ÐÓ°ÉÔ­´´

DRIVING south on Interstate 280 from San Francisco towards Silicon Valley, you soon pass a long, low structure stretching off for nearly 3 kilometres into the foothills of the Santa Cruz Mountains. Visible in satellite photographs of the Bay Area, this curious building houses a machine that accelerates electrons to feed a vast particle collider at the end of the line. But not all experiments probing the nature of matter at the Stanford Linear Accelerator Center are on such a grand scale.

Tucked away in a small laboratory at SLAC is an experiment less than 2 metres tall. It’s a modern-day version of a famous experiment carried out in 1909 by American physicist Robert Millikan to measure the electric charge on tiny oil drops. Millikan’s original research showed that electric charge always comes in indivisible chunks, the smallest of which is the charge carried by an electron. But my group and I believe our experiment at SLAC could reveal an arcane family of particles that possess a mere fraction of an electron’s charge.

No one has ever seen particles with a fractional electric charge and most scientists deny they can exist on their own. The only elementary particles that physicists can isolate and study are electrically neutral, such as photons and neutrinos, or have the same size of electric charge as the electron. These include the positron, the antimatter partner of the electron, as well as its heavier relations, the muon and tau.

However, there is no reason why there shouldn’t be fractional charges. No accepted theory in physics objects to their existence and we believe that the searches carried out so far may not have been sensitive enough to detect them. In fact, physicists believe the quarks that make up protons and neutrons have ¹/₃ or ²/₃ of an electron’s charge. And although fractional-charge experiments take particular care in looking for these values, no evidence for lone quarks has surfaced. Instead they always appear bound together in combinations to form particles that have either no electric charge or a multiple of the electron’s charge. This uncertainty both nags at and sings a seductive song to physicists.

There are good reasons to heed this song. Not only would the discovery of stable particles with a fractional electric charge open a new field in particle physics, it would improve our theoretical models, which have often skirted around the issue of fractional charge. Such a finding might also answer some questions about the early universe. For example, if we discovered single quarks it would substantially change our understanding of the nucleus and the strong force. One day we might even find a technological use for fractional charges.

My team and I began the hunt about 10 years ago. And over the years we have had to build ever more sensitive equipment. But in many ways, the first search for fractional charge particles started with Millikan’s experiment. To establish the smallest chunk of electric charge, Millikan peered through a microscope at a fine mist of charged oil droplets falling through air under the influence of gravity. Using a stopwatch, he timed individual drops as they fell between two metal plates and worked out their downward terminal velocity. Keeping his eye on the same drop, he switched on an electric field and watched the droplet rise towards a charged plate. To calculate the charge on the drop, he simply compared its upward and downward terminal velocities, the difference being proportional to the charge.

Millikan and his colleagues managed to measure hundreds of drops, but their approach was painstakingly slow: it took nearly an hour to prepare and select a suitable drop. Worse still, the measurements were made by hand and might have been prone to bias. So if signs of fractional electric charge had shown up, they might have been dismissed or even exaggerated.

All this makes Millikan’s original approach unsuitable for searching for something as rare and mysterious as fractional electric charge. We needed to adapt the experiment so that it could analyse over 100 million oil drops and measure the charge of each in a fraction of a second. Twenty years ago, this would have been no more than a dream. But in the last decade, low-cost equipment from the personal computer industry has made our long-odds experiment possible. For example, to generate drops quickly and with the same diameter (6 – 25 micrometres), we use micro-machined ejectors based on ink-jet printer technology. And we have replaced Millikan’s tedious observation and hand-recording of the drops with an automated imaging system that uses CCD cameras connected to powerful computers. Even the data we generate, which runs to gigabytes per week, would have cost a fortune to store and analyse before the advent of cheap hard drives, CD-ROMs and DVDs.

To make our measurements as accurate as possible, we wanted to measure each drop several times. This meant slowing the drops right down as they fell through the air. To do this we used a simple trick: we blew a steady stream of air upwards to cushion their fall. This involved rearranging the metal plates so they were vertical rather than horizontal as in Millikan’s original set-up. If we hadn’t done this, we would have had to punch holes in the metal plates to let the air through. This would have distorted the electric field acting on the drops, pulling them faster in certain directions and spoiling our calculations.

In our new set-up, we alternate the electric field between the plates so that as the drops fall, they wiggle from side to side. To calculate the charge, we measure the drops’ terminal velocity in both directions and work out the difference, which is related to the charge. This has allowed my colleagues – Valerie Halyo, Peter Kim, Eric Lee, Irwin Lee and Howard Rogers – and I to pin down the charge of each drop to an accuracy of about ¹/₃₀ of an electron charge.

