The Properties and Sensing Mechanism of PdNPs-decorated Silicon Nanobelt Devices for H2 Sensing at Room Temperature

Citation Author(s):
Yu-Sheng
Lin
National Yang Ming Chiao Tung University
Jeng-Tzong
Sheu
National Yang Ming Chiao Tung University
Submitted by:
Yu-Sheng Lin
Last updated:
Mon, 07/08/2024 - 15:59
DOI:
10.21227/1npe-tw43
Data Format:
Research Article Link:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This study reports the properties and the sensing mechanism of Pd nanoparticles (PdNPs) decorated n+/n-/n+ double-junction silicon nanobelt (SNB) device as hydrogen (H2) gas sensor. The SNB devices are prepared via CMOS process. Plasma-enhanced atomic layer deposition (PEALD) is adopted for PdNPs deposition as sensing material on the Al2O3 dielectric of SNB devices. The PdNPs-decorated SNB devices working at room temperature are characterizedat H2 concentration ranging from 10 to 1000 ppm. Instead of using the traditional steady-state response, the slope of response is presented to estimate concentration and shorten the response time. More than 60% improvement in response time has been achieved for 10 to 1000 ppm H2 detection. To reduce recovery time, device localized Joule heating (DLJH) with a bias of 11 V for 240 s is demonstrated to restore the device back to the baseline.At a bias of 1 V, H2 sensing at room temperature consumes only 68.39 μW.

Instructions: 

Fig. 1. (a) The schematic image and the top-view SEM image of PdNPs-decorated SNB device with five parallel Si nanobelts and the device process flows. (b) The cross-sectional TEM images of the active channel of SNB device showing PdNPs on the surface Al2O3 dielectric layer. (c) The longitudinal EDX-line-scanning of the SNB device showing the elemental profile of the PdNPs-decorated SNB device at n- region.

Fig. 2. (a) The ID–VBG characteristics of the pristine SNB device, after H2 plasma treatment, the deposition of PdNPs and VD = 11 V for 240 s treatment. (b) The real-time measurements of the sensitivity upon exposure to 100 ppm H2 at VD = 1 V and various VBG. (c) The band diagrams of the Pd-decorated SNB device before and after H2 exposure.

Fig. 3. COMSOL simulations of DLJH. (a) The temperature distribution of SNB devices along the channel at VD = 11 V and VBG = 0 V. (b) The temperature profiles along the channel at various VD. (c) The power consumption and the corresponding temperature of the PdNPs-decorated SNB device at n- region at various VD and VBG = 0 V.

 Fig. 4. The real-time measurements of H2 at various concentrations atVD = 1 V. (a) The real-time current and the slope of responses. (b) The steady-state response vs. the maximum slope of response with respect to the square root of H2 concentration from 100 to 1000 ppm. (c) The steady-state response vs. the maximum slope of response with respect to the square root of H2 concentration from 10 to 100 ppm. (d) The comparison between the gas steady-state response time and the maximum slope response time.