As a result of their IC compatibility, high acoustic velocity, and high thermal conductivity, aluminum nitride (AlN) resonators have been studied extensively over the past two decades, and widely implemented for radio frequency (RF) and sensing applications. However, the temperature coefficient of frequency (TCF) of AlN is −25 ppm/°C, which is high and limits its RF and sensing application. In contrast, the TCF of heavily doped silicon is significantly lower than the TCF of AlN. As a result, this study uses an AlN contour mode ring type resonator with heavily doped silicon as its bottom electrode in order to reduce the TCF of an AlN resonator. A simple microfabrication process based on Silicon-on-Insulator (SOI) is presented. A thickness ratio of 20:1 was chosen for the silicon bottom electrode to the AlN layer in order to make the TCF of the resonator mainly dependent upon heavily doped silicon. A cryogenic cooling test down to 77 K and heating test up to 400 K showed that the resonant frequency of the AlN resonator changed linearly with temperature change; the TCF was shown to be −9.1 ppm/°C. The temperature hysteresis characteristic of the resonator was also measured, and the AlN resonator showed excellent temperature stability. The quality factor versus temperature characteristic was also studied between 77 K and 400 K. It was found that lower temperature resulted in a higher quality factor, and the quality factor increased by 56.43%, from 1291.4 at 300 K to 2020.2 at 77 K.

Nowadays, the development of real-time water quality monitoring sensors is critical. However, traditional water monitoring technologies, such as enzyme-linked immunosorbent assay (ELISA), liquid chromatography, mass spectroscopy, luminescence screening, surface plasma resonance (SPR), and analysis of living bioindicators, are either time consuming or require expensive equipment and special laboratories. Because of the low cost, self-sustainability, direct current output and real-time response, microbial fuel cells (MFCs) have been implemented as biosensors for water toxicity monitoring. In this paper, we report a microscale MFC biosensor to study the dose–response curve of exoelectrogen to toxic compounds in water. The microscale MFC biosensor has an anode chamber volume of 200 μL, which requires less sample consumption for water toxicity monitoring compared with macroscale or mesoscale MFC biosensors. For the first time, the MFC biosensor is exposed to a large formaldehyde concentration range of more than 3 orders of magnitudes, from a low concentration of 1 × 10−6 g/L to a high concentration of 3 × 10−3 g/L in water, while prior studies investigated limited formaldehyde concentration ranges, such as a small concentration range of 1 × 10−4 g/L to 2 × 10−3 g/L or only one high concentration of 0.1 g/L. As a result, for the first time, a sigmoid dose–response relationship of normalized dose–response versus formaldehyde concentration in water is observed, in agreement with traditional toxicology dose–response curve obtained by other measurement techniques. The biosensor has potential applications in determining dose–response curves for toxic compounds and detecting toxic compounds in water.

No abstract