The quantum metrology triangle (QMT) tests the mutual consistency of the three quantum electrical effects through the closure condition KJRKQS=2K_J R_K Q_S = 2, where KJ=2e/hK_J = 2e/h (Josephson constant), RK=h/e2R_K = h/e^2 (von Klitzing constant), and QSQ_S is the charge quantum transported by a single-electron pump, expected to equal the elementary charge ee. The best closure to date, the NIST electron counting capacitance standard ECCS-1 (2007), reached only about 0.9 parts per million, and the PTB ECCS experiment was discontinued when the 1 pF cryogenic vacuum capacitor revealed unexplained low-temperature, frequency-dependent resonance structures of several to tens of μF F−1\mu\mathrm{F\,F^{-1}} that made the required extrapolation of the audio-band capacitance to the effective dc charging frequency impossible at the needed 1 μF F−11\ \mu\mathrm{F\,F^{-1}} accuracy. The open problem left by that discontinuation is: (i) the physical origin and elimination of the low-temperature resonance structures in cryogenic vacuum-gap capacitors, (ii) a validated determination of the effective low-frequency capacitance with uncertainty well below 1 μF F−11\ \mu\mathrm{F\,F^{-1}} and at the level required by the 10−810^{-8} closure budget, and (iii) the overarching objective, an experimentally realized QMT closure with total relative standard uncertainty of order 10−810^{-8}, whether via a re-engineered ECCS or via the direct route of self-referenced single-electron current sources with quantum-Hall- and Josephson-traceable small-current metrology. No work to date has explained the resonance structures, the ECCS principle has not been revived, and the best direct-route measurement (a 2 GHz silicon pump at 320 pA, 0.16-0.18 ppm combined uncertainty) remains more than an order of magnitude from the target.

Background

Modern electrical metrology rests on three quantum effects that link electrical quantities to fundamental constants. The Josephson effect relates a voltage to a frequency, U=f/KJU = f/K_J with KJ=2e/hK_J = 2e/h (ee is the elementary charge, hh Planck's constant); the quantum Hall effect (QHE) relates resistance to RK=h/e2R_K = h/e^2; and single-electron tunnelling (SET) - the controlled transfer of individual electrons through nanoscale 'pumps' exploiting Coulomb blockade, the electrostatic energy cost of placing one electron on a tiny island - relates a current to a frequency, I=QSfI = Q_S f, where QSQ_S is the transferred charge quantum, predicted to be exactly ee. Since no theory forbids small deviations, the quantum metrology triangle (QMT) tests their joint consistency by checking the product KJRKQS=2K_J R_K Q_S = 2. An exact closure would confirm that the three quantum electrical standards - voltage, resistance, and current - are mutually consistent.

The electron counting capacitance standard (ECCS) pioneered at NIST was the first route to closing the triangle: an SET pump transfers NN electrons one by one onto a cryogenic vacuum-gap capacitor ('cryocap') at millikelvin temperatures in a dilution refrigerator; the resulting voltage is measured against a Josephson voltage standard (JVS) and the capacitance against a transformer-based bridge whose impedance is traceable to a quantum Hall resistance (QHR) standard. The final NIST result (ECCS-1, 2007) closed the triangle with a standard uncertainty of about 9 parts in 10710^7. The PTB variant used a metallic five-junction 'R-pump' (tunnel junctions with series micro-resistors that suppress co-tunnelling errors) and a 1 pF coaxial vacuum-gap capacitor with stainless-steel electrodes separated by 5 mm; its 2012 preliminary closure reached about 2 parts in 10610^6.

