Mixed Dimensional Heterostructures

Mixed Dimensional Heterostructures

People:  Chengmei Zhong, Jack Olding, Suyog Padgaonkar

Sunlight provides more energy in 1 hour than all humans use in a year, eclipsing all other renewable and nonrenewable energy sources combined in terms of theoretical, extractable, and technical potential. However, the combined output of all current solar photovoltaics caps at 227 gigawatts, which only meets 1.5% of global energy demand at full capacity. Furthermore, the economic incentive of solar photovoltaics lags far behind fossil fuels, necessitating strong efforts to lower the cost of solar energy.

One promising approach lies in the use of mixed dimensional (MD) van der Waals (vdW) heterostructure materials to replace the current market leader in solar cell materials, crystalline silicon, which inherently has weak absorption near its band edge in the near-IR region that limits its efficiency, especially in thin films. MD vdW heterostructures are comprised of a layered 2D material that interacts through vdW forces with a material of a different dimensionality. MD vdW heterostructures can be incorporated into p-n semiconductor heterojunctions, which show promise for photovoltaic applications when materials are carefully selected to optimize the charge transfer processes.

In collaboration with the Hersam and Lauhon groups, we are interested in the synthesis and study of the photophysics of mixed dimensional heterostructures. Below, we have outlined some examples of the mixed dimensional heterostructures that we have studied (or are studying):



Schematic of the monolayer MoS2–pentacene van der Waals heterojunction probed by transient absorption spectroscopy.

Schematic of exciton dynamics in the MoS2-only film and MoS2–pentacene heterojunction. (a) Summary of relaxation pathways for the monolayer MoS2 exciton in the absence of pentacene, based on the dynamics of the bleach of the B-exciton feature in the TA spectrum. (b) Mechanism of hole transfer (τ2 = 6.7 ps) and charge recombination (τ2 = 5.1 ns) in the MoS2–pentacene p–n heterojunction. The depletion region is shown by band-bending of pentacene at the heterojunction (band offset 1.1 V). The effect of band bending is negligible in atomically thin MoS2. The electrons and holes shown in MoS2 form an exciton with binding energy of ∼0.5 eV.































We recently studied devices with pentacene, a p-type 0D organic molecule electron donor, and MoS2, an n-type 2D transition metal dichalcogenide (TMD) acceptor,  Using ultrafast transient absorption spectroscopy, we discovered that this p-n junction is limited by the tendency of pentacene to devolve into two low energy triplet states on a sub-picosecond scale through singlet fission. Thus, the choice of pentacene as the n-type material limits the efficiency of charge-separation in potential optoelectronic devices involving pentacene –informing the design of new potential heterojunctions.


Schematic of a p-n junction composed of PTB7, an organic polymer, and CVD-grown MoS2 studied using ultrafast transient absorption measurements.

Different 0D p-type materials, such as QDs or organic polymers, that do not engage in this ultrafast charge carrier dissipation process may be more promising in creating a longer-lived charge-separated state to increase the overall photovoltaic efficiency of the device. For example, we are currently studying the use of traditional organic photovoltaic (OPV) polymers such as PTB7 with chemical vapor deposition (CVD)-grown MoS2. (Shown above)

This approach to the synthesis of different heterojunctions can be generalized to incorporate any semiconducting materials of differing dimensionality (0D/1D/2D) such as organic dyes, polymers, nanorods (NRs) and carbon nanotubes, which can be tuned to provide sensitivity across the ultraviolet (UV), visible, and near-infrared (NIR) spectral regions. When combined with 2D transition metal dichalcogenides (TMDCs) that exhibit band-like charge mobilities to yield a p-n junction, these MD heterojunctions have the potential in optoelectronic applications. Further optimization and rational design of these MD heterojunctions requires in-depth mechanistic studies of the photodriven charge carrier dynamics at the MD heterojunction interface. For example: simultaneous CT of both charge carriers between the materials generally yields the best device performance while other processes such as energy transfer (EnT), carrier trapping, and biexcitonic annihilation all provide significant competition, thus reducing device efficiency. To bolster the efficiency of these heterojunctions, changes in photophysical properties across the material interface need to be thoroughly investigated and understood. One fundamental question we are attempting to address is how changes in the heterojunction interface modify the rate (and yield) of charge-separation within these devices. One straightforward method to addressing this question is to vary dimensionality of one of the semiconductor materials.


(a) Energy level diagram of a CdSe NC/MoS2 heterojunction. (b) Photoluminescence (dashed) and absorption (solid) spectra of nearly isoenergetic CdSe NCs of varying dimensionality. (c) SEM image of CdSe NPLs deposited on exfoliated MoS2 with the adapted Langmuir-Blodgett method. Scale bar = 200 nm. (d) Photoluminescence map of CdSe NPLs on CVD-grown MoS2, where MoS2 flakes are outlined with dashed blue lines.


Colloidal nanocrystals (NCs) offer an easy route to modifying the dimensionality of a semiconductor. Above, we have outlined some work examining how 0D, 1D, or (quasi)-2D CdSe NCs interface with monolayer MoS2. The primary goal of this project is to address the following question: How do changes in the quantum confinement of CdSe NCs affect competition between CT and EnT in a CdSe NC/MoS2 heterostructure?