Quantum confinement of electrons underpins many devices, from quantum cascade lasers to single-photon emitters. It is usually engineered deliberately by growing thin films or heterostructures. In layered materials such as MoS2, quantum wells arise naturally when the number of layers is reduced. However, they are not expected in bulk crystals: electronic states should be three-dimensional. This work shows that even in a bulk sample, mechanical exfoliation can leave a thin detached fragment that behaves as a quantum well. This connects the topic to studies of band topology and quantum phenomena in layered materials, where such effects are typically sought in monolayers or heterostructures.
Samples of synthetic Nb-doped MoS2 (Nb concentration ~2×$10^{17}$ cm^-3) were exfoliated once in ultrahigh vacuum at a pressure of 5×$10^{-8}$ mbar and immediately transferred to the ARPES chamber. Measurements were performed at the SOLARIS synchrotron (Kraków, Poland) on the URANOS beamline at a temperature of 110 K with photon energies of 40, 50, 70, and 100 eV. To enhance weak features, the curvature method was applied, which does not shift peak positions. Additionally, DFT calculations were performed with the QUANTUM ESPRESSO package using the PBE functional and Grimme's D3 van der Waals corrections. Slabs of 19 MoS2 layers with a vacuum gap of 30 Å were modeled, as well as configurations with the top three or four layers separated by distances of 0.5, 1, and 2 Å from the bulk.
At the Γ point, additional discrete states were found below the bulk valence band parabola. Their energy does not change when the photon energy is varied from 40 to 100 eV, proving the absence of kz dispersion and hence vertical quantum confinement. The number of states varies from 1 to 8 across different samples. The energy spacing between adjacent states decreases as the number of states increases, and the energy dependence on the index n is linear, not quadratic. This rules out the formation of a two-dimensional electron gas and points to a parabolic well potential. The well width L is estimated from the number of states: L ≈ 0.615·$n_{max}$ nm, giving a range from 0.6 nm ($n_{max}$=1) to 4.9 nm ($n_{max}$=8). DFT calculations show that when the top three layers are separated from the bulk by 1–2 Å, the energies of the corresponding three QCS shift and gain spectral weight, consistent with experiment.
This work demonstrates for the first time that mechanical exfoliation of bulk MoS2 can unintentionally create quantum wells from detached multilayers. This explains the previously observed but misinterpreted appearance of additional states in ARPES spectra. Unlike neighboring studies where quantum wells are created by controlled growth of thin films or heterostructures, here wells arise spontaneously due to weak van der Waals bonds. The result is important for interpreting photoemission data from bulk layered materials: observed features may not be defects but manifestations of quantum confinement. It also opens a simple route to creating quantum wells without complex epitaxy.
One could attempt to control the thickness of the detached multilayer by varying the force and method of exfoliation, or by transferring a predetermined number of layers onto a substrate. It would be interesting to investigate intersubband transitions in such wells for optical applications, since the parabolic potential gives equidistant levels. Also promising is to study how doping, strain, and temperature affect the QCS. Similar wells might arise in other layered materials such as WSe2 or black phosphorus.
The result will influence the interpretation of ARPES data for bulk transition metal dichalcogenides and other layered materials. It may also be useful in developing optoelectronic devices based on MoS2, where quantum wells are used to tune energy levels.
Conduct a systematic study of the dependence of the number of QCS on exfoliation parameters (force, speed, angle) and on the initial crystal thickness. Measure intersubband transitions by optical methods and compare with calculations.
This work touches on the fundamental problem of controlling quantum confinement in layered materials, where there is still no reliable way to predict how many layers remain after exfoliation. This is related to the broader challenge of tuning electronic properties through dimensionality and interfaces.
This section is addressed to the author of the paper and their colleagues. Neighbouring works were found by meaning-based search across our archive, not by word overlap. We do not review: we show the field around the work.
What the work does
Using ARPES, the study reveals quantum-confined states in bulk MoS2 after mechanical exfoliation. These states show no dispersion along kz, and their energies scale linearly with the state index, pointing to a parabolic quantum well potential. DFT calculations confirm that the well arises from the detachment of the top layers from the bulk.
Where its strength is
The work uncovers an unexpected mechanism for creating quantum wells in layered materials, linking experimental data to a parabolic potential model and opening a path toward controlled use of exfoliation-induced wells.
Where it could go next
Worth trying
Why this may matter
Understanding unintentionally formed quantum wells in layered materials could aid in developing new methods for creating quantum devices without complex growth processes, as well as in interpreting unexpected features in ARPES spectra of other layered materials.
Nearby works in our archive
The archive is sparse around this work: the closest review is only 0.50, against a usual 0.64. It looks like few have drilled nearby.
The minimum quantum well width was 0.6 nm, comparable to the diameter of a DNA molecule (about 2 nm), yet it accommodates just one layer of MoS2.