The obstacle that ended the PTB experiment is a frequency-scale mismatch. The capacitance bridge measures at audio frequencies (kilohertz), while the ECCS charges the capacitor at an effective frequency near 10 mHz, so the ac result must be extrapolated to dc with relative uncertainty of about 1 μF F−11\ \mu\mathrm{F\,F^{-1}} (one part in 10610^6) or better. High-accuracy frequency sweeps instead revealed sharp, double-peak-like resonance structures in Ccryo(f)C_{\mathrm{cryo}}(f) with amplitudes of several to more than 10 μF F−110\ \mu\mathrm{F\,F^{-1}} (the largest, near 1.5 kHz, about 60 μF F−160\ \mu\mathrm{F\,F^{-1}} peak-to-peak). A roughly two-year, eight-step diagnostic campaign excluded the needle valve and refrigerator operation, the filtered cryogenic coaxial cabling, the needle switch, galvanic grounding, and electro-acoustic coupling in the semi-rigid lines, and showed the structures vanish when the capacitor is warm: the cold cryocap itself is the resonant element. A driven-mechanical-resonance picture (weakly damped oscillations of capacitor components at millikelvin temperature, excited by the bridge voltage) was suggested but remained speculative, no model was confirmed, and no redesign or remediation was demonstrated; the project was stopped.

Meanwhile the metrological bar rose. The CODATA 2014 adjustment of fundamental constants reduced the uncertainties of the possible correction factors for KJK_J and RKR_K to about 2 parts in 10810^8, so a QMT closure would need a total uncertainty of order one part in 10810^8 to improve confidence in the Josephson and quantum Hall relations. After the 2019 revision of the SI, in which ee and hh are exact defining constants and KJK_J, RKR_K are fixed conventional values, the QMT instead tests the SET relation QS=eQ_S = e itself - whether single-electron devices really transfer exactly one elementary charge - and enables a direct realization of the ampere from ee and frequency. Neither the NIST nor the PTB ECCS could plausibly reach 10−810^{-8}. The alternative architecture proposed at PTB measures an SET-generated current directly in terms of RKR_K and KJK_J: a 'self-referenced' quantized current source - several pumps in series combined with single-electron detectors that count transfer errors during current generation, with radio-frequency SET electrometers pushed toward 100 kHz detector bandwidth - sources about 100 pA, which an ultrastable low-noise current amplifier (ULCA, a transimpedance amplifier with 1 GΩ\Omega effective transresistance, 1.4 fA/Hz\sqrt{\mathrm{Hz}} noise, calibrated daily against the QHR) converts to a voltage measured against a JVS. This route was projected to reach about 0.1 μA A−10.1\ \mu\mathrm{A\,A^{-1}} (one part in 10710^7) - still an order of magnitude short of the 10−810^{-8} target - and has not been completed as a triangle experiment either.

Problem Statement

The open problem is the experimentally realized closure of the quantum metrology triangle at the accuracy level at which it is metrologically significant today: a test of KJRKQS=2K_J R_K Q_S = 2 (equivalently, of whether the single-electron charge quantum QSQ_S equals ee) with a total relative standard uncertainty of order one part in 10810^8, together with resolution of the specific obstacles that stopped the most advanced such experiment. The common objective uniting the sub-questions is a complete QMT experiment with a full, experimentally quantified uncertainty budget; the sub-questions are the two routes toward it left open by the PTB electron-counting capacitance standard (ECCS) record.

(1) Capacitance (ECCS) route: determine the physical origin of the pronounced low-temperature, frequency-dependent resonance structures observed in cryogenic vacuum-gap capacitors at millikelvin temperature (amplitudes of several to tens of μF F−1\mu\mathrm{F\,F^{-1}} in the audio band, vanishing at room temperature, localized to the cold capacitor by systematic exclusion of cabling, switches, grounding, and refrigerator operation) - specifically, identify the mechanism well enough either to eliminate the structures by capacitor geometry, materials, or operating-condition redesign while preserving the metrological properties (stability, voltage coefficient, leakage) that an ECCS requires, or to establish that they are intrinsic to vacuum-gap capacitors at base temperature. On that basis, provide a validated method to determine the effective dc (charging-frequency) capacitance Ccryo(f→0)C_{\mathrm{cryo}}(f \to 0) - extrapolating or otherwise bridging from audio-frequency bridge measurements to the about 10 mHz effective charging frequency of an ECCS - with relative uncertainty well below 1 μF F−11\ \mu\mathrm{F\,F^{-1}} - the historical minimum for a viable ECCS extrapolation - and at the level required by the 10−810^{-8} closure budget.

