TL;DR
This paper uses a novel photocurrent spectral atomic force microscopy (PCS-AFM) technique to map the electrical and optoelectronic properties of molybdenum disulfide (MoS₂) nanosheets as a function of their layer number. They reveal a complex interplay between tunneling barriers and light absorption that dictates photoconductivity, offering crucial insights for designing next-generation 2D electronic and optoelectronic devices.
What Problem Does This Paper Address?
Atomically thin molybdenum disulfide (MoS₂) is a promising material for electronics and optoelectronics due to its unique 2D quantum confinement effects. However, a critical gap in understanding exists regarding how its optoelectronic properties, particularly at metal-semiconductor junctions, scale with the number of layers. Furthermore, the spatial uniformity of these properties across a MoS₂ flake and how they are affected by the nanoscale interface with metal contacts remained largely unaddressed. This lack of understanding hinders the rational design and optimization of MoS₂-based devices.
Methods
The researchers employed a sophisticated technique called Photocurrent Spectral Atomic Force Microscopy (PCS-AFM). This method combines the high spatial resolution of Atomic Force Microscopy (AFM) with the ability to measure both dark current and photocurrent under varying illumination wavelengths. Specifically:
- Conductive AFM (C-AFM): Used to map the current flow between a conductive PtIr-coated AFM tip and the MoS₂ nanosheet under applied bias voltages. This allowed for the characterization of dark current transport.
- Photocurrent Spectral AFM (PCS-AFM): Extended C-AFM by illuminating the sample with a supercontinuum laser filtered to specific wavelengths. This enabled the measurement of photocurrent and its spectral dependence.
- Spatially Resolved I-V Curves: By sweeping the bias voltage across different regions of the MoS₂ flake, thousands of local current-voltage (I-V) curves were generated simultaneously with topographic imaging. This provided unprecedented detail on spatial variations.
- Raman Spectroscopy: Used to accurately determine the number of MoS₂ layers in different regions of the sample.
- 2D Fourier Analysis: Applied to topographic and current images to analyze spatial frequency variations and understand the nature of the tip-sample contact.
This approach is novel in its ability to simultaneously probe layer-dependent electronic and optoelectronic properties with nanoscale resolution on a single, continuous MoS₂ flake, avoiding the complexities and potential damage associated with fabricating individual devices.
Datasets
The paper does not describe the use of pre-existing datasets for training or evaluation. Instead, the authors generated their own experimental data by performing measurements on exfoliated MoS₂ nanosheets of varying layer thicknesses (from 1 to 20 layers) on an Indium Tin Oxide (ITO)-coated glass substrate.
Key Insights and Ideas
- Fowler-Nordheim Tunneling Dominance: Dark current measurements revealed a diode-like behavior well-described by Fowler-Nordheim (FN) tunneling. This suggests that carrier transport at the nanoscale MoS₂-metal junction is primarily governed by tunneling through a barrier.
- Linear Scaling of Tunneling Barrier: The effective tunneling barrier height was found to scale linearly with the number of MoS₂ layers. This provides a predictable way to tune the electronic properties by controlling layer thickness.
- Non-monotonic Photoconductivity: The photocurrent response showed a surprising trend: it decreased for the first few layers (up to 4 layers) but then increased again for thicker layers. This is a key finding that challenges simple expectations.
- Competition Between Barrier and Absorption: The authors propose a model where the non-monotonic photoconductivity arises from a competition between an increasing tunneling barrier (which hinders carrier collection) and increasing light absorption (which generates more carriers) as the layer number increases.
- Direct vs. Indirect Bandgap Impact: Monolayer MoS₂ (1L), with its direct bandgap, exhibited the highest photoresponse due to more efficient photoexcitation, even though its tunneling barrier was low.
- Spatial Inhomogeneity: 2D Fourier analysis revealed spatial variations in the MoS₂/substrate contact that exceeded those imposed by the substrate itself, suggesting subtle crinkling or imperfect contact at the nanoscale.
Key Results
- Effective Barrier Height: The effective barrier height for FN tunneling was found to be approximately 0.61 eV for a monolayer, scaling linearly with layer number.
- Photocurrent Minimum: A minimum in photocurrent was observed around 4 layers of MoS₂, attributed to the interplay of barrier height and light absorption.
- Layer Number Dependence: The photoresponse ratio (normalized to the 4L value) was approximately 15:4:2:1:4 for 1L, 2L, 3L, 4L, and 10L, respectively, highlighting the significant impact of layer number.
- Barrier Symmetry: The 1L MoS₂ regions showed symmetric photoresponse with respect to bias voltage, while 2L, 3L, and 4L regions exhibited noticeable asymmetry.
- Sublinear Power Dependence: Photocurrent showed a sublinear dependence on incident laser power (exponents ranging from 0.56 to 0.67), attributed to exciton-exciton annihilation (EEA) in atomically thin MoS₂.
Limitations and Future Work
- Model Simplifications: The proposed model for photoresponse assumes an effective bias voltage increase due to photoexcitation and a change in effective barrier height. While it captures the trend, further refinement might be needed for precise quantitative predictions.
- Tip-Sample Contact Variability: While efforts were made to ensure stable tip-sample contact, the paper acknowledges that small amounts of crinkling in the MoS₂ flake under the scanning tip can lead to more variable electrical contact and current inhomogeneity.
- Future Work: The authors suggest that their findings provide a fundamental understanding of layer-number-dependent electronic and optoelectronic behavior at the nanoscale MoS₂-metal junction. This knowledge can be leveraged for the selection of appropriate metals and the precise control of MoS₂ layer numbers to achieve a wide range of electrical property modulation for future device applications. Further investigation into the role of exciton dynamics and their interaction with tunneling barriers could also be fruitful.
Summary updated