Inside Furman is published monthly during the school year by the Furman University Department of Marketing and Public Relations. For story ideas, e-mail John Roberts, editor.
Rewriting Newton's Law?
Research in Physics Department is shedding new light on law of gravity
Like Einstein's Theory of Relativity, Newton's Law of Gravity is one of those fundamental, bedrock laws of science.
High school science students recite the formula from memory. And most school children are familiar with the story of how Newton developed the theory. It came as an epiphany after an apple fell from a tree and struck the young mathematician on the head.
History, however, tells a somewhat different story.
During parts of 1665 and 1666, Trinity College in Cambridge where Newton was studying, was closed due to the plague. Newton put these months of idleness to good use. Returning to his family farm, the 23-year-old developed the law of gravity (probably after seeing an apple fall from a tree). Greater than expected
Newton 's Law, which seemed to explain the orbit of planets, went mostly unchallenged for the next three centuries until scientists – charting the star movement in far off galaxies – postulated that conventional laws gravity did not apply there. Star velocity (the speed at which a star moves through space) seemed to disprove Newton's Law.
In 1983, Moti Milgrom, an Israeli scientist, modified Newton 's Law to account for the odd star velocity observations in galaxies. The new formula, called MOND (MOdification of Newtonian Dynamics), has been gaining interest since.
But testing the new formula proved difficult. Doing so would require tracking the movement of thousands of stars over thousands of years.
That's where Wade Shepherd and Bill Baker come in.
Earlier this summer, Shepherd, who works in Furman's Department of Computing and Information Services, and Baker, a professor of physics, hypothesized that 16 new computers in a Department of Physics laboratory could be networked using gridMathematica, a recently released software. By harnessing the collective calculating power of the computers, they thought they might be able to develop a program that could simulate the evolution of a star cluster (thousands of stars that are bound by gravity) over thousands of years.
And by creating these star cluster simulations using both Newton's Law of Gravity and MOND, Shepherd, Baker and two physics majors, Joel Olive and John Brady, believed they could help settle a decades-old debate: Does MOND work?
Creating such a simulation, explains Baker, requires billions of calculations, enough to overwhelm many computers.
“These machines talk to one another and divide the problem into manageable parts,” says Baker, adding that it would take one computer 30 hours to complete a simulation of a 1,000-star star cluster. The “parallel computer cluster” can complete the same simulation in three to four minutes. “We are talking about massive amounts of data and calculations.”
When the computers are not being used to create star cluster simulations, they are used for more traditional laboratory activities, says Baker.
“At very little cost and by using existing resources, we will be able to create a program that can be used to carry out large-scale simulations in many areas, such as classical and quantum mechanics, structural and dynamic modeling, materials science and astrophysics,” says Baker. “This is groundbreaking research.”
Last month, the “Simulation of Modified Newtonian Dynamics” project received the InnoVision 2003 “Innovation in Education Award” during an awards ceremony. More than 500 attended the dinner, which was hosted by the Palmetto Expo Center in Greenville.
Baker presented a poster about his group's findings to an enthusiastic audience at a November 7-8 meeting of the American Physical Society at the University of North Carolina-Wilmington. Another presentation, before members of the American Astronomical Society, is planned for January in Atlanta.
Although Baker and Shepherd have completed dozens of simulations, they are working to fine-tune the program and hope to add more computers and develop simulations on a much larger scale.
But what do the preliminary results show? Will Newton's Law of Gravity be rewritten? Does MOND have merit?
“Our initial results show a significant difference between simulations of star clusters that use Newton's Law of Gravity and MOND. In the MOND case the mean velocity of the stars in the clusters is about 75 percent greater,” says Baker. “While the clusters evolve in essentially the same manner up to core collapse, afterwards, during evaporation, the velocity difference becomes very apparent.”
Translation: More research must be done.