(2) Direct single-electron-current route: realize a QMT experiment that measures an SET-generated current directly in terms of RKR_K and KJK_J at the 10−810^{-8} level. This requires a quantized current source with in-situ error accounting (a self-referenced arrangement of pumps in series with single-electron detectors whose bandwidth covers the pump error rates, of order 100 kHz), a current level high enough (of order 100 pA or more) to keep counting statistics and amplifier noise below 10−810^{-8} over practical integration times, and a small-current measurement chain - ultrastable transimpedance amplification traceable to the quantum Hall resistance with output voltages referred to a Josephson standard, or an equivalent scheme - whose own scaling accuracy, calibration drift, and unexplained offsets are controlled at 10−810^{-8}, a level one to two orders of magnitude beyond the best existing small-current measurements (0.16-0.18 ppm) and the best previous triangle closure (0.9 ppm).

A complete answer is an end-to-end experiment, on either route (or a demonstrably equivalent architecture), whose published uncertainty budget for the closure quantity KJRKQS/2−1K_J R_K Q_S / 2 - 1 reaches a total relative standard uncertainty of order 10−810^{-8} with every systematic component experimentally quantified; this end-to-end closure, on either route, is the sole completion criterion. The capacitor sub-problem of route (1) is a separately resolvable, load-bearing step rather than a complete answer: an experimentally validated identification of the resonance mechanism - including the negative outcome that the structures are intrinsic to vacuum-gap capacitors at base temperature - together with a validated method for Ccryo(f→0)C_{\mathrm{cryo}}(f \to 0) at the accuracy stated above answers sub-question (1) as a self-contained result, but is a necessary and not a sufficient condition for the completion criterion; an 'intrinsic' outcome forecloses the ECCS route and leaves route (2) as the path to that criterion.

