FAQ-second

Frequently Asked Questions concerning the Redefinition of the Second

CCTF Task Force on Updating the Roadmap for the redefinition of the second

 

Table of Contents

 

What is the timeline for redefinition?

The timeline is set by meetings of the General Conference on Weights and Measures (CGPM), which are held every four years. The 2025 CCTF roadmap update keeps 2030 as the decisive milestone.

  • CGPM 2026: Reviews progress and Draft Resolution B for continued work toward a proposal at CGPM 2030.
  • CCTF 2028 & 2029: Evaluate criteria fulfilment, shortlist and finalize proposal with one possibility.
  • CGPM 2030: Decision on the new definition and its implementation date, provided there is consensus and the mandatory criteria are met.

If decided, implementation of the new definition would be expected to follow after CGPM 2030.

 

Why does the community want to redefine the second?

From a metrological point of view, we have reached the point where the best realizations of the current definition (i.e. caesium cold atom fountain clocks) have been surpassed by several optical frequency standards already constituting secondary representations of the second. Some of these have achieved two orders of magnitude lower systematic uncertainties and correspondingly lower frequency instabilities than caesium fountain clocks, with continuing progress being made.

The accuracy of frequency measurements is, therefore, limited by the current definition, rather than being limited by the technological level reached by the most advanced scientific teams. This challenges the recognition of the SI second as the universal common time unit and may hinder scientific and technological progress. As in 1967, this prompts the community to examine alternative options and choose a new definition, in-line with contemporary technological capacities, and able to remain on top of the metrological pyramid for a long time.

 

What are the options to redefine the second?

The SI units are currently defined using seven constants of nature with fixed numerical values. The defining constant for the second, ΔνCs, is a microwave frequency characteristic of the Cs atom. Two options are being considered for a new definition based on an optical transition, to realize improved accuracy:

  • Option 1: Select a single optical atomic transition in place of ΔνCs and fix its frequency as the new definition of the second.
  • Option 2: Create a new defining constant from a weighted geometric average of optical frequencies of several atomic transitions.

A third option, fixing another fundamental constant such as the electron mass, has also been considered. However it is currently unsatisfactory , since no such constant is known accurately enough to be competitive with even the present definition of the second.

Each of these options will impact the definitions of other base units, except for the mole. For more technical details, you can refer to https://iopscience.iop.org/article/10.1088/1681-7575/ad17d2/pdf.

 

What will happen to the caesium standards after the redefinition?

After the redefinition, caesium standards will have an additional uncertainty (of the order of 1 x 10−16), but this will only be significant in the uncertainty assigned to the very best caesium standards such as fountains. The caesium fountains will become a secondary representation of the second.

 

What actions might I need to take prior to the redefinition?

Depending on how legal time is defined in your national legislation, it may be necessary to work with your legislators to revise this definition. The current timetable allows at least four years for this to take place.

Education of the user community and stakeholders about the redefinition is most effectively done by their local NMI. The CCTF plans to provide resources to help with this responsibility.

The time laboratories are encouraged to consider realizing the SI second using the new definition in their forward planning. Improvements to time-transfer links may also be needed to fully realize the benefits of better clocks.

 

Will I need to replace my caesium standards after the redefinition?

No. Although optical frequency standards will become the primary representation of the second, caesium will remain a secondary standard. Traceability for caesium standards will still be provided through Circular-T.

 

Will caesium standards still be accepted into UTC after the redefinition?

Yes. Their exact status will depend on the chosen redefinition option, but after the redefinition the former caesium primary standards will be considered as secondary standards and be included in UTC just like secondary standards are currently used to steer EAL.

The pool of ~ 450 clocks contributing to UTC already mixes commercial thermal beam caesium clocks with hydrogen masers, with its resulting time scale EAL being steered by the primary and secondary frequency standards measurements to get TAI and finally UTC. The redefinition will have no immediate impact on this situation.

 

What are the criteria for deciding if we are ready for the redefinition?

A roadmap towards redefining the SI second to be based on an optical transition was adopted at the 21st meeting of the CCTF in 2017 and was refined and updated by a CCTF taskforce established in 2020. To choose the best new definition and establish a timeline, criteria and conditions were defined as part of the roadmap to assure that the redefinition:

  • offers an immediate improvement in accuracy by 10 – 100×, with a possible larger improvement in the long term;
  • ensures continuity with the definition based on caesium;
  • ensures continuity of the availability of the new SI second, enabling an immediate improvement of the quality of TAI and UTC(k);
  • is acceptable to NMIs and stakeholders, enabling broad dissemination of the unit to users.

Along these lines, eight mandatory criteria were established that must be met before a redefinition. The fulfilment of the mandatory criteria relies on the progress of high-performance and reliable optical frequency standards and time and frequency transfer techniques required for the realization of the new definition and its dissemination to users, including the contributions of optical frequency standards to International Atomic Time (TAI).

Additionally, six ancillary conditions were established that call out essential work that must have achieved a reasonable amount of progress at the time of redefinition, with a commitment of stakeholders to continue their efforts on the associated activities.

Fulfilment indexes were also defined to evaluate the fulfilment level of the mandatory criteria, to draw attention to the remaining work to fulfil the criteria and ultimately, to decide when it is time to change the definition.

