Australasian Science: Australia's authority on science since 1938

Crystals So Flexible They Can Be Tied in a Knot

Credit: Leigh Prather

Credit: Leigh Prather

By John McMurtrie & Jack Clegg

The ordered structure of most crystals makes them brittle and inflexible, but the discovery of crystals with elastic properties opens a range of new uses in emerging technologies.

Crystals are beautiful objects. They have been admired for millennia because of their intriguing properties, and in many cultures are thought to be imbued with magical properties. Crystals, do however, underpin a wide variety of modern technologies, including semi-conductors and lasers, which are used in everything from mobile phones to space-shuttles.

The large crystals that we encounter in our natural environment – like quartz, salt and gemstones – are intriguing because of their clean faces, sharp corners and symmetrical features. Often it is how crystals interact with light that captures our attention. Crystals are translucent and can be highly coloured. Light also reflects off the faces of crystals, causing them to glint and gleam.

As early as the 17th century, scientists used these properties to start to understand the behaviour of light and answer fundamental questions of physics. Newton used prismatic crystals to split and recombine white light into its coloured constituents, while Danish scientist Rasmus Bartholin used calcite crystals to discover how light could be polarised – the technology that underpins 3D movies.

As well as diffracting and refracting lights, crystals also diffract X-rays and neutrons, which allows us to use these forms of radiation to probe the atomic structure of materials. The discovery of this led to the 1915 Nobel Prize in Physics to William and Lawrence Bragg. Researchers now routinely grow crystals of new materials to analyse their crystal structure. More than one million crystal structures have now been reported.

Crystals are composed of atoms or molecules that are arranged in highly ordered arrays that repeat infinitely within the boundaries of the crystal. It is the repeating molecular nature of crystals that give rise to both the inherent beauty of crystals and their useful properties.

The highly ordered state of crystals makes them hard, brittle and inelastic. Mechanical stresses such as bending, stretching or compression cause crystals to crack, break or even shatter irreversibly (try googling “hitting quartz with a hammer”). The crystalline allotrope of carbon, diamond, is extremely hard while the non-crystalline allotrope, graphite, is so soft it is used in pencils.

While the optical, electronic and magnetic properties of crystals make them attractive for use in technological applications, their inflexibility has limited their use in emerging technologies like flexible electronics and optical devices. But that may be about to change.

In recent years it has become apparent that the simple view that crystals are hard and brittle needs some revision. About 10 years ago, researchers in India reported that they had found crystals of a simple organic molecule that didn’t shatter when bent. While these crystals could bend, they did so irreversibly. This type of mechanical response is termed “plastic deformation”, and is observed in metals and polymers.

Even more recently has been the discovery of crystals that can bend reversibly, returning to their original shape after mechanical deformation. This type of mechanical response is termed “elastic deformation”. By definition, every material undergoes some amount of elastic bending before breaking, but in many materials it is not significant enough to be observed.

Elastically flexible crystals are important because of the reversibility of the bending. Knowing how to engineer them will allow the development of smart materials that respond to changes in their environment, like temperature, pressure or the application of force, and will be used in emerging technologies such as aeroplane and spacecraft components or in sensors and electronic devices.

Our recent research, published in Nature Chemistry (, suggests that not only is significant elastic flexibility in crystals possible but it might not be as rare as previously thought. We found that crystals of the well-known compound Cu(acac)2 are so flexible they can be tied reversibly into a knot.

Cu(acac)2 isn’t a new compound. It was first made in the late 1800s, and is frequently made in chemistry teaching laboratories all over the world. What is new about these crystals is the way the molecules are arranged within the crystals. In this case the molecules, which are flat, stack upon each other into long chains that run the length of the crystal. There are relatively strong interactions between each molecule in the stack, but very weak interactions in the two orthogonal directions.

One of our co-authors at The University of Queensland, Dr Michael Pfrunder, grew some crystals of the closely related compound Cu(Bracac)2 for X-ray analysis and noticed that they were much more flexible than normal crystals. After finding this first example of flexible crystals we hypothesised that because we can control the chemistry of the molecular building blocks on the atomic level we might also be able to control the flexible properties of the crystal on the macro-scale.

To test this hypothesis, PhD student Anna Worthy made a number of derivatives of our first lead compound by systematically changing small parts of the molecules that make up the crystals. While we discovered the first example accidentally, we’ve now made eight examples of elastically flexible crystals.

To our surprise, the most flexible sample we have found so far was the most well-known: Cu(acac)2. Anna was able to grow blue crystals of this compound, which were about the width of a fishing line and up to 5 cm long. These crystals can be bent significantly many times, and spring back into an unbent state with no signs of breaking or cracking. We filmed Anna showing this ( Most notably, when bent they also keep the useful properties of crystals.

We then set out to quantify just how flexible the crystals were. We performed mechanical testing on the crystals using fairly common techniques in materials science and engineering, including nano-indentation, three-point bend tests and tensile-strain tests.

  • Tensile-strain tests involve clamping two ends of a material and measuring how much force is required to stretch a material. Our crystals could be reversibly stretched by up to 4.4% of their length before breaking, and had a tensile strength of 8.0–22 MPa, which is very similar to a non-crystalline material like polyethylene.
  • Three-point bend tests measure how far a sample can be bent under the application of a certain stress. We found that the crystals had a flexure modulus in the range of 2–8 GPa, which is very similar to nylon (2–5 GPa). This test also allowed us to demonstrate the reversibility of the bending by repeating the bending several times on the same sample. Above 60 MPa of strain we saw some evidence of irreversible plastic deformation.
  • Nano-indentation allows us to measure the hardness of individual faces of the crystals. The results showed us that different faces of the crystal had quantifiable differences in their hardness. One face had a hardness of 200–240 MPa while another was significantly harder (380–400 MPa). This is of similar hardness to many organic molecules and polymers, but much lower than inorganic materials and ceramics.

Because the samples maintain their crystalline properties even when bent, we could use X-rays generated by the Australian Synchrotron to analyse changes in the crystal structure at the atomic level when the sample was bent (for a video see By carefully aligning a bent crystal perpendicular to the X-ray beam, Dr Arnaud Grosjean was able to use micro-focused radiation to map how the spacing and orientation of individual molecules changed along the cross-section of the crystal.

We found that the individual molecules reversibly rotate as the crystals are bent, allowing for the crystal to expand on the outside of the loop while simultaneously getting smaller on the inside. It is this simultaneous compression and expansion that is required for elasticity.

The method that we developed to measure the changes in the crystals during bending could also be used to explore flexibility in any other crystals. This is an exciting prospect given that there are millions of different types of crystals already known, and many more yet to be discovered.

The ability of crystals to bend flexibly has wide-ranging implications. For example, when a crystal is bent or twisted, the translational symmetry relating individual molecules to one another throughout the crystal is lost, so the sample is no longer a crystal by traditional definitions. Bending the crystal also changes its optical and magnetic properties, and our next step is to explore these changes with a view to identifying applications for new technologies.

A/Prof John McMurtrie is ARC Future Fellow in the School of Chemistry, Physics and Mechanical Engineering at Queensland University of Technology. A/Prof Jack Clegg is ARC Future Fellow in the School of Chemistry and Molecular Biosciences, The University of Queensland.