Viruses such as SARS-CoV, MERS-CoV, and SARS-CoV-2 have posed serious threats to humanity in recent decades, with the COVID-19 pandemic wreaking havoc on both health and the economy. Early detection of these viruses is crucial for mitigating their effect on the immune system. In this study, we propose a dielectric-modulated biosensor realized using junctionless negative capacitance (NC) FinFET with strained silicon (StrS) integration for label-free electrical detection of neutral and charged biomolecules. The biosensor design integrates StrS within the source/drain (S/D) extension region, characterized by a Gaussian doping profile. The device employs a uniaxial stress of 1.4 GPa applied along the channel direction, which modulates the band structure to enhance carrier mobility. Additionally, a ferroelectric (FE) layer is incorporated in the gate stack, introducing a NC effect. This configuration reduces the OFF current while increasing the ON current, thereby improving the selectivity of the sensor. A 'T-shaped' metal support is incorporated, featuring cavities on both sides for the biomolecule's accommodation. Our investigation focuses on the label-free detection of various biomolecules characterized by their equivalent dielectric constant (K), e.g. Streptavidin (K = 2.1), Biotin (K = 2.63), APTES (K = 3.57), Keratin (K = 8), etc. Discrimination between different biomolecules is achieved by analyzing variations in threshold voltage (VTH), OFF current, and ION/IOFF ratio as sensing metrics. The limit of detection (LoD) of the proposed sensor was meticulously estimated through a machine learning approach, achieving an impressive sensitivity with a calculated value of 0.47, quantified in terms of the 'K' of the biomolecules. A comprehensive analysis also involves the sensitivity and selectivity analysis with different FE thicknesses. Our study found that Pyridine (K = 12) biomolecules have the highest VTH and OFF current sensitivity (189.57% and 85.51%, respectively), as well as the highest VTH and OFF current selectivity over Streptavidin (1.3 and 0.811, respectively) at FE thickness of 1.8 nm. Furthermore, this paper explores the impact of low-frequency noise (i.e. flicker noise) on the noise power spectral density (PSD) of the proposed biosensor in the presence of different biomolecules. Our findings reveal that an increase in the gate coupling capacitance corresponds to the rise in K of biomolecules, leading to a decrease in noise PSD ( SID/ID2) and a subsequent reduction in noise. Using well-calibrated TCAD models, the acquired results reveal that incorporating StrS leads to a similar to 41% increase in the ON current over its baseline case.
