The Intricate Love Affairs between MoS₂ and Metallic Substrates

Abstract

Mechanical exfoliation of 2D materials yields high-quality crystals popular with researchers in fundamental scientific disciplines but its scalability is severely limited. This method generates 2D monolayers tens or hundreds of microns in lateral sizes on most substrates, often after an elaborate surface treatment. [1] Gold-mediated exfoliation of chalcogenides, chlorides, thiophosphates, black phosphorus, and black arsenic, with a robust control of the near-unity monolayer yield at a millimeter-/centimeterscale, has recently emerged as a viable solution to the scalability issues, [2] and has been adopted in various branches of applied research and engineering. [3] In the case of transition metal dichalcogenides (TMDCs), the root of the preferential mono layer exfoliation has been attributed to the strong interactions between gold and chalcogenides, which have been explored in different facets of science for decades. [4] However, it has recently been shown that the interaction between TMDCs and Au is non-covalent and van der Waals (vdW) in its nature, inferred from the sizeable S-Au equilibrium distance (3.5 Å) and binding energies in the Au-MoS 2 heterostructure. [2c,5] The vdW interaction therefore facilitates the transfer of the TMDC monolayers onto non-metallic substrates, which restore their semiconducting characteristics exploitable in optoelectronics, photovoltaics, and related themes. [2b] The polymer-free nature of this transfer, which leaves surfaces free from residual contamination, is of significant advantage also. [6] Despite these research efforts, it is currently unknown whether this method can also be applied to other metals, predicted to exhibit even stronger binding with MoS 2 than Au. [7] Here, we study the ability of different metallic substrates to exfoliate large-area monolayer MoS 2. We find that gold is by far the best substrate, outperforming all other metals by at least two orders of magnitude in terms of the lateral size of the MoS 2 , thanks to the unique ability of Au to resist oxidation and the sizeable interfacial strain in the Au-MoS 2 heterostructure. A moderate exfoliation yield is achieved for other precious metals, including Pt, Pd, and Ag, while hardly any exfoliated material is found on base metals, including Cu, Ni, Co, Cr, and Ti, which suffer from significant oxidation of their surface upon exposure to air. A correlation between the maximum lateral Mechanical exfoliation yields high-quality 2D materials but is challenging to scale up due to the small lateral size and low yield of the exfoliated crystals. Gold-mediated exfoliation of macroscale monolayer MoS 2 and related crystals addresses this problem. However, it remains unclear whether this method can be extended to other metals. Herein, mechanical exfoliation of MoS 2 on a range of metallic substrates is studied. It is found that Au outperforms all the other metals in their ability to exfoliate macroscale monolayer MoS 2. This is rationalized by gold's ability to resist oxidation, which is compromised on other metals a...

Can Metals Other than Au be Used for Large Area Exfoliation of MoS2 Monolayers?

“…Instead, uniform strain is the determining factor and other metals that do not produce such uniform strain are unable to exfoliate large-area ML TMDs. [26,27] Given that there are more than 800 layered crystals are predicted to form ML flakes by exfoliation [28] of which only ≈40 2D materials have been realized, [12] a study and demonstration of other metalmediated large-area exfoliation is important for the practical realization of many future 2D materials.…”

Spectroscopic Ellipsometry Investigation of Au‐Assisted Exfoliated Large‐Area Single‐Crystalline Monolayer MoS2

“…[43] In contrast, the broadening and splitting of the E 1 2g mode indicate the degree of heterogeneity of strain distribution. [22] In our measurement, the E 1 2g mode is broadened yet not split, and it is still nearly symmetric and can be fitted with a single peak. This is different from the cases of 1L MoS 2 on Ag and Pt films, [22,43] which implies uniform strain distribution in MoS 2 caused by regular Au clustering.…”

“…[22] In our measurement, the E 1 2g mode is broadened yet not split, and it is still nearly symmetric and can be fitted with a single peak. This is different from the cases of 1L MoS 2 on Ag and Pt films, [22,43] which implies uniform strain distribution in MoS 2 caused by regular Au clustering. As for the out-of-plane vibration A 1g mode at 402.8 cm À1 , it splits into two peaks located at 395.6 and 402.6 cm À1 , which is consistent with the observation of Velický et al [22,44] The higher frequency peak of the A 1g mode of 1L MoS 2 on Au is almost identical to that on SiO 2 /Si (approximately regarded as the corresponding Raman frequency of bare 1L MoS 2 ), indicating that there may be regions where the 1L MoS 2 is not in direct contact with Au.…”

“…This is different from the cases of 1L MoS 2 on Ag and Pt films, [22,43] which implies uniform strain distribution in MoS 2 caused by regular Au clustering. As for the out-of-plane vibration A 1g mode at 402.8 cm À1 , it splits into two peaks located at 395.6 and 402.6 cm À1 , which is consistent with the observation of Velický et al [22,44] The higher frequency peak of the A 1g mode of 1L MoS 2 on Au is almost identical to that on SiO 2 /Si (approximately regarded as the corresponding Raman frequency of bare 1L MoS 2 ), indicating that there may be regions where the 1L MoS 2 is not in direct contact with Au. [44] The lower frequency of the A 1g mode is mainly caused by the charge in the 1L MoS 2 transferred from Au in intimate contact as the A 1g mode possesses a strong doping dependence.…”

“…Fortunately, Au-film-assisted exfoliation of macroscopic TMDC monolayers with near-unity yield was introduced very recently due to the excellent oxidation resistance, large uniform interfacial strain of the Au film as well as the appropriate relative difference (%0.12 eV per unit cell) between the interfacial binding energy of the Au-MoS 2 heterostructure and the interlayer binding energy of MoS 2 . [20][21][22][23][24] This novel technology makes it possible to conduct an SE study of the as-exfoliated single-crystalline monolayer MoS 2 , which is exactly the motivation of this work.…”