2.3. Experimental Setup

The testing setup comprised mass flow controllers directed by PC, gas cylinders providing synthetic air, a reference calibration mixture (100 ppm NO₂ balanced in nitrogen), a test chamber, and readout electronics. A more detailed description of the whole setup can be found in previous papers [41,42]. The operational bias voltage was set to −0.5 V vs. platinum RE for all the tested sensors, which were placed on the bottom of the test chamber with gold spring probe contacts. The samples were exposed to various concentrations of nitrogen dioxide ranging from 0 up to 10 ppm under the following conditions: 298 K, 40%RH, 1013.25 hPa, analyte flow rate 1 L/min. Unless otherwise stated, each sensor sample was placed in the same position in the test chamber under the same fluidic conditions.

Sensor performance was investigated in two ways. We focused on steady-state measurements to analyze the mean value of the current response and its fluctuations using our very sensitive laboratory apparatus operating at higher sampling rates. As this is a very time-consuming procedure, the measurement was only performed for one sample for each type of SPE. The second standard approach simultaneously measures six sensor samples with a commercial potentiostat circuit (LMP91000, Texas Instruments, Dallas, TX, USA). This circuit can be used for a wide range of electrochemical sensing applications, and it was used here to obtain a general overview of the effect of the SPE type on the sensor parameters.

The measurement was performed as follows: firstly, a particular NO₂ concentration (e.g., 1 ppm) with a total flow rate of 1 L/min (298 K, 40%RH) was set. After a few minutes, when the sensor current response had achieved a steady state value corresponding to the particular NO₂ concentration level, simultaneous measurements of both direct current and noise were carried out by a two-channel device. The same procedure was then applied to other NO₂ concentration levels.

Every sensor was a part of a potentiostat circuit based on a rail-to-rail operational amplifier (OPA2144, Texas Instruments, Dallas, TX, USA) with the WE grounded configuration [41,44,46]. The proper noise measurement process is highly dependent on using apparatus and a measurement circuit suitable to the task. Thus, the potentiostat circuit and a low noise transimpedance amplifier were combined to form a single device powered by a battery, providing simultaneous current fluctuation and direct current measurement. The AC voltage output was connected to an amplifier equipped with highly selective filters (AM22, 3S Sedlak, s.r.o., Brno, Czech Republic). It was acquired by a 12-bit AD convertor (HS3, TiePie engineering, Sneek, The Netherlands), which also acquired the DC voltage output as the second channel. To minimize the influence of any power peaks and electric field disturbances, the test chamber with the sensor and our potentiostat-circuit device was placed in a Faraday cage, and all of the devices, including the laptop, were powered by batteries.

A group of six sensors was prepared for each SPE type. These sensors were tested simultaneously in the test chamber using a customized evaluation board equipped with six configurable LMP91000 potentiostats and ΔΣ AD converters (LTC2485, Linear technology, Milpitas, California, USA). All the sensors were tested under identical conditions (298 K, 40%RH, 1013.25 hPa, analyte flow rate 1 L/min), and the signal from each sensor was recorded every other second.