For more technical details, you can refer to https://iopscience.iop.org/article/10.1088/1681-7575/ad17d2/pdf.

Will GNSS time transfer be accurate enough for optical clocks to fully contribute to UTC(k) laboratories to benefit from the improvement in the definition of the second?

At present, most links to UTC are formed using GNSS time-transfer. The stability of this method is sufficient to realize the full performance of the best microwave primary frequency standards at practical averaging times of one day. However, GNSS time-transfer is not well-matched to optical clocks, which are 10 to 100 times more accurate than microwave clocks. Although there are prospects for improvements to GNSS time-transfer, these will not be enough to bridge the gap. Optical fibre links provide the required performance and have been demonstrated at the continental scale at around 1000 km. Inter-continental links are more difficult but are an active area of research (for example, advanced Two-Way techniques, optical links in space and transportable optical clocks). The full benefit of optical clocks cannot be exploited with current GNSS links.

What will happen after the redefinition if a new systematic frequency offset for the selected species is identified?

In case of Option 1:

Before 2019, when the SI unit of the kilogram was defined by an artefact “Le Grand K”, the answer to the question, “What would happen if ‘Le Grand K’ were to fall on the ground and get damaged?” was “The weight of the Universe would change”. A similar situation would happen if a new systematic effect were discovered after the definition of the new reference frequency value for the SI second was made. The frequency ratios between different standards would change. The time intervals, if they were measured using frequency standards other than the definition of the SI second, would change as well. This ominous prospect is, however, of low probability and of negligible impact. This is because firstly, rigorous criteria were set for the redefinition to take place, including multiple and various frequency comparisons with very high levels of agreement, and secondly, because the accuracy level for the typical realization of SI second in the laboratories, scientific experiments and technological applications is several orders of magnitude worse than the uncertainty of the measurements on which the definition is based and the possible frequency change due to any new systematic effect. In fact, in 1997, a similar situation to that presented in this question occurred: the effect of black body radiation on the clock frequency of Cs-133 was added to the SI second definition, which resulted in the effects described above, but had no practical impact on the routine TF metrology since the uncertainties of frequency transfer and comparisons were much larger than the changes of the frequency ratios. Thus, a frequency offset of a magnitude invisible in present investigations may affect the interlaboratory realization of time interval and frequency at the highest scientific level but have no influence on routine time measurements and interlaboratory comparisons.

In case of Option 2:

If the frequency offset is caused by an effect that influences all included species in the chosen definition, the outcome will be similar to option 1 above. In case it is an effect influencing one, or a subset, of the included species, the overall effect will be smaller and related to the weight of the affected species. Eventually, the weights can be adjusted for all species used in the definition, which is a faster process with option 2b.

What are the benefits and potential weaknesses of options 1, 2 (a&b) and 3?

What are the benefits of Option 1 (Single Optical Transition)?

  • Option 1 would offer a straightforward, static definition that would follow an identical approach to the current definition based on caesium as the primary frequency standard. The single defining microwave transition would be replaced by a single optical one that could realize a new primary frequency standard with higher accuracy.
  • Option 1 is easily understood and accepted.
  • Option 1 leaves available the possibility of using secondary representations of the second based on other clock transitions for frequency calibrations.

What are the potential weaknesses of Option 1?

  • A consensus on the choice of the optical transition for a new primary frequency standard would be hard to find, since a single transition would have to be chosen from several competing options.
  • If a better standard were to emerge with significantly lower uncertainty, there would be a need for another redefinition, which is a long and complex process.

What are the benefits of Option 2 (Ensemble of Optical Transitions: 2a Static weights, 2b Dynamic weights, Dynamic ensemble)?

  • Option 2 acknowledges that several transitions could be used in the realization of a redefined second and provides a rational method to do so without singling out one transition.
  • Being able to redefine the second with an ensemble of transitions could facilitate a consensus on the redefinition since it could be inclusive of efforts at more institutes, and a decision would not have to be made on which transition would be the “best” one to choose.
  • Option 2b would allow the weights for the defining transitions to adapt as technologies improve, following clock progress faster and for a much longer period. However, discussions remain regarding the establishment of a process for setting the weights. It would also allow the inclusion of new transitions after the redefinition when they reach and/or surpass the required uncertainty levels.

What are the potential weaknesses of Option 2?

  • Both Options 2a and 2b are different from the status quo, more difficult to understand and propose a fundamental change in the common understanding of a single standard as a definition of a unit.
  • A redefined SI second under Option 2 would be harder to realize by a single laboratory in complete isolation.
  • Under Option 2, there would be no primary frequency standard, and all individual clock transitions would have “representation uncertainties”, just like secondary frequency standards have today under the current definition.
  • Option 2b would introduce additional complexity in terms of managing updates to the weights and the value of the defining constant N while ensuring the quality of the defining ensemble of transitions. A process for managing these updates has not yet been established.

Why is Option 3 less favored?

Option 3 would involve fixing the value of another fundamental constant such as the mass of the electron or the Rydberg constant. However, there is currently no fundamental constant that is known with low enough uncertainty that could be used as a new defining constant without degrading the accuracy of the definition of the SI second.

 

Version 2.2
Last updated 2026-10-13 11:12:03 +0200