CONTROLS ON THE PRIMARY POROSITIES OF  BRECCIAS

GEOL 2500

Harmon D. Maher Jr.

Dept. of Geography and Geology, University of Nebraska at Omaha

 

Nature of this project

The context for these web pages is that of a one-credit undergraduate research project for students at the University of Nebraska at Omaha. There are two primary motivations behind this project: 1) to learn about the scientific process through experience, and 2) to learn about what controls primary breccia’s porosity. A view of science that I like in particular is that it is a sophisticated form of play.  So we are going to play with collections of broken up material, with a focus on the space between the pieces. The percentage of space between the pieces is known as intergrain porosity. The puzzle is what determines how much. In nature it varies greatly.

There is some simple satisfaction of puzzle solving, without real regard to any utility. Yet, the skills learned in aimless play can be useful when efforts switch to more purposeful efforts. In addition, when novelty is highly valued in play, serendipity produces unintended useful discoveries. I hope as we explore this particular puzzle, the participants will find it particularly rich and rewarding. This puzzle is also particularly accessible to the uninitiated (i.e. not much background is required).  Below is some preparation material that is intended to be useful as introduction and preparation for this project.

Definition for breccia

Breccias are collections of angular rock fragments visible to the human eye, where the fragments have moved and/or rotated some significant distance with respect to each other.  Additionally, the clasts are typically in contact with each other, instead of floating in some type of matrix. A basic assumption is that the rock fragments (clasts) are angular because some continuous and original material got broken up.

What types of breccias exist geologically?

Link to images of different breccias.

Fault breccias: As two blocks of rocks slide past each other, wear and tear, abrasion, produces angular debris. A great variety of factors determine what that fault breccias looks like, including: speed or movement, geometry of the slip surface (roughness – think sandpaper), the character of the rocks involved, the presence of water, and the stress field involved.

Hydrothermal (dilational) breccias: hot fluids under high pressure aid in opening up small rock flaws into larger cracks. Typically they also deposit mineral matter in the cracks. Repeated periods of such cracking can produce a breccias. This requires the creation of lots of new space – hence the term dilational. 

Karst breccias: Some rocks, such as carbonates in general, and limestone in particular are relatively soluble.  These might also be generally called dissolution breccias. This produces blocks of rock in several ways. First, solution can produce large cavities whose roof and sides are unstable.  Remembering that typically rocks come prefractured, pieces of the wall and roof fall off and collect within the cavity.  In addition, if insoluble stuff is interlayered or dispersed within soluble material, then the solution of the soluble material removes the support leaving the rubble. Again the prefractured character of the material is important. With the support removed the material collects on the solution cavity (cave) floor. Here it can be mobilized by underground stream flow or by debris flows. Chert nodules are a common and relatively insoluble material found in limestones that often ends up as clasts in karst breccias. Gypsum, which is even more soluble than limestone can also produce good breccias, often of carbonate clasts.

Salt dome breccias: Salt, halite, is very weak, and indeed will mobilize to form columns that rise and pierce the overlying sedimentary layers. As they do so more brittle layers within the salt dome get broken apart by the solid-state flow of the salt. In addition, at the sides of the salt dome, the insoluble country rocks get broken and plucked by the flowing salt to form a breccia with a matrix of salt. As salt is also very soluble, at higher levels where the rising salt dome encounters fresher water in the surrounding rocks, the salt can be dissolved. Ironically, here it is the ductile deformation of one rock type that breaks apart another stronger rock and contributes to the formation of a breccias.

Volcanic breccias: One of the most spectacular geologic events is an explosive volcanic eruption. Rock is highly fragmented, and mobilized, resulting a great variety of products, including volcanic breccias, also called volcanic agglomerates. Some of these breccias have good porosity, and as hot fluids from the volcanic pile migrate through they often deposit minerals within that porosity.

Intrusive breccias: It is not uncommon to find angular blocks inside intrusive rocks such as granite. These are called xenoliths, and represent pieces of country rock that the magma has moved through and intruded. Magma under pressure can pry open cracks. Less common is to find such an abundance of xenoliths, that they are touching, and can be considered an intrusive breccias. One possible origin of such breccias is that heavier xenoliths sink and collect at the bottom of a magma chamber. Another is that a magma conduit or reservoir gets ‘pressed’, the magma squeezed out elsewhere, leaving only the xenoliths and trapped magma between the clasts.

