Professor

Laboratory of Nanotechnology In Civil Engineering (NICE)

Department of Civil and Environmental Engineering

Northeastern University


Thanks for the note and the interesting thoughts. This work is tremendous!! One comment that I would like to make it the changes that will likely happen in the future with research in this area. Someone once told me that "Biology" research is mostly aimed at human health and disease. He then went on to say that over the next few decades a significant shift will move toward non-disease, but still extremely important research in biology. I thought that was very interesting (and I believe it). This work is a great example of this and work we are doing continues to push more into non disease areas although we still do a number of things linked to disease. Just a thought after reading this excellent work above! Best wishes and keep up the exciting findings. Phil

keten's picture

Dear Phil,

I couldn't agree more with your comments. As you mentioned, applications of mechanics to disease related problems will continue to be an area of great interest to mechanicians and researchers across many disciplines. The complexity of living systems and pathologies are astonishing and one can expect health and disease to be a field that will continue to grow for many decades. On the other hand, it is also exciting to see that advancements in our understanding of the physics of biomolecules and biological systems, such as ion channels, have motivated synthetic approaches to replicate nature's building blocks, factories and functionalities. There are many unsolved mechanics problems related to production and use of bioinspired and biomolecular materials. As the applications to nontraditional areas such as energy, environment and structural materials begin to emerge, I believe that the the community will continue to embrace and promote a broader definition of nano and biomechanics. It seems like an exciting time to be working on the interface between biological and engineered materials.

Thank you for very much for your insightful comments.

- Sinan

Luis Ruiz Pestana's picture

Thanks Steven,

You raise some very good points that are central to the issue of supramolecular systems.

In the first place you wonder if other types of weak interactions are capable of forming collaborative networks, similarly to hydrogen bonds. There have been some recent experimental developments in the field of self-healable supramolecular materials where networks of weak interactions such as metal-ion complexation [1], or pi-pi stacking [2] have been exploited. For a recent review of the different available strategies I recommend [3]. There are also biological systems that display remarkable mechanical properties using alternative weak bonds such as dopa-metal bonds in mussel threads [4] or sacrificial ionic bonds in bone [5].

I think that cooperativity can be achieved by other types of weak interactions although the morphology of the bond network has to be adapted to exploit the different peculiarities of the interaction (a good example is presented in [5]). The lack of studies focusing on the cooperativity or strength of networks of this kind of interactions can be partially due to the fact that the morphologies can be complex compared to hydrogen bond networks.

Regarding your second comment, I agree with the statement that high strength materials can be designed using stronger bonds. However the synthesis routes will most likely differ from autonomous self-assembly. The self-assembly process can be hindered by kinetic trapping of the system in local minima, and kinetic effects are enhanced by increasing the strength of the interactions (increases the relaxation time of the system) or their specificity (higher specificity leads to lower growth rates). Thus, the use of strong interactions will most likely result on poor yield of target systems.

Putting this concern aside, you raise a very interesting theoretical question; Does the cooperativity of the cluster decreases by increasing the strength of the individual bonds? I think this is an open question but I am going to give you my personal view based on a simplistic interpretation of the results presented by Keten and Buehler in [6]. The first premise about cooperativity is that the external force has to be applied to the bond cluster as to produce a failure mode where several bonds can share the applied force. In this work, the number of h-bonds in a cluster that break simultaneously under applied external force (Ncr) scales with the individual bond energy (E0) as Ncr~1/E0. This scaling suggests that stronger bonds lead to lower cooperativity, in the sense that less number of bonds will break simultaneously. So it seems that the strong bonds can become egotistic load bearers after all.

As for the theoretical strength of spider silk if all the h-bonds are replaced by covalent bonds. There is a study where the mechanical properties of spider silk are improved by infiltrating metals into the protein matrix [7]. In the mineral-infused silk the hydrogen bonds are replaced by metal coordinated or covalent bonds, enhancing the strength of the material. It is also hypothesized a certain structural change, through which some parts of the largest beta sheet crystals (weaker than the smaller ones) would become amorphous regions, leaving the modified silk with optimum size crystals. I think that reversible metal-coordination bonds can improve the silk but I suspect that by introducing covalent bonds you may suppress the main energy dissipation mechanisms of the silk decreasing its toughness. Also, stress concentration effects around the newly covalent-bonded beta sheet crystals (very hard inclusions) can also play an important role.


1. Burnworth, M., et al., Optically healable supramolecular polymers. Nature, 2011. 472(7343): p. 334-337.

2. Burattini, S., et al., A Healable Supramolecular Polymer Blend Based on Aromatic π-π Stacking and Hydrogen-Bonding Interactions. Journal of the American Chemical Society, 2010. 132(34): p. 12051-12058.

3. Wojtecki, R.J., M.A. Meador, and S.J. Rowan, Using the dynamic bond to access macroscopically responsive structurally dynamic polymers. Nat Mater, 2011. 10(1): p. 14-27.

4. Harrington, M.J., et al., Iron-Clad Fibers: A Metal-Based Biological Strategy for Hard Flexible Coatings. Science, 2010. 328(5975): p. 216-220.

5. Hartmann, M.A. and P. Fratzl, Sacrificial Ionic Bonds Need To Be Randomly Distributed To Provide Shear Deformability. Nano Letters, 2009. 9(10): p. 3603-3607.

6. Keten, S. and M.J. Buehler, Geometric Confinement Governs the Rupture Strength of H-bond Assemblies at a Critical Length Scale. Nano Letters, 2008. 8(2): p. 743-748.

7. Lee, S.-M., et al., Greatly Increased Toughness of Infiltrated Spider Silk. Science, 2009. 324(5926): p. 488-492.


Guy M. Genin's picture

Hi Luis and Sinan, 

Your post and your work have explained very nicely how optimal stiffness and toughness can be achieved using very weak bonds.  But why doesn't nature use strong bonds?  Are weak bonds actually better, or does nature simply make the best of its limited tools?


- - - - - - - -
Guy M. Genin
Department of Mechanical Engineering and Materials Science, Washington University in St. Louis
Department of Neurological Surgery, Washington University School of Medicine



Luis Ruiz Pestana's picture

Dear Guy,

That is a very good point that you raise, and somehow in
line with Steven's previous comment.

My view in this issue is that weak bonds are a requirement
for self-assembly, so at the minimum, they are a necessary "burden" that
biological systems have to endure. However, they also provide some direct
advantages over materials based on strong, covalent bonds.

Traditional engineered materials (e.g. metals, ceramics,
etc.) based on covalent interactions, are usually not suited for many
applications where multifunctionality or controlled reversible transformations
upon stimulation are required.

It is for these applications that weak interactions may
offer some advantages. For example, while collaborative clusters of weak bonds
can still display high strength, the energetic cost to reform a weak bond is
almost negligible compared to the cost of establishing a new covalent bond.
This makes weak bonds especially appealing for self-healing materials.

Biological materials are always in need of multifunctionality
and reorganization capabilities, so in this context, weak interactions usually
outperform covalent bonds. When nature is in need of higher strength, it
usually recurs to clever strategies such as composites and hierarchical architectures
to improve the mechanical performance while retaining the multifunctionality,
rather than stronger bonds.



Subscribe to Comments for "Journal Club Theme of May 2013: Mechanics of biological and bio-inspired supramolecular assemblies"

Recent comments

More comments


Subscribe to Syndicate