"A Rag and a Bone and a Hank of Hair"*
New Dating Technique Reaches Back 3,000,000 Years

First, these atoms could tell you approximately how old something is. Perhaps even where it originated. You could tell if a rock in Antarctica is debris from a meteor collision with the moon - or just a rock. You might follow rainwater contaminated by a nuclear bomb test as it works its way into soils and aquifers. Or discover precisely when a glacier covered a particular part of the earth's surface.

The process used to count these tell-tale atoms is the 21st century version of carbon dating, an area where Penn has long played a leading role. This new procedure, however, requires some special equipment. You need a mass spectrometer, one part of which is a particle accelerator. Particle accelerators are generally huge: CERN (in Switzerland) is a mile in diameter. Jeff Klein and Roy Middleton, the Penn physicists who are working with this process quickly point out that their machine is actually a nuclear accelerator, which is smaller and works in a range of only a few million volts. "It's also Army surplus. We got it for the price of moving it to Penn."

Accelerated mass spectrometry (AMS) separates atoms by their mass (roughly, their weight). Knowing the mass of the atoms you're analyzing should enable you to identify and count them. However, contaminants - scattered particles and other stable atoms or molecules with the same mass as the atoms you're interested in - can confuse the issue. An ordinary mass spectrometer has no way to screen out these contaminants. "That's why you need an accelerator." In several stages of progressive screening, an accelerator can get rid of contaminants.

When a sample is put into the machine, it is electrically charged and turned into fast-moving negative ions. In the accelerator, seven to eight million volts of electricity speed them up even faster. As they flash through a thin screening foil, their electrons are stripped out, changing the ions' negative charge to positive. This change destroys contaminating molecules. Large magnets then filter out more unwanted particles, until the atoms that are left can be identified by composition and counted.

AMS dating derives from and greatly improves on carbon-14 dating. The older technique assumes that all living things derive all their carbon from the atmosphere. Carbon-14 atoms (radioactive carbon isotopes) have been formed steadily at about the same rate ever since earth first had an atmosphere. A plant picks up small amounts of the isotope during the course of its life, but when the plant dies, its carbon-14 begins to decay. By comparing the concentration of the isotope in the plant when you find it with the concentration normally in the atmosphere, you can determine the plant's age. With AMS, scientists can use much smaller samples, determine sample age more rapidly, and extend the time range. The age limit with conventional carbon-14 dating is about 40,000 years. Using AMS, Klein says, "can take you back as far as three million years."

Another kind of application for AMS is found in biomedical work. Diagnosing disease, for example, often involves giving a patient a protein or carbohydrate with a radioactive tracer, then following the tracer to see where it goes and what it does in the body. Clearly, the disadvantage to this technique is the hazard posed by the radioactivity. Using AMS, however, requires such low concentrations of radioactive isotopes that the biological hazard is eliminated. Researchers can also do tracer experiments over periods of years or decades. One on-going experiment, for example, studies bone loss in osteoporosis by feeding people calcium-41 and tracing the amounts in their bodies for many years. When do people begin to lose the calcium-41 concentrated in their bones? Again, answers can be found without hazard to the human subjects.

While AMS has many applications and many advantages as an investigative technique, it does have one daunting disadvantage: it's expensive. Jeff Klein puts the budget of the Penn lab at about $500,000 per year. (That hardly makes it Big Science. CERN's budget approaches $10 million per year.) Normally, the lab's primary sources of funding are the National Science Foundation, NASA for extraterrestrial studies, and the National Institutes of Health for biomedical research. Until 18 months ago, Klein points out, Penn's accelerator was also doing nuclear structure physics. The lab then had about $1.5 million a year from the Nuclear Physics division of the NSF. However, "nuclear physics is becoming less an interest of the Foundation."

Some of these cutbacks in funding result from the natural evolution of knowledge. "You should put your money out on the frontiers of knowledge." And the perception of where those frontiers lie changes over time. "The maturing of what we know about atomic nuclei is partly responsible for the change in support, which tends to go now to particle physics and work related to cosmology." But funding also shifts because there are "good times and bad times for science." Physics tends to be especially hard hit right now because of its expense and the complication of its work, which can take years or decades to achieve results. "It's relatively esoteric, too, and often has no immediate applications flowing from the work."

Biomedical research does have obvious paybacks, of course. And geomorphic processes are intensely interesting to people who construct and test climate models. AMS can be usefully applied in many other areas, but sometimes, Klein admits, "all that you get back is knowledge." And who needs knowledge if it isn't immediately useful? Well, we all might. As the nineteenth-century physicist Michael Farraday is reported to have said, "A newborn baby isn't immediately good for anything." With a bit of investment, however, he/she could find a cure for cancer or AIDS.

*The quote is by Rudyard Kipling.