Igneous breccias pipes: These are of great interest because one type, Kimberlite pipes, can contain diamonds. The diamonds and some associated material were carried upward all the way from the mantle. However, the mechanics of these pipes are poorly understood (at least by me at this point). They consist of highly brecciated, and often mineralized and altered material. They are usually attributed to a ‘jet’ of high pressure volcanic gases that rise quickly through the overlying crust, brecciating it as they do so.

Impact breccias: OK, come to think of it, an impact is an even more spectacular geologic event than an explosive eruption. As the crater is created, the pulverized rock is thrown clear, forming an apron of debris. Naturally the big chunks stay closer, and the smaller chunks get flung farther. Close to the crater, surge deposits, masses of fragments flowing in a coherent matter, but hugging the ground can form. That angular fragments can flow as a semicoherent aggregate is perhaps at first counterintuitive, but think of sand flowing in an hour glass as providing some initial insight into the possibility. The fragments can bear marks of their formation by shock waves from the impact.

Colluvial breccias: Breccias form at the base of cliffs or scarps of varying types by mass wasting processes. Remembering that rocks at the surface are already fractured, it is no surprise that as water in a crack freezes it can pry a rock loose. The loose rock falls and collects at the cliff base as talus. I remember well such a collection from my childhood at Bear Mountain in New York, because in places the blocks were so big that the resultant porosity could be climbed through as a cave. The variety of processes that contribute to these colluvial deposits of debris that flank slopes are actually varied, and in many places include debris flows, which are water saturated very thick slurries of angular slope debris that snake their way down hill. Another, unigue example is the collection of debris that collects on the unprotected, deep water side of a fringing reef. Storm events often play a critical role in producing these reef breccias or talus can end up being particularly porous, and are a well known type of petroleum reservoir/trap. A special variant of these type of breccias deposits are avalanche deposits, especially very large avalanches that can have counter intuitive runout distances. The Huascaran avalanche that buried a village in the Andes is an excellent, but tragic example.

Quarry breccias (aggregate): These are artificial breccias, formed by crushing machines. What process is used to produce the optimal product is an interesting question. Something may be learned from the literature on aggregate production.

Summary on breccias: A great variety of breccias exist, mainly defined on the basis of  how and where they were formed. Breccias can also be polygenetic, hybrid. For example, hydrothermal brecciation can occur along faults, and at the earth’s surface fault breccias can become fault scarp talus material as one side gets lifted.  One natural question that arises – how do you tell one breccia type from another? The type of clasts, the nature of matrix between the clasts, the shape of the breccias body, and the geologic context are all important clues as to origin.

Link to images of different breccias.

Why are breccias of interest?

Of course to some breccias are of little interest. Breccias are of academic interest simply because they inform about the geologic history of the area they occur in. They are indicative so some distinct event, and are information rich. However, as you might guess from the above array of possibilities, it can be challenging unraveling their origin.

Breccias are often of economic interest because they often have significant porosity/permeability which can host water, oil, gas, and ore mineralization. In this case being able to unravel the breccia’s origin and extent is of distinct interest.  A key trait of the breccias in this case is their porosity, and the permeability. Breccia porosity can be surprisingly variable and complex. That is because of the multiple variables involved: including clast shapes and size distributions, the way the clasts are arranged (packed), the degree of infilling between the clasts (by mineralization or finer sediment), or of solution of the clasts and/or matrix (producing secondary porosity). What controls breccia’s primary porosity is the focus of this research project. One can view this as a very challenging geometric puzzle, especially in contrast to the well worked out and more elemental puzzle of what controls the porosity of more rounded sedimentary grains.

