Layered van der Waals (vdW) materials, readily exfoliated to single-unit-cell thicknesses, exhibit heightened surface sensitivity. Environmental adsorbates and substrates significantly influence their properties. Surface contamination is a major obstacle in manufacturing vdW heterostructures, hindering interlayer adhesion and limiting achievable configurations. Many applications of vdW materials occur under ambient conditions, making contamination unavoidable. However, our understanding of the chemical composition of this contamination remains limited, often described only generally as 'hydrocarbon and adsorbed water'. Contaminants can significantly hinder intrinsic material properties, impacting wetting behavior and electrochemical activity. Previous studies have reported one-dimensional (1D) ordered structures, such as parallel stripes (4–6 nm period), on graphitic material surfaces. Explanations have varied, suggesting self-assembled molecular nitrogen or monolayers of organic chain-like molecules. However, a consensus on the chemical composition and origin of these structures remains elusive. While some studies propose alignment along zigzag or armchair directions of the hexagonal lattice, and anisotropic friction is attributed to a contamination layer, the precise nature of the contaminating molecules hasn't been definitively determined. Anisotropic friction domains, displaying similar properties across various vdW materials (graphene, hBN, MoS₂, WS₂), have been reported but their origin is unclear; some theories propose periodic strain, though lacking direct evidence. This study aims to clarify the chemical composition and structure of the ubiquitous contamination layer, and its connection to anisotropic friction domains.

A significant body of research has documented the observation of 1D ordered structures, often appearing as parallel stripes with a period of 4–6 nm, on the surfaces of various graphitic materials. Early studies suggested that these structures might result from the self-assembly of molecular nitrogen. However, subsequent investigations proposed alternative explanations, pointing towards monolayers of organic chain-like molecules as the building blocks of these stripe patterns. Despite the numerous observations, a definitive consensus on the chemical composition and the fundamental origin of these molecular layers remains elusive. Researchers have extensively explored the anisotropic friction observed on graphene and other 2D materials. The commonly observed parallel stripes are often linked to the anisotropic frictional properties. Several studies have indicated that a contamination layer on the surface is the source of this anisotropy. However, the precise nature of the molecules composing this ordered surface layer remained undetermined. Furthermore, while some studies suggested periodic strain in the uppermost layer as an explanation for anisotropic friction, this hypothesis lacked direct experimental confirmation. The overall lack of clarity regarding the chemical composition of the contamination layer and the absence of conclusive evidence for periodic strain ripples highlight the need for further investigation into the root causes of anisotropic frictional domains on vdW materials.

The researchers employed a range of surface-sensitive techniques to investigate the composition and structure of the contaminant layer on various vdW materials. Atomic force microscopy (AFM), both in PeakForce QNM® and contact modes, was used to image surface topography and friction properties. Low-temperature scanning tunneling microscopy (STM) at 9 K provided atomic-resolution images of the contaminant layer. Grazing-angle infrared (IR) spectroscopy was used to identify the vibrational modes of the molecules, enabling their chemical characterization. X-ray photoelectron spectroscopy (XPS) provided information on the elemental composition of the surface. Finally, ellipsometry was used to monitor the growth kinetics of the contaminant layer over time. Sample preparation involved exfoliation of graphite, few-layer graphene, hBN, and MoS₂ flakes onto Si/SiO₂ substrates using the standard scotch tape method. Samples were stored under ambient laboratory conditions for 4–5 days to allow for contaminant layer formation. For IR, XPS, and ellipsometry measurements, bulk HOPG crystals were used, with pristine surfaces obtained by mechanical cleaving. The presence of the contaminant layer and its characteristic stripe structure was always verified using AFM before other measurements. To compare with the naturally occurring contaminant layer, the researchers also prepared samples with artificially deposited monolayers of dotriacontane (C₃₂H₆₆) and arachidic acid (C₁₉H₃₉COOH) using vapor deposition techniques. DFT calculations were performed using the VASP software package to model the arrangement of alkanes on graphite surfaces, simulating STM images to compare with the experimental results. The DFT calculations considered both edge-on and flat-on orientations of the alkanes. AFM measurements were performed using a Bruker Multimode 8 AFM in various imaging modes (contact, lateral force, and PeakForce QNM). STM measurements were conducted using an RHK PanScan Freedom microscope under ultra-high vacuum conditions at both room temperature and 8.9 K. IR spectroscopy used a Bruker IFS 66/v vacuum FTIR spectrometer with a grazing angle reflection accessory. XPS measurements were conducted using a Kratos XSAM 800 XPS instrument. Finally, ellipsometry measurements were performed using a Woollam M-2000DI rotating compensator ellipsometer.

