Primary thermometry in the intermediate Coulomb blockade regime

We investigate Coulomb blockade thermometers (CBT) in an intermediate temperature regime, where measurements with enhanced accuracy are possible due to the increased magnitude of the differential conductance dip. Previous theoretical results show that corrections to the half width and to the depth of the measured conductance dip of a sensor are needed, when leaving the regime of weak Coulomb blockade towards lower temperatures. In the present work, we demonstrate experimentally that the temperature range of a CBT sensor can be extended by employing these corrections without compromising the primary nature or the accuracy of the thermometer.

Ex-situ Tunnel Junction Process Technique Characterized by Coulomb Blockade Thermometry

We investigate a wafer scale tunnel junction fabrication method, where a plasma etched via through a dielectric layer covering bottom Al electrode defines the tunnel junction area. The ex-situ tunnel barrier is formed by oxidation of the bottom electrode in the junction area. Room temperature resistance mapping over a 150 mm wafer give local deviation values of the tunnel junction resistance that fall below 7.5 % with an average of 1.3 %. The deviation is further investigated by sub-1 K measurements of a device, which has one tunnel junction connected to four arrays consisting of N junctions (N = 41, junction diameter 700 nm). The differential conductance is measured in single-junction and array Coulomb blockade thermometer operation modes. By fitting the experimental data to the theoretical models we found an upper limit for the local tunnel junction resistance deviation of ~5 % for the array of 2N+1 junctions. This value is of the same order as the minimum detectable deviation defined by the accuracy of our experimental setup.