High-resolution TENGS for earthquakes ground motion detection

Highlights

• Highly sensitive and cost-effective TENG developed were used to detect earthquakes.

• LoRa protocol-based wireless transmission system was implemented to transfer SEIS-TENG signals.

• SEIS-TENG showed similar results compared to the professional earthquake simulation devices.

Abstract

Towards a deeper consciousness of the human being in preserving the natural environment of our planet, huge effort is nowadays being exerted by scientists all over the world to use green and free energy. The main motivation for this effort is to protect humanity, today and in the future, from the big natural disasters that often affect the poorest areas of the globe. In this work, TENGs (triboelectric nanogenerators) with an inertial mass on their surface are proposed to be used as self-powered seismic sensors (SEIS-TENGs) in the detection of seismic waves with global connectivity to the Internet of Things (IoT). For this purpose, various TENGs with several characteristics such as low cost, flame retardancy, and frequency-dependent sensitivity were fabricated with different materials such as Paper, Polyvinylalcohol (PVA), Polyvinyledenefluoride (PVDF) and PDMS. The TENGs based on PVA10PPA-PEI/PVDF exhibited the highest sensitivity of 280 Hz and the ones based on Paper/PDMS were very low cost and easy to manufacture. Pulses, artificially generated with an electro-mechanical machine operating in fatigue mode, were remotely transmitted by using LoRA communication protocol and could be monitored in the TTS (The Thing of Stack) platform. Finally, a realistic application of SEIS-TENG was carried out by simulating earthquakes with a triaxial shaking table and the seimsic waves were measured using SEIS-TENGs. Interestingly, they exhibited a similar response to the acceleration, velocity, and displacement, and with today’s commercial accelerometers such as micromechanical systems (MEMS). Furthermore, the number of quakes detected by SEIS-TENG was very similar to that of the MEMS. The results obtained demonstrate that the SEIS-TENGs are very promising to be employed in a global IoT seismic network. They will allow measuring ground oscillations to prevent the catastrophes caused by earthquakes.

Introduction

There is an increasing interest in the last decades of finding optimal and automated smart-sensing systems [1], [2], [3] able to prevent huge catastrophes produced by periodical severe seismic episodes [4], [5], [6]. Up to now, we do not have capability enough to accurately predict the onset of big earthquakes and their consequences. However, the evolution of machine learning, artificial intelligence (AI) algorithms and big data analytical tools [7], [8], [9] together with the fabrication of high-power and high-speed computer processors [10,11], fast and high-resolution data acquisition systems (DAQs) [12,13] and highly evolved seismic networks [14,15] is helping to identify and classify with much more accuracy critical variables relevant in the prediction of these dramatic seismic effects.

Towards these possible predictions, obtaining these variables is essential. With this aim, different seismic sensors have been set up at different networks all over the globe [16]. High-sensitivity, broadband and strong-motion seismographs [17], [18], [19], [20] are often equipped with short-period velocity seismometers [21], accelerometers [22] and tiltmeters [23] constituting permanent seismic networks [24,25]. They are placed on the ground with their DAQs and electronics integrated, inside small rooms or houses built in regions either with medium or high seismic activity. Others, such as large N-arrays [26,27] including micro-electromechanical systems (MEMS) [28], [29], [30], distributed acoustic systems (DAS) [31,32] and nodes [13], [33] (each including a geophone, a DAQ and a battery) make use of thousands of sensors separated up to several kilometers of distance one each other. In addition, fiber optic sensing [34], [35], [36], [37] is distributed and continuous and operates in base of the backscattering of the light as it encounters either some crystal impurities or changes in temperature when propagating along the fiber. All these seismometers constitute a broad network with interconnected units via Wi-Fi [38,39], GSM [40,41] and GPS communication. However, they present some disadvantages [1]: smartphone MEMS are not suitable for small earthquake detection while high accuracy MEMS are very expensive and saturate with strong motion. In addition, DAS cannot detect activity for distances further than 100 km long (this is short to cover ocean areas) and their arrays of nodes do not detect deep events. Traditional seismographs ignore rotational motions while high-resolution ones use high-cost and sophisticated instrumentation. Others, such as fiber optics sensing (different from DAS) [42–43] devices present low spatial resolution and they are not accurate in locating specific events although they can reach much farther distances than DAS, being still efficient in detecting large and moderate earthquakes. High-tech permanent networks are expensive and their distance separation between units is in the sub-km range. Temporary ones with high accuracy and sensitivity demand long time shipment and deployment and as a result, important events after the main shock may be lost. Hence, novel sensors integrated as a whole system capable of overcoming these limitations are globally under research by various researchers [44].

Triboelectric nanogenerators (TENGs) exhibit excellent sensitivity to mechanical stress, high mechanical-electrical conversion efficiency, tunable size, and low cost. In this work, TENGs are proposed as seismic sensors to identify different types of seismic waves and as a result, to measure different ground oscillations. Six different types of TENGs were studied, which exhibit various characteristics such as fire retardancy, low cost, and ease of fabrication with the following chemistries; PVA (0, 1, 5, and 10 wt% PPA-PEI) /PVDF, Paper@ 50PA /PVDF-HFP and Paper /PDMS. Furthermore, the TENG with the highest sensitivity was determined and used for further proof-of-concept demonstrations. Artificial oscillations with electro-mechanical and hydraulic-mechanical vibration testing machines were produced with sinusoidal excitation frequencies up to 50 Hz. High frequency pulses generated were transmitted remotely under LoRA protocol and could be read in the The Things of Stack (TTS) platform by anyone with access to the internet. Moreover, the fabricated TENGs based sensors were employed to detect real ground oscillations simulated using with a 3-axis shaking table at CEDEX and results were compared with those obtained using commercial accelerometers [45] such as the CMG5 one of GURALP [49]. Notably, a strong agreement was obtained between the TENG based sensor and commercial accelerometers in various measured parameters such as; the number of quakes and their frequencies, the acceleration, velocity, and displacement magnitudes. This work responds to society's real need to develop sensors that can detect earthquakes rapidly and at a low cost while exhibiting high accuracy.