Brecciated or granular material can have useful or significant material properties, foremost of which might be high surface area. What are the optimal shapes and packing to keep the porosity high enough so that fluids can migrate through while maximizing surface area to promote the reaction? Another useful property of breccias can be their abrasive and frictional characteristics. Think of grit and abrasion. Clearly clast hardness is a factor in the ability to abrade, but clast shape and size distribution can be expected to play a role also. What might be the best material for a non-skid surface?  What is learned in one endeavor can be useful in another.

Do breccia shapes reflect origin?

Given that breccias are collections of broken pieces, another way of phrasing this question – what are the various ways something can be broken up? We can approach this question from a theoretical or reasoned perspective, or from an observational and/or experimentally perspective. We will discuss the first, and explore the second in great detail.

There are different ways to break something up: explosion, implosion (geologically rare), crushing, or shear. More fundamentally, fractures are of two types – tensile (mode 1)  and shear planes (mode 2). In nature hybrids also occur. Tensile mode fractures are often at right angles to each other, and shear fractures are typically at 60 degrees from each other. These patterns are very commonly seen in regular arrays of approximately planar fractures seen in rocks which are called joints.

As important as how you break it is what you break. An important thing to remember is that most geologic materials are layered, anisotropic, and this influences very strongly how they break up. Think of layered sediment, or slate. One might imagine that if you start with a planar fabric, layered sediments for example, that one produces rectangular blocks, where as something more homogenous (isotropic) produces more pyramidal.

To add another element, the direction of the ‘breaking forces’ relative to any layering can be important.  Think of hammering on layered rocks. Trying to collect rock specimens can teach one something about the importance of layering (material anisotropy) in fragmentation. Hammering sub-perpendicular to the layering can result in a big block, whereas hammering parallel often leads to smaller pieces of different shape. This can help one understand why you can get all sorts of clast shapes. Theory provides some very good guidelines as to what should happen in different situations (although we likely won’t explore this aspect very much in this project).

The fracturing may be a separate event from that which produces the breccia. The rocks typically come pre-fractured. The above mentioned joints are very common features. Ductile flowage can cause brittle deformation of more competent material. This is seen in deformed gypsums with internally deformed limestone layers. If the relatively soluble gypsum gets dissolved, the more insoluble broken residue is left.  Dissolving limestone can result in a residue of broken up chert. In this case the clast geometry and size is a function of the initial fracturing process.

Another aspect to consider is clast modification. For example, clast transport or continued shearing may tend to take off corners, leading to clasts that are less angular, more rounded.

How do you characterize angular clast geometry?

This is a rich and difficult question. One might ask why one would want to characterize clast geometry. Scientist measure things so they can compare samples, describe the character of populations, establish relationships between different measures, predict, and model behavior. The language one use is mathematics. In this case we want to see how clast shape, size and shape-size distribution in breccias influences porosity.

For clasts that are approximately spherical it is easy, the radius capture quite a bit. For an ellipsoid it is a bit trickier – one needs to measure the long, short and intermediate axis.  Volume can be computed from these, and ratios of the axis can capture shape. The aspect ratio for example is simply the long divided by the short axis. How about for a ‘blocky’ clast? A rectangular block could be characterized in a similar way to an ellipsoid. However, for a more irregular block it is obviously more difficult.  Ideally, the characterization should focus on shape and/or size and provide possible insight into how the clasts were generated.

One could simply count the number of fracture faces. Four would appear to be the minimum for a pyramidal shape (unless the faces are curved). 6 might be expected to be common (tabular blocks). 5 could reflect triangular blocks with parallel faces. A clast with many faces  would be more rounded.  A consideration would be whether the faces are approximately planar, are more curved, or even more complex in its geometry. For certain materials such as obsidian, curved (or conchoidal) fracture faces are common.

Another possibility could be aggregate internal face angle amount for a cross section. Working with cross sections can be very advantageous, since one can easily slab breccias samples for inspection. One needs to consider how the cross section view may be biased, and we will discuss that later.  What will the number of face angles reflect. A triangular sections would yield 180, parallelograms 360, and as something gets more fracture faces, it can approach infinity. The higher the number of faces and the aggregate angle, the closer an approximation to a sphere.  Indentations may be a problem, but one can always take the smaller angle. Try to visualize cutting through a given shape block in different directions, and what the resulting shapes in cross section would be.

