[Paper Review] Ultra-sensitive Parity-Time Symmetry based Graphene FET (PTS-GFET) Sensors

Lu Ou, Shaolin Liao|arXiv (Cornell University)|Jul 9, 2020

Mechanical and Optical Resonators29 references4 citations

TL;DR

This paper proposes a novel ultra-sensitive Parity-Time symmetry-based Graphene FET (PTS-GFET) sensor that leverages the exceptional point (EP) of PT-symmetric RLC resonators and the tunable conductance of a GFET to detect gas concentrations below 50 ppb. By adjusting the gate voltage to maintain the system at the EP, the applied voltage directly indicates the gas concentration, achieving high sensitivity through EP-enhanced frequency splitting and noise suppression.

ABSTRACT

A novel ultra-sensitive Parity-Time symmetry based Graphene FET (PTS-GFET) sensor is studied for gas concentration detection. The PTS-GFET sensor effectively integrates the sensitivity of the PT symmetry around its Exceptional Point (EP) and the tunability of the GFET conductance. The change of GFET conductance with the gas concentration can be brought back to the EP of the PTS-GFET by tuning the gate voltage on the GFET. Thus, the applied gate voltage indicates the gas concentration. The minimum detectable gas concentration has been derived and estimated based on the experimental data, which shows that PTS-GFET can detect gas concentration below 50 ppb.

Motivation & Objective

• To develop a highly sensitive gas sensor capable of detecting sub-ppb-level gas concentrations.
• To overcome the limitations of low-frequency GFET sensors, particularly $1/f$ noise, by operating at electromagnetic frequencies.
• To integrate the extreme sensitivity of PT-symmetric systems near the exceptional point (EP) with the tunability of graphene FETs (GFETs).
• To enable real-time, quantitative gas concentration detection by correlating gate voltage adjustments with gas-induced conductance changes.

Proposed method

• Model the PTS-GFET as a pair of coupled RLC resonators, one with negative conductance ($-G_1$) and the other with tunable conductance ($G_2$) from the GFET.
• Use a Vector Network Analyzer (VNA) to monitor the reflection spectrum and identify the EP condition where $G_1 = -G_2$.
• Apply a feedback mechanism via gate voltage $V_G$ to dynamically tune $G_2$ and maintain the system at the EP despite changes in gas concentration.
• Utilize the quantum Hamiltonian formalism for coupled RLC resonators to derive the eigenfrequencies and analyze the system's sensitivity near the EP.
• Derive the normalized PT frequency splitting sensitivity $s_ ext{Ω}$ as a function of gate voltage noise and conductance fluctuations.
• Estimate the minimum detectable gas concentration using experimental data on $ rac{ ext{d}g}{ ext{d}V_G} $ and $ rac{ ext{d}g}{ ext{d}c_ ext{gas}} $, and noise parameters.

Experimental results

Research questions

• RQ1 Can PT-symmetry at the exceptional point (EP) be harnessed to enhance the sensitivity of graphene FET-based gas sensors?
• RQ2 How does the tunability of GFET conductance via gate voltage enable dynamic compensation for gas-induced conductance changes?
• RQ3 What is the theoretical minimum detectable gas concentration achievable when combining EP-based frequency splitting with GFET tunability?
• RQ4 How do gate voltage noise and conductance fluctuations limit the sensor's resolution and detectability?

Key findings

• The PTS-GFET sensor achieves a minimum detectable gas concentration of less than 50 ppb under optimal conditions.
• The system maintains operation at the exceptional point (EP) by dynamically tuning the GFET gate voltage to compensate for gas-induced conductance changes.
• The normalized PT frequency splitting sensitivity $ s_ ext{Ω} $ is derived as $ rac{ ext{d} ext{Re}( ext{Ω}_+ - ext{Ω}_-)}{ ext{d} ext{Δ}g} imes rac{ ext{Δ}g}{ ext{σ}} $, with $ ext{σ} $ representing conductance noise.
• With a gate voltage noise standard deviation $ ext{σ}_{V_G} = 1 $ mV and $ s_ ext{Ω} < 0.02 $, the sensor achieves sub-50 ppb detection for NO₂.
• The sensitivity is enhanced by operating at electromagnetic frequencies, eliminating $1/f$ noise and enabling high-quality factor filtering.
• Theoretical modeling using the quantum Hamiltonian $ ext{ℋ}_{PT} $ confirms the system’s behavior near the EP and enables precise sensitivity analysis.