So far we have measured 64 million droplets of silicone oil and found no evidence of fractional charge. In total we have searched 90 milligrams of oil, which means there is less than one particle with a fractional charge for every ten thousand billion billion neutrons and protons. That is, fractional charge particles are extremely rare, at least in silicone oil.

But what about in other materials? While other groups in the US, Italy and the UK have looked for fractional charges in niobium, mercury and iron, they have hunted through a mere 16 milligrams of matter. Researchers have also hunted through cosmic rays and among the particle debris produced in accelerators. So far they have found nothing, but the results indicate that any fractional charge particle must be at least a hundred times heavier than a proton – otherwise accelerators would have found one.

With all these failures, maybe we should have given up on fractional charge particles several years ago. But we didn’t quit. We couldn’t help wondering if the previous searches had been looking in the wrong place. Oil, iron and niobium are all processed materials, and the fractional charge particles may have been lost during processing. We reasoned it would be better to look in a raw material that had not been refined chemically or by other means.

We chose to look in meteoritic material from asteroids. If lone quarks or any other isolated fractional charges were created in the violent birth of the universe, they could have been incorporated into the planets and asteroids when the solar system formed 4.6 billion years ago.

Geological and chemical processes such as melting would have filtered these particles out of the Earth’s surface, wiping out any trace of them on the planet. However, many of the meteorites landing on Earth were formed at the same time as the solar system and haven’t undergone any refining. We believe meteorites called carbonaceous chondrites might contain fractional charge particles because they contain some of the most pristine material in the solar system. In particular we have been using samples from the famous Allende meteorite that fell in Mexico in 1969. Samples of Allende are plentiful and cheap, costing just a few dollars per milligram.

This isn’t the first search for fractional charges in meteorites. In the early 1980s, Gareth Jones of Imperial College in London and his colleagues searched through samples of iron-nickel and stony meteorites and found nothing. But we think carbonaceous chondrites might be a better place to look because they are good electrical insulators. This means they might hold a fractional charge more securely in their mineral matrix than a good conductor such as iron.

A few years ago we began developing a way to suspend powdered meteorite in mineral oil and then make small droplets of the suspension. We grind the meteorite so that most of the particles are less than one micrometre across and mix the powder with oil. To that we add surfactant chemicals, which help keep the powder in suspension. In tests we found we were able to make drops about 25 micrometres in diameter and containing 2 to 4 per cent meteorite. If any droplets contained stray fractional charges, they would then show up in our measurements.

Things seemed to be going well, but then we hit problems. First we found that the oil-meteorite drops carry charges hundreds and thousands of times the charge of a single electron – far greater than the charge on pure silicone oil droplets, which somehow avoid becoming ionised due to friction as they pass through air. This is a problem because it makes it far harder to pin down the accuracy of the measured charge on the drops. The total charge affects the precision with which we can measure the terminal velocity of the drop, and this uncertainty feeds into our calculation. Instead it is far better to use droplets carrying up to just 10 times the charge of an electron. We solved this by passing the droplets through highly ionised air, which neutralises excess ions in the oil drops and lowers their charge.

A much more serious problem was that making uniform droplets from oil-meteorite suspension every day is much more difficult than making uniform droplets of pure oil. The larger particles in the meteorite powder gradually settle above the hole through which the drops are sprayed. Eventually the hole becomes completely clogged and the spray stops working. To get round this we have to remix the suspension and adjust the spray – it is quick to fix but an annoying interruption. Once again we thought of quitting.

However, the concept of fractional charge particles continued to seduce us and we have persevered. We have found that with patience we can operate the drop generator and collect data for several days at a time. What’s more, we are designing a feedback system that senses changes in the drops’ properties and adjusts the operating conditions to match.

So far we have searched through less than a milligram of the carbonaceous chondrite meteorite without any sign of fractional charges, but our goal is to study several milligrams. We even dream of searching through 100 milligrams of meteorite, but we don’t yet know how to accomplish this. Perhaps this dream may have to be left to the next generation of physicists who heed the seductive song of fractional charge particles.

At present, to my knowledge, we are the only researchers looking for fractional charge particles in matter. Those people who used to hunt for them now work in other areas. That’s because most experiments today look for things theorists say ought to exist. No theories call for, or even hinge on, the existence of fractional charge particles roaming freely around the universe. But we believe the scientific pay-off of finding something totally new and unexpected is worth it.