Industry has developed one answer where the ratio of circumference of a circle which has the equivalent cross sectional area is divided by the actual clast circumference of a clast with that cross sectional area. Details may be found at the website http://www.azom.com/details.asp?ArticleID=3882.

This may or may not be one of the approaches we take. It is crucial to remember that this endeavor is open-ended. It is an exploration, and we are not sure exactly where we will end up.

How do you characterize a population of angular clasts?

Obviously there will be variation in the shapes and sizes of breccias clasts, and that variation will be crucial to breccia porosity. For example, one could easily expect that if there was a wide range of breccias clast sizes, that the smaller pieces could fit inbetween the bigger, thus reducing the porosity. On the other hand if the clasts are all about the same size then porosity would be greater. Variation in shape may also be important.

If you can reduce the clast characterization down to one number, then sampling a breccias clast population, and constructing histograms can provide a valuable first step. For example, a histogram of aggregate internal face angles could be quite insightful. Is it unimodal or polymodal, symmetric or asymmetric in distribution, what is the average and the  standard deviation are all that questions that can be answered by analyzing the distribution. That prepares the way for looking at such questions such as – does a greater variability in clast shape tend to increase or decrease porosity.

What is a fractal distribution of sizes?

It is already established that fractured and fragmented material typically has a fractal distribution (e.g. Turcotte, ).  What does that mean? Most succinctly it means that if one plots clast size on the x axis, and the frequency of clasts of that size or greater on the y axis, where both axes are log scale (1, 10, 100, 1000 etc.) that there is a straight line relationship. This succinct statement probably does not provide much insight unless you are already familiar with fractals. Another way of describing fractals is that there is a continuous distribution from gobs of little sized pieces, to lots of medium sized pieces, to a few big pieces. Characteristic of fractals is that the patterns look the same at large scale, as an intermediate scale, as at small scale. Theoretically the pattern looks similar every time you look at an even smaller scale. This translates into not being able to tell even approximately how large an element in the fractal pattern is. However, in my experience words just don’t suffice, and the best way to learn about fractals is to see them, to construct them, and to measure and plot them up.  This is one direction that we can take.

Given that fragmented material has a fractal distribution, breccias clast distribution should also have such a distribution. Of course, different from the mathematical relationship, real fractal distributions have a size range limit.  You can’t form a talus clast larger than a cliff. Once you get down to the crystal size of a the rock being broken up, the way the crystal breaks may be quite different than the way the rock breaks and so one wouldn’t expect the fractal distribution to continue at smaller scales. For example, if you imagine a cliff 100 m high of coarse grained granite, you might expect the range over which talus clasts might follow a fractal relationship to be from 100 m to a couple of centimeters.

Whether or not we pursue this aspect of breccias will depend on where our discussions and interest take us. There is a fair bit of literature that explores the fractal character of fragmentation, that we should tap into if we go this direction.

How do you  make an experimental breccia?

Here creativity will be particularly useful in coming up with materials to use, and experimental designs to set up. It might be helpful to think on idealized versus natural experimental breccias.
Idealized breccias would be where the shape and/or size distribution is carefully controlled. This would be useful if one wanted to explore the effect of a certain variable on porosity. For example, one might use rectangular blocks of different aspect ratios to see what influence aspect ratio has. What would you predict – would collections of squarish blocks or ‘bricks’ tend to have greater porosity and why?  One could also use an experimental design where two end member idealized populations are mixed to different proportions to see how porosity changes as a function of the mixing. Imagine an aggregate of dominoes and dice. One could predict the maximum porosity  at a 50-50 mix if clast shape variability and shape mismatch. Obviously these idealized breccias are quite artificial, but one may still be able to develop useful generalizations that would apply to natural breccias.

If one could find a natural breccia where the pieces are not glued together (by cementation and mineralization), then simple collection of truly natural (field) samples could be done. These natural aggregates could be used in experimental set-ups. One easy possibility for this is the collection of talus slope breccias. We now have three such samples, and hope to acquire more in the future. For practical reasons (like moving the sample) one would have to sample material that produces smaller clasts. In doing so it is important to realize that one may be introducing a bias. Also, there are plenty of geologic examples of where properties are a function of measurement scale, and this (scaling properties) would need to be considered.

