Hydraulic Fracture

15 Second Summary: We are interested in the material properties that govern the scale, extent and morphology of hydraulic fractures in brittle materials.  Our goal is to study fluid driven crack propagation in a variety of brittle hydrogels, where the time scale of these fractures is slow enough that we can study the dynamics of the fracture as well.

For details:  Will Steinhardt

Hydraulic fracturing (also known as “hydrofracking” or “fracking”) is one of the most important innovations of the last 10 years.  It is reshaping our energy landscape and in turn having major impacts on climate change, our economy and even geopolitics.  The process involves pumping a mixture of water, sand, and other chemicals into a hole drilled miles underground and pressure great enough that the water forces open and extends existing fractures and/or creates new fracture surfaces, thus greatly increasing the permeability of the rock surrounding the well and thereby its production.  In spite of the importance of this process, there are still many extremely important scientific, economic, and environmental questions that are unanswered due to the fact that these fractures cannot be directly observed.

The goal of this project is to try to understand some of the factors that control the propagation and morphology of hydraulic fracture by creating and studying them in novel tough hydrogels.  Heavily cross linked (and thus very tough) hydrogels have been shown to fracture in a way that is in many measurable ways identical to classically brittle materials like PMMA and glass [Livne et al. 2004].  Since these materials are often used to model the mechanics of geologic materials, we believe that similar research on the mechanical properties of fractures can be done in hydrogels, where there are many experimental advantages, namely that fracture propagation occurs at lower, more manageable pressures, these fractures propagate slower, and most importantly, that we can design these gels chemically or through embedding different types of index matched particles to tune the properties of the gel while still being able to image throughout it.

In our preliminary experiments, we have been able to produce two major classes of hydraulic fractures in Polyethylene Glycol (PEG) hydrogels.  First is a manner characterized by large smooth areas separated by both annular and radial lineations.  These fractures are visually similar to those seen in the classically brittle materials and normally occur at lower failure pressures (and propagate a slower speed).

Microscopy image of smooth fracture

Microscopy image showing the smooth fracture structure.  Note the clear presence of both the thicker linear features that are predominantly parallel to fracture propagation as well as the arced features that generally perpendicular to the propagation direction.

The other class of fractures is much rougher, propagates much faster, and is usually associated with violent or complex fractures.  These rough fractures are in many ways visually similar to some of the more complex fractures produced in PMMA that has been fractured by gases in a complex manner [Alpern, Marone, and Elsworth 2012].

Microscopy image of rough fracture surface

Microscopy image showing the complex structure of the rough fractures.  Note the lack of smooth areas compared to the smooth fracture image above.

Additionally we have observed dynamic transitions between these two behaviors that are an area of active research.

Movie at 10000 fps showing fracture of a brittle gel with air. There is a clear transition from slow, smooth fracture to faster, rough fracture.