AFM imaging revealed ubiquitous parallel stripes (3.0-5.2 nm period) on various vdW materials (graphite, hBN, MoS₂) after several days of ambient exposure, absent on freshly prepared samples. These stripes displayed domain-like behavior with 60° rotational symmetry. Low-temperature STM at 9K provided atomic-resolution images of the contaminant layer, revealing linear molecules arranged in stripes. The molecule length, measured to be approximately 3.27 ± 0.15 nm, suggests normal alkanes with 24–25 carbon atoms. Grazing-angle infrared spectroscopy identified the characteristic C-H stretching and scissoring modes of alkanes, confirming the presence of saturated hydrocarbons. Importantly, other functional groups (–COOH, –OH, halides) were absent. Tunneling conductance (dI/dV) measurements supported this identification. Comparing IR and STM data with a vapor-deposited dotriacontane (C₃₂H₆₆) monolayer further confirmed the identification of normal alkanes in the contaminant layer. The contaminant layer appeared as a monolayer with a quasi-periodic 1D superlattice along the stripes, with a period of 1.7–2.5 nm. DFT calculations of alkane chains on a bilayer graphene surface, considering both edge-on and flat-on orientations, qualitatively reproduced the main features of the experimental STM images, particularly the quasi-periodic superlattice and the observed inter-molecule distances. The edge-on orientation was found to be more consistent with the experimental data. The study found a direct correlation between the self-organized stripes of the alkane layer and anisotropic friction domains. The 60° rotational symmetry of the friction domains is directly linked to the three zigzag directions of the underlying hexagonal lattice of the vdW materials. Controlled manipulation of friction domains using AFM was demonstrated, showing that the domains can be reliably patterned, independent of host material thickness. Ellipsometry measurements revealed that an unordered contaminant layer initially forms, subsequently replaced by the self-organized alkane monolayer within 10–14 days.

The identification of mid-length normal alkanes (20–26 carbon atoms) as the primary constituents of the ubiquitous contaminant layer on various vdW materials addresses a long-standing gap in our understanding of surface contamination. The prevalence of these alkanes in the environment, ranging from natural minerals to byproducts of combustion and lubricants, explains their presence on these materials. The self-organization of these alkanes into an ordered monolayer explains the previously reported stripe structures and anisotropic friction domains. This self-organization likely stems from a balance of molecule-molecule and molecule-substrate interactions, with the longer-chain alkanes gradually displacing shorter chains over time. The finding that friction anisotropy is not dependent on the thickness of the vdW material excludes theories based solely on intrinsic strain ripples. The ability to controllably manipulate and pattern friction domains by AFM opens exciting possibilities for applications requiring tailored frictional properties. The results suggest that this type of alkane contamination might be universally present on vdW materials with hexagonal lattices, having significant implications for device performance and stability.

This study definitively identifies a self-organized monolayer of mid-length normal alkanes (C₂₀-C₂₆) as the main component of ubiquitous surface contamination on several vdW materials. This contamination layer is responsible for the previously reported stripe structures and anisotropic friction domains. The ability to precisely manipulate these domains by AFM offers a route toward controlling the frictional properties of vdW materials. Future research could focus on exploring the impact of this contamination on other material properties and investigating potential strategies for mitigating or exploiting this ubiquitous phenomenon in various technological applications. The formation and stability of the alkane layers themselves present intriguing opportunities for the design of protective or functional coatings.

While this study provides a comprehensive understanding of the composition and structure of the ubiquitous contaminant layer, it primarily focuses on specific vdW materials with hexagonal lattices. Further research is needed to confirm the universality of this contamination on other vdW materials with different lattice structures. Additionally, while the study successfully manipulated friction domains, the long-term stability of these artificially created patterns under various environmental conditions could be further investigated. The study also mainly focuses on characterizing the contaminant layer formed under laboratory conditions. Further investigations are warranted to assess the composition and self-organization behavior of these alkanes under various ambient environments.