Ammonia Sensing via Pseudo Molecular Doping in UV-Activated Ambipolar Silicon Nanowire Transistors

The potential of adsorbed gaseous molecules to create shallow electronic states for thermally excited charge carrier transport and to engineer silicon transistor properties has been largely overlooked compared to traditional substitutional impurities. This paper successfully modifies the electrical properties of ambipolar silicon junctionless nanowire transistors (Si-JNTs) using the reducing properties of ammonia (NH₃) for selective detection. Physisorption of NH₃ induces a dual response in both <i>p</i>- and <i>n</i>-type conduction channels of ambipolar Si-JNTs, significantly altering current and key parameters, including the "on" current (<i>I</i>_on), threshold voltage (<i>V</i>_th), and mobility (μ). NH₃ interaction increases conduction in the <i>n</i>-channel and decreases it in the <i>p</i>-channel, acting as an electron donor and hole trap, as supported by Density Functional Theory (DFT) calculations. This provides a pathway for charge transfer and ″pseudo″ molecular doping in ambipolar Si-JNTs. This NH₃-mediated molecular doping and conduction modulation in Si transistor enabled, for the first time, the electrical detection of gaseous NH₃ at room temperature across a wide concentration range (200 ppb to 50 ppm), achieving high sensitivity (200 ppb) and precise selectivity under ultraviolet (UV) light. UV illumination dynamically modulates current and reveals distinct sensing features in the <i>p</i>- and <i>n</i>-channels of the dual-responsive Si-JNTs. The ambipolar Si-JNT sensor exhibits a fast response time of 1.91 min for 0.8 ppm of NH₃ in the hole conduction channel and a high sensitivity of 80% for 0.8 ppm of NH₃ in the electron conduction channel. This dual-channel approach optimizes sensor performance by leveraging the most responsive parameters from each channel. Furthermore, the ambipolarity of Si-JNTs broadens the parameter space for developing a multivariate calibration model, enhancing the selectivity of Si-JNT sensors for NH₃ detection.