References

  1. H. Scherer, J. Schurr, F. J. Ahlers, 'Electron counting capacitance standard and quantum metrology triangle experiments at PTB', Metrologia 54 (2017) 322-338. DOI: 10.1088/1681-7575/aa65f9. https://doi.org/10.1088/1681-7575/aa65f9 (preprint arXiv:1701.04759, https://arxiv.org/abs/1701.04759)
  2. M. W. Keller, N. M. Zimmerman, A. L. Eichenberger, 'Uncertainty budget for the NIST electron counting capacitance standard, ECCS-1', Metrologia 44 (2007) 505-512. DOI: 10.1088/0026-1394/44/6/010. https://doi.org/10.1088/0026-1394/44/6/010
  3. M. W. Keller, 'Current status of the quantum metrology triangle', Metrologia 45 (2008) 102-109. DOI: 10.1088/0026-1394/45/1/014. https://doi.org/10.1088/0026-1394/45/1/014
  4. H. Scherer, B. Camarota, 'Quantum metrology triangle experiments: a status review', Measurement Science and Technology 23 (2012) 124010. DOI: 10.1088/0957-0233/23/12/124010. https://doi.org/10.1088/0957-0233/23/12/124010
  5. B. Chenaud, C. Hernandez, A. Delgard, C. Chaubet, 'Influence of coaxial cable resistance on impedance measurements at low frequency and guidelines for phase determination of ac electrical signals', Review of Scientific Instruments 94 (2023) 115112. DOI: 10.1063/5.0172200. https://doi.org/10.1063/5.0172200
  6. L. Fricke, M. Wulf, B. Kaestner, F. Hohls, P. Mirovsky, B. Mackrodt, R. Dolata, T. Weimann, K. Pierz, U. Siegner, H. W. Schumacher, 'Self-Referenced Single-Electron Quantized Current Source', Physical Review Letters 112 (2014) 226803. DOI: 10.1103/PhysRevLett.112.226803. https://doi.org/10.1103/PhysRevLett.112.226803
  7. D. Drung, C. Krause, U. Becker, H. Scherer, F. J. Ahlers, 'Ultrastable low-noise current amplifier: A novel device for measuring small electric currents with high accuracy', Review of Scientific Instruments 86 (2015) 024703. DOI: 10.1063/1.4907358. https://doi.org/10.1063/1.4907358
  8. F. Stein, D. Drung, L. Fricke, H. Scherer, F. Hohls, C. Leicht, M. Goetz, C. Krause, R. Behr, E. Pesel, K. Pierz, U. Siegner, F. J. Ahlers, H. W. Schumacher, 'Validation of a quantized-current source with 0.2 ppm uncertainty', Applied Physics Letters 107 (2015) 103501. DOI: 10.1063/1.4930142. https://doi.org/10.1063/1.4930142
  9. F. Stein, H. Scherer, T. Gerster, R. Behr, M. Goetz, E. Pesel, C. Leicht, N. Ubbelohde, T. Weimann, K. Pierz, H. W. Schumacher, F. Hohls, 'Robustness of single-electron pumps at sub-ppm current accuracy level', Metrologia 54 (2017) S1-S8. DOI: 10.1088/1681-7575/54/1/S1. https://doi.org/10.1088/1681-7575/54/1/S1
  10. H. Scherer, H. W. Schumacher, 'Single-Electron Pumps and Quantum Current Metrology in the Revised SI', Annalen der Physik 531 (2019) 1800371. DOI: 10.1002/andp.201800371. https://doi.org/10.1002/andp.201800371
  11. S. P. Giblin, A. Fujiwara, G. Yamahata, M.-H. Bae, N. Kim, A. Rossi, M. Moettoonen, M. Kataoka, 'Evidence for universality of tunable-barrier electron pumps', Metrologia 56 (2019) 044004. DOI: 10.1088/1681-7575/ab29a5. https://doi.org/10.1088/1681-7575/ab29a5
  12. S. Li, Q. Wang, W. Zhao, S. Huang, 'From μ0 to e: A Survey of Major Impacts for Electrical Measurements in Recent SI Revision', IEEE Transactions on Instrumentation and Measurement 69 (2020) 5956-5965. DOI: 10.1109/TIM.2020.3010351. https://doi.org/10.1109/TIM.2020.3010351
  13. D. Matsumaru, Jia Zhengsen, M. Maruyama, N.-H. Kaneko, 'A Voltage-tracking scheme for Quantum-Metrology-Triangle Experiments', 2020 Conference on Precision Electromagnetic Measurements (CPEM) (2020). DOI: 10.1109/CPEM49742.2020.9191858. https://doi.org/10.1109/CPEM49742.2020.9191858
  14. S. P. Giblin, G. Yamahata, A. Fujiwara, M. Kataoka, 'Precision measurement of an electron pump at 2 GHz; the frontier of small DC current metrology', Metrologia 60 (2023) 055001. DOI: 10.1088/1681-7575/ace054. https://doi.org/10.1088/1681-7575/ace054
  15. A. Fujiwara, G. Yamahata, N. Johnson, S. Nakamura, N. Kaneko, '(Invited) Silicon Quantum Dot Single-Electron Pumps for the Closure of the Quantum Metrology Triangle', ECS Transactions 112 (2023) 119-130. DOI: 10.1149/11201.0119ecst. https://doi.org/10.1149/11201.0119ecst
  16. T. Gerster, N. Ubbelohde, K. Pierz, T. Weimann, F. Hohls, H. W. Schumacher, 'Reproducible Single-Electron Pumps for a Quantum Current Standard', in: 150th anniversary of the Metre Convention - From Units to the Universe (2025). DOI: 10.59161/juri7463. https://doi.org/10.59161/juri7463
  17. L. K. Rodenbach, J. M. Underwood, N. T. M. Tran, A. R. Panna, M. P. Andersen, Z. S. Barcikowski, S. U. Payagala, P. Zhang, L. Tai, K. L. Wang, D. G. Jarrett, R. E. Elmquist, D. B. Newell, A. F. Rigosi, D. Goldhaber-Gordon, 'A unified realization of electrical quantities from the quantum International System of Units', Nature Electronics 8 (2025) 663-671. DOI: 10.1038/s41928-025-01421-2. https://doi.org/10.1038/s41928-025-01421-2
  18. W. Poirier, S. Djordjevic, 'A universal quantum electrical standard is getting closer', Nature Electronics 8 (2025) 632-634. DOI: 10.1038/s41928-025-01442-x. https://doi.org/10.1038/s41928-025-01442-x