## How the Planck Constant is Better than a Kilogram Artifact, or How the History of Measuring Physics Constants Will Lead to a “New” International System of Metric Units

##### Abstract

For over 200 years, the metric system has been the standard for comparing measurements in science and industry. Formal procedures were adopted about 125 years ago to create the International System (SI) of units, and it has been steadily improved. In the next few years, the SI will be completely redefined to make all units more reproducible for the foreseeable future. The base concepts of time, length, mass, charge, temperature, amount of substance, and luminosity will have SI units of seconds, meters, kilograms, coulombs, kelvins, moles, and candela, respectively, all linked to fundamental physical constants. Conducting tests to obtain consistent and better values for physics constants has a long history, going back to Galileo trying to measure the speed of light. Many physics constants now have values in SI units of 8 digits or more, but that accuracy improved slowly. The Planck constant h has the shortest history of them all, since it was only conceived in the modern quantum approach to atomic theory. This talk will use the Planck constant as an example of how the uncertainty in constants measurements decreases, but not without jumps and disagreements, even increasingly smaller ones, that still have significant effects. Change always involves controversy. The early Planck values “quickly” changed by one percent as physics itself developed. Even in the 1970's and 80's there was controversy over voltage units, arising in the changeover from standard chemical cells to the quantum-based Josephson effect. Today's hot topic concerns changing from the last artifact, a kilogram mass standard, to a Planck constant (or Avogadro constant) definition. The recent discrepancies over Planck and Avogadro results are a factor of 100 smaller than those over voltage, showing how electronic metrology is still progressing, and that research on accurate measurements is never complete.