1/f Noise Characterization of Bilayer MoS₂ Field‐Effect Transistors on Paper with Inkjet‐Printed Contacts and hBN Dielectrics

Abstract

1/f noise represents the dominant source of noise in the low‐frequency range in several physical systems, including field‐effect transistors. Its investigation can provide very important information on the fabrication process, highlighting the steps that are more prone to the introduction of defects. Here, 1/f noise in bilayer MoS2 transistors on paper with inkjet‐printed Ag contacts and hBN dielectric is investigated. These devices are promising building blocks for future low‐cost, flexible, and easily recyclable disposable electronics. The analysis of 1/f noise, performed following Hooge's empirical approach, results in a Hooge parameter ≈1–10, which is comparable to those reported for bilayer MoS2 transistors on SiO2. The present results indicate that the noise properties of the investigated devices are stable against substrate bending and are mainly determined by the printing of the dielectric, while not being sensibly affected by the use of the paper substrate. These results are promising for the further development of low noise 2D material‐based flexible electronics on paper.

Physics-based bias-dependent compact modeling of 1/f noise in single- to few-layer 2D-FETs

“…LFN in 2D-FETs has been thoroughly inspected focusing mainly on its experimental characterisation. 33–51 In more detail, LFN has been researched in MoS 2 , 33–37,39,44,46,47,49–51 MoSe 2 , 38 WS 2 , 40 MoTe 2 , 41,42 ReS 2 45,48 and BP 44 FETs, respectively while single-, 33,35,36,39,40,43,46,49,51 bi- 34,36,37,45,49,50 and a few- 37–40,44,47,48,51 layer (SL, BL, and FL, respectively) devices have also been investigated as well as liquid-gated 48 and paper-printing 50 2D-FETs. Experimental random telegraph noise (RTN) has been recorded only at very low temperatures in 2D-FETs so far, 33 while it is usually the case in small devices, where the number of traps is small; 25 RTN is out of the scope of the present work.…”

“…LFN in 2D-FETs has been thoroughly inspected focusing mainly on its experimental characterisation. [33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51] In more detail, LFN has been researched in MoS 2 , [33][34][35][36][37]39,44,46,47,[49][50][51] MoSe 2 , 38 WS 2 , 40 MoTe 2 , 41,42 ReS 2 45,48 and BP 44 FETs, respectively while single-, 33,35,36,39,40,43,46,49,51 bi- 34,36,37,45,49,…”

“…can be diminished. A hexagonal boron nitride (hBN) layer between the channel and the dielectric has been evidenced to improve the 1/f noise, 41,48,50 while the latter can also be decreased after the application of polymer electrolyte encapsulation. 46 1/f noise from the contacts can also contribute to total noise, especially in the high current regime [36][37][38] and vacuum annealing might be a decent way to eliminate it.…”

“…43 Despite the aforementioned advancements in the experimental characterisation of 1/f noise in 2D-FETs, compact modelling developments, appropriate for circuit designs, are still in an embryonic phase. According to bibliography, [33][34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51] the wellknown ΔN, Δμ, and ΔR effects generate 1/f noise in 2D-FETs similarly to that in MOSFETs [25][26][27][28][29][30][31][32] and graphene FETs (GFETs). 52 1/f noise increases as the number of layers decreases 49 and becomes maximum at SL 2D-FETs where the ΔN effect prevails.…”

“…A much more intense ΔN CMF contribution to the 1/f noise has been recorded in 2D-FETs in comparison with MOSFETs, 27,[30][31][32] which again has been modelled by applying the simplified ∼(1 2 (g m /I D ) 2 concept, 35,40,41,43,46 where α c is the Coulomb scattering coefficient (in V s C −1 ), μ the low-field mobility of the device (in cm 2 (V s) −1 ) and C t,b the top(back)dielectric capacitance per unit area (in μF cm −2 ). Some other works follow exclusively the Δμ model to describe the behaviour of the1/f noise experiments 33,38,47,50 while in some occasions both ΔN and Δμ effects are required to successfully interpret the 1/f noise measurements. 39,42 More complicated ΔN 34 and Δμ 47 models have also been proposed but they are based on derivations not suitable for circuit design while they also employ linear region approximations.…”