Morphology of Fluid Flow Channels in Reactive Porous Media

Peter Kelemen, Geology & Geophysics
Jack Whitehead, Physical Oceanography

DOEI Project Funded: 2002

Proposed Research

Fluid flow channels form by reactions between migrating fluid and solid, porous media. Determining the factors that govern the spatial organization of focused porous flow is essential to understanding melt transport in the upper mantle, localization of volcanic eruptions, localization of seafloor hydrothermal vents, gas hydrate formation and dissolution along sedimented continental margins, and groundwater flow through calcareous sediments. However, the phenomena that control formation of channels have only recently become computationally tractable.

We will study the spatial distribution of fluid flow channels formed by reactive transport of fluid via laboratory experiments, mathematical analysis, and numerical modeling. Our goal is to better understand how the spatial distribution of channels formed by flow along a gradient in which the solubility of the solid increases changes as small channels coalesce into larger conduits, thereby determining the ultimate scale of flow localization.

Final Report

What were the primary questions you were trying to address with this research? (Or, if more appropriate, was there a hypothesis or theory that you were trying to prove or disprove?)
The primary goal is to learn how flow of magma is focused as it moves along, and whether drainage channels can branch out into more channels again.

What have you discovered or learned that you didn't know before you started this work?
We learned a little more about how to set these experiments up. Drainage channels have been produced, but the branching of channels has not been found at all.

What is the significance of your findings for others working in this field of inquiry and for the broader scientific community?
Nothing yet, the project was only intended to start this work.

What is the significance of this research for society?
Ultimately to know volcano mechanics better, to understand how magma interacts with host rock as it flows along.

What were the most unusual or unexpected results and opportunities in this investigation?
The biggest surprise to us is how hard it is to get flows to focus into channels when we want that to happen, and yet we know experiments where the channelization is easy to make. The branching of channels, which is seen in geological settings, is not found in our laboratory experiments.

What were the greatest challenges and difficulties?
Working with saturated salt solutions. They corrode metals easily, are tough on one’s skin, and their properties are not well quantified.

When and where was this investigation conducted?
The purpose was to design and begin to run new experiments in the Geophysical Fluid Dynamics Laboratory.

What were the key tools or instruments you used to conduct this research?
Densiometer, thermometers, temperature baths, custom made containers and lots of mathematics and searching out properties of materials.

Is this research part of a larger project or program?
Conceptually yes, but the projects are not funded.

What are your next steps?
One of us has started using a wax-like polyethylene glycol instead of saturated salt solutions and in this one channels are easy to make.

Have you published findings or web pages related to this research?
No, except Braun’s thesis does have discussion of the experiments, but the experiments enrich his understanding. There will be no publication of these results in their own right, this project is not able to support that level of scientific achievement.

Please suggest or provide photographs, illustrations, tables/charts, and web links that can help illustrate your research.
Report below.

Please provide some biographical information, such as place of birth, degrees earned, significant awards or honors, research interests, reasons why you became a scientist or why you are interested in this line of research, and any personal interests, hobbies, or details that you are willing to share. You can find a few good examples here:
Peter Kelemen is a field geologist and petrologist interested in how rocks are formed during magma flow. He just received the Bowen Award from the Volcanology, Geochemistry, and Petrology Section of the American Geophysical Union. After extensive field studies, he became interested in working with Jack Whitehead who is a fluid dynamicist and has developed more than forty new laboratory experiments in his career. They have jointly worked toward developing fluid mechanics studies to study melt-rock interaction starting about ten years ago, This project teaches about the effects of crystallization on flow focusing using saturated salt solutions. Mike Braun conducted extensive studies of magma drainage mechanisms for his thesis work and the preliminary experiments done here by Mike and other students in the geldynamics program produced new refined experiments.

