Logo of the Physikalisch-Technische Bundesanstalt

World record for validation of a quantum current source

PTB verifies the quantization of the current sourced by a single-electron pump with unprecedented accuracy

27.11.2015

Single-electron pumps are promising candidates for the realisation of a future quantum current standard and subject of intense metrological research. Using a novel, highly accurate current amplifier PTB researchers have succeeded in verifying the quantisation of a current of the order of 100 pA sourced by a single-electron pump with unmatched accuracy: the current agreed with the expected quantized value within the relative measurement uncertainty of only 2 . 10-7. This measurement uncertainty is about a factor of five lower than previously obtained results and, for the first time, better than the best ampere realisation in SI unit system.

 

Single-electron pumps are nanostructured electrical circuits that enable the transport of individually clocked charge quanta (electrons with elementary charge e). At PTB such circuits are manufactured on the basis of semiconductor heterostructures and investigated for future applications in metrology. With clock speeds ƒ up to the gigahertz range currents of more than hundred picoampere can be generated, quantised according to I = e . ƒ. Figure 1 shows schematically the principle of operation. Single-electron pumps are, thus, promising candidates for quantum current sources, enabling a future realisation of the unit for electrical current after the forthcoming redefinition of the ampere (see also PTB-News 2.2014).
PTB researchers used an in-house developed, high-accuracy current amplifier which is calibrated traceably to electrical quantum standards (“Ultrastable Low-noise Current Amplifier”, see also PTB-News 1.2015). It surpasses the accuracy of available state-of-the-art picoammeters by about a factor of hundred. The current generated by a single electron pump, operated ƒ = 545 MHz, was measured with this new instrument in dependence on external control parameters. The relative systematic uncertainty in these current measurements was only 1.3 . 10-7.
Figure 2 shows the measurement results under variation of the voltage on the control electrode controlling the right barrier of the “quantum dot” of the electron pump. Depending on the measurement time, statistical standard uncertainties per point down to 6 . 10-7 were achieved (represented by error bars in the figure). Within a relatively large range of about 11 mV the measurement values differed less than two standard deviations from the value of the inflection point of a theoretical model curve. Averaging the measured current values in this plateau region - in the figure shown as dashed box - gave the final result: The relative difference between generated current and the quantized nominal value, i.e. (I - e . ƒ)/(e . ƒ), was -0.94 ± 1.94 . 10-7. This result, thus, demonstrated that the generated current agreed with the quantised expectation value e . ƒ within an overall uncertainty of about 2 . 10-7 - for 545 million of transferred electrons per second this corresponds to an uncertainty of only around 100 electrons. It means a reduction in uncertainty about a factor of five compared to previously achieved results.
In addition, the accuracy achieved in this “quantum-ampere” demonstration for the first time excels the best possible experimental realisation of the ampere by “classic” experiments in the SI unit system: this achieves a total uncertainty of 2.7 . 10-7.

 

           

Figure 1: Operating principle of the investigated single-electron pump: Single electrons (shown as red balls) coming from the left-hand side of the conductor are trapped by a “dynamic quantum dot” (between the “hills” of the potential landscape shown) and then ejected to the other side. The whole cycle is repeated with the frequency of the ac voltage modulating the left barrier of the quantum dot.

 

 

 

Figure 2: Precision current measurement at a single electron pump (ƒ = 545 MHz corresponding to I = e . ƒ ≈ 87 pA) shown as relative deviation from the nominal value, as function of the voltage at the static control electrode. Error bars correspond on the points measured with different integration times correspond to standard statistical uncertainties. Also shown is a theory curve fitted to the measured data. The current values measured in the plateau region (data within the gray box) were averaged and yielded the final result.

 

 

Scientific publication:
F. Stein, D. Drung, L. Fricke, H. Scherer, F. Hohls, C. Leicht, M. Götz, 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, 103501 (2015)