<?xml version="1.0"?><?xml-stylesheet type="text/xsl"  href="../template.xsl"?><!DOCTYPE nsuarticle PUBLIC "-//NPG//DTD NSU//EN" "../nsu_article.dtd"><nsuarticle type="news">   <articleidlist> 	 <articleid type="uid">010719</articleid><storyno>-2</storyno> 	 <articleid type="doi">10.1038/nsu010719</articleid><storyno>-2</storyno>   </articleidlist>   <pubfm> 	 <pubdate> 		<dayofweek name="Friday"/> 		  <day>13</day> 		  <month>July</month> 		  <year>2001</year> 	 </pubdate> 	 <category>physics</category>   </pubfm>   <fm> 	 <title>Physicists better their time</title> 	 <aug> 		<prefix></prefix> 		<fnm>Philip</fnm> 		<snm>Ball</snm> 		<suffix></suffix> 	 </aug> 	 <keywdgrp> 		<keyword>time</keyword> 	 <keyword>measure</keyword><keyword>atomic clock</keyword><keyword>metrology</keyword><keyword>atom optics</keyword></keywdgrp> 	 <standfirst>New atomic timepiece clocks up more ticks.</standfirst>   </fm>   <body> 	 <p>In the most impressive feat of time management ever, scientists in Colorado have made a clock potentially so accurate that, had it started when the Earth was formed four and a half billion years ago, it might by now have slipped only a second or so. </p><p>Substantially more accurate than the best existing clocks, this development could allow physicists to measure the fundamental constants of the Universe with incredible accuracy.</p><p><figure align="center" filename="clock_160.jpg"><caption>Possibly the most accurate clock in the world - but not quite ready to hang on the wall.</caption><source>© Scott Diddams </source></figure></p><p>Scott Diddams of the National Institute of Standards and Technology (NIST) in Boulder and colleagues have, like others before them, used vibrations of atoms to measure time<bibr rid="b1">1</bibr>. But the ticks of their timepiece are provided by atomic oscillations of higher frequency than previously obtained.</p><p>Every clock needs a pacemaker. Early devices used a swinging pendulum; quartz watches use a vibrating quartz crystal. </p><p>In conventional atomic clocks, atoms rearrange their electron clouds when stimulated by microwaves oscillating exactly at the atoms' resonant frequency. The detection of the excited atoms then retunes the frequency of the microwave generator. The oscillator producing the microwaves becomes locked to the atoms' resonant frequency, which remains extremely stable.</p><p>Atomic clocks are now commercially available - for navigation, for instance. These clocks use the electronic resonances of caesium atoms to mark time. NIST (then called the National Bureau of Standards) first introduced caesium atomic clocks in 1951. In 1967 the unit of time was formally defined in terms of a microwave-frequency resonance of the caesium atom: one second is occupied by 9,192,631,770 such cycles. </p><p>The best caesium clocks are accurate to something like one second in 30 million years. This might sound good enough for anyone's purposes, but physicists still hanker after even more precise timepieces. </p><p>Some of the most crucial properties of time, space and matter reveal themselves only to experiments of phenomenal precision. For example, some researchers believe that the 'fine-structure constant' - which measures how strongly matter interacts with light - might not be constant at all but may vary very slowly as the Universe gets older. Better clocks might enable them to test this idea.</p><p>An atomic clock's accuracy is limited by the frequency of its 'ticks'. Atoms have resonances at higher frequencies than that used for caesium clocks. But ticks in such close succession are hard to count using existing electronic devices. </p><p>Diddams's group counted the cycles of an atomic resonance of mercury at the frequencies of visible light, which oscillates faster than microwaves.</p><p>Atom-cooling techniques developed in the past decade have transformed atomic timekeeping. The colder an atom, the less its resonance is blurred by its motion. The Colorado researchers excite the resonance of a single, trapped mercury atom. They use a visible-light laser boosted into the ultraviolet range.</p><p><media width="600" height="600" number="1" filename="clock1.jpg" filetype="Large Image" image="clock2_160.jpg"><caption>Mechanical analogy of the optical clock.</caption></media></p><p>They solve the tick-counting problem using a second laser that emits light in pulses just 25 femtoseconds long - a femtosecond is 10-15 seconds. This laser interacts with the visible light, converting it to an analogous signal whose timing is set by the pulse rate. It converts a high-frequency light signal to a lower-frequency microwave signal, whose oscillations can be monitored - just as gears turn fast revolutions into slower ones.</p><p>While the new clock is potentially 100 times more accurate than caesium clocks, Diddams stresses that he and his colleagues have not yet demonstrated this. Rather, they have shown only that the clock is more stable than existing ones. </p><p>Stability is very important for accurate timekeeping. But it doesn't by itself guarantee accuracy - a clock could very stably gain a minute a day, for example. To prove that the new clock is as accurate as predicted will require more work.</p></body>   <bm> 	 <refgrp> 		<bib id="b1" homeurl="http://www.sciencemag.org"><refau> 		  <snm>Diddams</snm>, 		  <inits>S. A.</inits> et al.</refau> <atl>An optical clock based on a single trapped <sup>199</sup>Hg<sub>+</sub> ion</atl>. <jtl>Science</jtl>,		  (<pubyear>2001</pubyear>).		  </bib></refgrp> <features><related_stories url="010614/010614-14"><title>Atoms shunted and shot</title><pubdate><dayofweek name="Friday"/><day>15</day><month>June</month><year>2001</year></pubdate></related_stories><related_stories url="990826/990826-1"><title>Measuring gravity with an atomic fountain</title></related_stories><related_stories url="990715/990715-1"><title>Landfall on the island of stability</title></related_stories></features><pic_idea>Image of laser set-up available at http://www.eurekalert.org/jrnls/sci/pages/diddams.html Otherwise, images of clocks.</pic_idea>   </bm> </nsuarticle> 
