We present a novel resonant MEMS sensor for the detection of particles with dielectric properties in liquid environments. The sensor consists of a piezoelectrically excited silicon microcantilever with integrated tailored electrodes on the surface that accumulate particles due to dielectrophoresis. By monitoring the mass-induced shift of the cantilever-related resonance frequency, the accumulated particle mass during the measurement cycle is deduced. We evaluated the mass-sensing performance by exposing the sensor to four artificially aged lubricating oils, thus containing different particle concentrations due to the oxidative degradation of the oil constituents. The measured resonance frequency shift showed a clear correlation to the degree of oil oxidation and is used to determine the accumulated mass by applying an automated particle detection scheme. The measured mass responsivity and limit of detection are in the range of 13 Hz/ng and 3.5 ng, respectively. Compared to standard laboratory methods, such as infrared spectroscopy, the analysis of the sensor response with respect to the oil quality shows that even subtle changes of the oil condition can be detected. These results pave the way for on-line monitoring of oil degradation in industrial applications, such as needed for the condition-based and predictive maintenance of lubricated machinery.

我们提出了一种新颖的共振MEMS传感器，用于检测液体环境中具有介电特性的颗粒。该传感器由压电激发的硅微悬臂梁组成，在其表面上集成有定制电极，这些电极由于介电电泳而积聚颗粒。通过监视质量引起的与悬臂有关的共振频率的偏移，可以得出测量周期内的累积颗粒质量。我们通过将传感器暴露于四种人工老化的润滑油中来评估质量感测性能，由于机油成分的氧化降解，因此含有不同的颗粒浓度。测得的共振频移显示出与油氧化程度的明显相关性，并通过应用自动粒子检测方案用于确定累积质量。测得的质量响应度和检测极限分别在13 Hz / ng和3.5 ng的范围内。与标准实验室方法（例如红外光谱法）相比，对传感器对机油质量的响应分析表明，甚至可以检测到机油状况的细微变化。这些结果为工业应用中油品降解的在线监测铺平了道路，例如润滑机械的基于状态和预测性维护所需要的。例如红外光谱，对机油质量的传感器响应分析表明，即使是机油状况的细微变化也可以被检测到。这些结果为在线监测工业应用中的油降解铺平了道路，例如润滑设备的基于状态和预测性维护所需要的。例如红外光谱，对机油质量的传感器响应分析表明，即使是机油状况的细微变化也可以被检测到。这些结果为在线监测工业应用中的油降解铺平了道路，例如润滑设备的基于状态和预测性维护所需要的。