However, most breccias are cemented together in some fashion. In some rare cases one might be able to unglue a breccia by dissolving the cement, but here is the additional challenge for our particular project of traveling to collect the breccias. While we will keep open the possibility of working with field samples, the plan at this time is to make our own breccias from scratch. The materials we use will require some thought. The ability to easily break and saw through the material may be advantageous, and so plaster could be one type of material to work with. Unfortunately plaster may become mush when exposed to water, but one could use wax, glycerin or oil to measure the porosity.  As we will see wax may have some particular utility.

It is not only the clast population are working with, but how they are assembled that is crucial. Air fall is one possibility, where the  clasts are simply dropped in place. This could mimic volcanic or ejecta or karst (roof fall) breccias. However, even something as simple as dropping it in place can have complexities. Is it dropped all at once, or a few at a time? Does any sorting take place (e.g. with distance from an impact or eruption)

Another possibility is that the clasts slide, or roll into place. This might be expected when mass wasting down a slope is involved. Replicating this in the lab so that the porosity can be measured may call for some creativity.

Debris flows are common ways breccias are deposited inside caves, in association with volcanic eruptions, and in association with slopes. These aggregates that move down slope can have complex internal processes that produce sorting and grading of clasts. One interesting observation from Svalbard is that muddy debris flows with large talus blocks in it are modified by subsequent rain, and water flow so that much of the mud is washed out, leaving only the larger aggregate. Again replicating debris flows so that we can measure the resulting porosity will be a challenge.
One last consideration is post-deposition modification. Depending on the strength of the clasts, the weight of overlying material material can crush grains and cause grains to slide past each other into a new position and compact the material. Typically this does not seem to be a major factor.   As mentioned above, there are many other post-deposition that infill the original porosity (internal sedimentation, cementation, pressure solution), but we are focusing only on the primary porosity for this project.

One ‘trick’ that we might use is to pour wax into our breccia aggregate. In this way we can saw through it in order to provide cross sections, allowing us to see the internal framework.

How do you characterize the arrangement of angular clasts?

The typical description for breccia clast arrangements I am aware of the arrangement is along the lines of qualitative, descriptive assessments – a good place to start. A framework texture is where clasts are all in contact, producing a frame that could bear weight, i.e. carry stress. The clasts themselves provide clast support. Matrix-support is where large angular clasts are not in contact, but floating in a matrix of some sort. These technically are not breccias, but diamictites. This distinction between framework and matrix support is unsatisfactory when it comes to deposits with a larger range of grain sizes and a fractal distribution. In this case, depending on the exact nature of the fractal distribution, larger and rarer clasts can flow in finer grained material, but that finer-grained material is all composed of clasts, and so technically it is clast supported. Yet, such material can form a debris flow.  If the clasts are clearly inequant, they can have a preferred orientation (evident as an alignment of the long axes of the clasts).  If there is a change in average clast size upward then the deposit may be graded. It is fairly common that colluvial deposits are graded.

Are there ways to quantify breccias clast arrangement, so that comparisons or predictions can be made? To my knowledge this is a relatively unexplored field.  Can we build on an existing approach.  In working with the porosity of well sorted (all about the same size) round grains (e.g. in a sandstone), the arrangement is known as packing (how the grains are packed together).  The tighter the packing the lower the porosity. The closest packing is where the clast centers of adjacent grains form a 4 sided pyramid (a tetrahedral). We might call that tetrahedral packing. A much more open packing is cubic, where the centers form a cube.  ‘Defects’ or departures from tetrahedral would represent additional porosity. So could one learn something by looking at (quantifying) center to center angles or distances of clasts? Of course working in 3-D can be very difficult, so one can think of how a 2-D slice would ‘sample’ this 3-D geometry.

Other possibilities may exist, and you are encouraged to explore them.

References: These are not required reading, but a resource we may tap into as the course proceeds.

 

Any feedback on this page (which will be under revision) is welcome.

harmon_maher@mail.unomaha.edu