RESEARCH SUMMARY:
A number of laboratory experiments were constructed in which water flowed through salt crystals and caused the crystals either to grow due to the water reaching saturation temperature, or to dissolve if the water was below saturation. We sought to view spatial organization of drainage channels in the salt matrix. A number of geological applications for flow focusing exist in magma systems, including processes leading to dunites in Ophiolites (ancient ocean floors that have been thrust up on land), flow focusing under spreading centers that leads to dunites, channelization of magma in surface volcanic systems, and erosion within limestone. The salt used was Ammonium Chloride, which has a solubility coefficient in water that is very dependent on temperature. A few experiments were conducted with syrup rather than salt solution. In the project a number of experiments were conducted by students, especially by Mike Braun in conjunctions with interests he developed in his thesis work. Funds were used principally to construct the equipment and purchase the materials. A small amount of funds were used to design the experiments and help with data analysis. The projects are summarized below: The reports are in the form of Geodynamics project reports or informal notes. At present, there are no plans for publications, although some of the results might be in future papers.

Lisa Lassner: Spring 2003. Used small 16cm x 16 cm x 16 cm tank for crystal growth experiment related to ledge formation on hydrothermal vents. We used supersaturated KNO3 liquid in the tank. A heater/cooler unit with a 50/50 antifreeze mix supplied cold fluid of about - 7 degrees C to a 1/2 cm diameter stainless steel tube placed horizontally in the tank. Crystals then grew on the tube presenting ledge-like features. (Photo of ledge is in the Gizmos' section of our website.)

Christina Kaba: Spring 2003. Built a flat tray about 40 cm square. One end of the tray was constructed of 1/2" aluminum about 15cm wide, the remainder of the tray was acrylic. Beneath the aluminum section was a chamber connected to a heater cooler unit that supplied cold water at about 2 degrees C , which cooled the aluminum. A thin acrylic cover was placed above the entire tray and spaced above it using small glass beads. Supersaturated NH4CL was supplied to a manifold and uniformly distributed it in a line above the aluminum plate. The super-saturated liquid hit the cold aluminum, crystallized and eventually formed channels through the crystallized salt. The glass beads provided channels to the non-cooled sections of the tray and other channels developed.

Mike Braun: (I) summer 2002: Built 60cm (l) x 20 cm (h) by 1/2" (w) of thick acrylic. At the top and bottom were 1/2" diameter stainless steel tubes that ran the length of the tank. Heating and cooling units supplied water to the tubes at fixed temperatures. A saturated sodium nitrate solution filled the tank. The lower stainless tube was 5 deg. C and the upper tube was 40 deg. C. The top fluid was in contact with solid crystals and became saturated and hence dense. It sank to the bottom where rising cold liquid would not be saturated upon rising and getting warmer and it would form dissolution channels through the salt. These eventually form mineature vents with rising lower salinity liquid emerging from their tops as shown in shadow graphs. (Photo is available on our website in photo gallery 'experiments'.)

Mike Braun: (II) Spring 2003: An expansion of his first experiment. Used supersaturated NH4Cl rather than Sodium Nitrate. Built much more complicate tank with a 1' thick acrylic test section to insulate the fluid. A bottom tank about 78 cm (l) x 14cm (h) x 13cm (w) was a holding tank for the liquid and was heated by a 1/2' stainless steel tube. The NH4Cl was continuously pumped into this tank. Above it was placed the test chamber about 76cm(l) x 27 cm(h) x 1/2" (w). Four stainless steel 1/2" tubes (two at the top and two at the bottom) were recessed into the 1" thick sidewalls. They ran horizontally and supplied either warm or cool water from two heater coolers. The test chamber was filled with glass spheres to aid in channel development from the crystallizing salts. Finally, above the test chamber was a long tank that caught the rising fluid and recycled it to the supply system. (Photo in Gizmo section of website.)

Mike ran short of time with experiment II. He had planned to study the crystallization process with different flow rates and concentrations. He wanted to observe the change in liquid composition by sampling the supply fluid before it was cooled and its concentration as it left the system to be recycled. He had hoped to observe the flow channels in detail.