The most powerful engine on Earth is not under the hood of Alex Zinardi’s Indy car nor is it one of the space shuttle’s booster rockets. It is not man-made, but instead lies deep beneath the surface of the planet where it transports tremendous heat between the core and mantle. This colossal machine manifests itself at the surface through sea-floor spreading and plate tectonics. Now the first steps towards building a model of such an engine have been taken in a Cornell laboratory. Since the turn of the century, scientists have puzzled over the shape and placement of the continents. While the fierce debates of the first half of this century concerning the theory of plate tectonics and sea-floor spreading have long since been resolved, the search continues for a complete and accurate physical explanation.
At Cornell, a group of researchers associated with the Materials Science Center has been studying the properties of solid and molten wax layers under tension. The results of their experiments show a remarkable resemblance to plate spreading at mid-ocean ridges. This past summer, undergraduate Nathan Gemelke from the University of Wisconsin continued research done by Jeron Carr of Cornell under direction of Professor Eberhard Bodenschatz of the Laboratory of Atomic and Solid State Physics. Gemelke was a participant in the Materials Science Center’s summer research experience for undergraduates (REU), a research program funded by the National Science Foundation.
The group studied the structure and dynamics of rift formation in sheets of solid wax floating on a bed of molten wax. The experimental apparatus consisted of a metal tray filled with paraffin wax called Shellwax 158 to a depth of 8 centimeters. A bath of circulating hot water (77¡C) beneath the tray kept the wax a few degrees above its melting point. A flow of cold air (25¡C) across the surface of the melt caused the uppermost layer to solidify, forming a thin crust.
During the initial cooling stage, two steel plates froze in, becoming anchored in place as the wax solidified. These plates were used to transmit the force from a microstepping motor to the wax plates, generating enough stress to pull the plate apart slowly. In order to minimize variation from trial to trial, a straight incision was made perpendicular to the spreading direction with a sharp knife, providing a starting edge for the simulated crustal rifting. In the actual experiment, the plates were pulled apart at rates varying from less than 20 microns (1 micron equals1/10000 of a centimeter) per second to more than 100 microns per second. Fluorescent lamps positioned deep in the molten wax illuminated the experiment from below, and enabled pictures to be taken via an overhead video camera. Each trial ended once the anchor plates reached the sides of the tray.
Results from various trials show that the pattern of rift formation depends heavily on the spreading rate. At slow speeds (less than 20 microns per second), Bodenschatz’s group found that the rift remained relatively straight, exhibiting only minor variations from the rift axis. The spreading was slow enough to allow the cold air to solidify the upwelling liquid wax at the rift almost instantaneously; hence the rift was frozen over.
At slightly faster speeds (20 to 35 microns per second), faults, similar to transform faults, began to form. A transform fault is a special kind of tectonic fault where the rift itself is offset along the fault line. The displacement, however, is in a direction contrary to that expected from the relative motion along the interface. Transform faults on Earth are found near mid-ocean ridges, the large scale analog of the wax rifts. A mid-ocean ridge, such as the mid-Atlantic ridge, is an undersea mountain range formed by the upwelling of molten rock from within Earth. Transform faults split the ridge in numerous locations.
One theory attempts to explain the occurrence of transform faults at oceanic ridges by considering the spherical nature of Earth’s surface. A “straight” spreading rift must necessarily bend towards the ends of its axis to account for the changing circumference of Earth (the same problem is encountered on a conventional Mercator map of the world: the map is distorted towards the poles). Bodenschatz’s experiment, however, seems to indicate that there may be additional processes involved in the formation of transform faults, as it took place in a flat, rectangular system, transform faults being observed even in the absence of curvature.
Another interesting feature that developed at medium speeds was the creation of an overlapping spreading center (OSC). OSCs occur when two offset rift segments grow, in effect creating two separate spreading edges in a local area. These spreading centers compete against each other in the region of overlap, resulting in a rotating piece of solid wax trapped between the rift segments. In some experiments, one of these segments detached and moved away with the spreading plate, forming two new OSCs at the corners of the remaining rifts. Overlapping spreading centers have also been observed in association with rifts of Earth’s crust.
At fast spreading speeds (greater than 35 microns per second), transform faults and fracture zones dominated the rift pattern. Scientists observed a slight ridge (2 millimeters tall) at the rift axis, a situation analogous to the elevation of the seafloor close to mid-ocean ridges. As with newly created crust, recently solidified wax is still relatively hot. However, as it moves away from the spreading axis, it ages, cools, and contracts. Fracture zones are regions of steep drop-off of the crust due to such age gradients. Fracture zones occur along extensions of transform faults, where the rift is displaced. Thus, along the fracture zone, the age of the crust on one side differs from that on the other. Since both wax and Earth’s crust contract as they cool, an age gradient results in a difference in crustal volume: the older crust is denser than the younger, it rides lower on the underlying molten wax, and a “fracture” occurs. Bodenschatz’s group observed such fracture zones in fast spreading wax layers.
The results of these experiments should provide new methods for testing the theory of sea-floor spreading. According to Bodenschatz, it may soon be possible to develop more accurate models of this geological phenomenon: “We believe that this study may be a first step towards the development of a laboratory scale system which may provide the possibility for a quantitative test of the models developed for mid-ocean rift formation.” With such a tool in hand, scientists should gain a better understanding of the processes taking place hundreds and possibly even thousands of miles below Earth’s surface.
Readers who want learn more about the research of Bodenschatz’s group are invited to visit their website. A complete description of the experiment and results, as well as motion-picture videos of various stages of the experiment, can be found at the internet URL: http://milou.msc.cornell.edu/waxtectonics.html.