Small Structures - 2021 - Tang - Recent Advances in Synthesis and Study of 2D Twisted Transition Metal Dichalcogenide

REVIEW
www.small-structures.com
Recent Advances in Synthesis and Study of 2D Twisted

Transition Metal Dichalcogenide Bilayers
Bijun Tang, Bingyu Che, Manzhang Xu, Zhi Peng Ang, Jun Di, Hong-Jun Gao,
Haitao Yang,* Jiadong Zhou,* and Zheng Liu*
due to their potential applications in vari-

In recent years, twisted 2D transition metal dichalcogenides (TMDs) have ous fields such as optics, electronics, and
attracted tremendous research interest thanks to their intriguing properties that catalysis.[1–3] Numerous 2D materials have
are essential in developing future electronic and photoelectronic devices in this been developed in the past decade, rang-
ing from insulators to superconductors.
modernizing era. The aim of this review is to introduce recent developments in
However, most of the works reported so
the preparation of twisted TMD (t-TMD) bilayers, their properties investigated by far are limited to the monolayer sample,
photoluminescence (PL) spectroscopy, Raman spectroscopy, and other techni- while overlooking the bilayers, though
ques, as well as the related distinctive physical phenomena. First, various countless fascinating physical phenomena
strategies for synthesizing t-TMD bilayers via the two-step stacking method and have been observed in the past few
one-step chemical vapor deposition method are reviewed. Then, the commonly years.[4–6] For example, graphene bilayers
and graphene/boron nitride (BN) heterobi-
used characterization techniques in probing the interlayer coupling between the
layers are found to possess various intrigu-
twisted layers are introduced. Moreover, the intriguing physical properties ing physical properties such as Fermi
associated with twisting such as interlayer excitons and correlated electronic velocity renormalization,[7] new van Hove
phases are summarized. Last but not least, challenges and future research singularities,[8,9] unconventional quantum
directions are briefly discussed in light of recent advances in the field. Hall effects, and Hofstadter’s butterfly
pattern.[10–12] These can be attributed to
the weak van der Waals (vdW) interaction
between the two layers. The interlayer
1. Introduction coupling in graphene depends sensitively on the interlayer twist
angle. In 2018, the “magic” angle of 1.1 in twisted bilayer gra-
Since the first successful isolation of graphene from graphite phene (TBG) was reported, which gives rise to exotic phenom-
in 2004, 2D materials have attracted considerable attention ena, including intrinsic unconventional superconductivity
and correlated insulator states.[13,14] This groundbreaking find-
ing immediately triggered substantial research interest in the
B. Tang, Dr. M. Xu, Z. P. Ang, Dr. J. Di, Prof. J. Zhou, Prof. Z. Liu study of twisted bilayer 2D materials.[15–18] We have looked
School of Materials Science and Engineering
Nanyang Technological University into the related literature and completed the publication statis-
Singapore 639798, Singapore tics, i.e., number of publications, as shown in Figure 1a. In
E-mail: [email protected]; [email protected] recent years (from 2010 to 2020), publications concerning twisted
B. Che, Prof. H.-J. Gao, Prof. H. Yang bilayers have increased rapidly, especially after the report of TBG,
Beijing National Laboratory for Condensed Matter Physics and Institute of showing a considerably growing interest.
Physics As an indispensable member of the 2D family as well as a
Chinese Academy of Sciences
Beijing 100190, China perfect complement to graphene, 2D transition metal dichalco-
E-mail: [email protected] genides (TMDs) have long been the center of research atten-
Prof. J. Zhou tion.[3,19–22] Their popularity can mainly be attributed to three
Key Lab of Advanced Optoelectronic Quantum Architecture and reasons: 1) 2D TMDs, with a generalized formula of MX2 (where
Measurement (Ministry of Education) M refers to a transition metal of groups 4–10 and X represents a
Beijing Key Lab of Nanophotonics & Ultrafine Optoelectronic Systems,
and School of Physics
chalcogen), exhibit versatile properties. 2) 2D TMDs have desir-
Beijing Institute of Technology able structures similar to graphene, which consist of covalent
Beijing 100081, China bonding in the constituent layers and vdW interactions with
Prof. Z. Liu neighboring layers, rendering them ideal for the exploration
CINTRA CNRS/NTU/THALES of monolayer or few-layer properties.[23] 3) Structures and
UMI 3288, Research Techno Plaza, 50 Nanyang Drive, Border X Block, properties of TMDs shared in common provide an additional
Level 6, Singapore 637553, Singapore
degree of freedom to study and tailor the resulting properties
The ORCID identification number(s) for the author(s) of this article of TMD heterostructures.[24–26] Development of twisted TMD
can be found under https://doi.org/10.1002/sstr.202000153. (t-TMD) bilayers not only provides an ideal platform to study
DOI: 10.1002/sstr.202000153 the fundamental physical problems such as the many-body
Small Struct. 2021, 2, 2000153 2000153 (1 of 17) © 2021 Wiley-VCH GmbH

www.advancedsciencenews.com www.small-structures.com
Figure 1. Overview of the recent study on twisted 2D materials. a) Publication trend regarding the study on twisted 2D bilayers from 2010 to 2020 (by Dec
29, 2020). Source: Web of Science. Search index: TOPIC: (twist) AND TOPIC: (bilayer). b) Schematic illustration of the preparation of t-TMD bilayers via
stacking method (route I) and one-step CVD method (route II).
phenomenon,[27–30] but also promotes the realization of future correlated electronic phases are highlighted. Finally, the prospects
electronic and photoelectronic devices.[31] Up to now, remarkable for future research on t-TMDs are offered.
progress has been achieved with the study on t-TMDs bilayers,
where ultrathin t-TMD bilayers have been successfully synthe-
sized and the associated interlayer coupling has been systemati-
2. Synthesis of 2D t-TMDs
cally studied.[32–37] For instance, researchers have found out that There are mainly two routes to produce t-TMD bilayers, which
the photoluminescence (PL) intensity ratio of the trion and the are the stacking method and the one-step CVD method, as shown
exciton of MoS2 is highly dependent on the twist angle of in Figure 1b. The stacking method can be regarded as a two-step
bilayers.[32] To put it another way, twisting the stacking configu- method as well, whereby the TMD monolayer is initially grown
ration of the bilayer is a promising method to alter the intrinsic or exfoliated on a substrate, followed by transferring the mono-
properties of the materials. layer onto another monolayer through a stacking or stamping
To investigate their intriguing properties, researchers have process (route I). By doing so, the twist angle of the bilayers
devoted great efforts to synthesize various t-TMDs by different can be manipulated. Having said that, it is possible to use this
methods. Generally, the stacking method and one-step chemical route to create a heterogeneous bilayer that combines two dis-
vapor deposition (CVD) method are the two major strategies tinct TMDs. The other method is relatively more straightforward.
adopted, which will be elaborated in the subsequent part. One-step CVD, as the name suggests, can grow twisted bilayers
Apart from the reliable production, property characterization in a one-time CVD process as it does not require growing, trans-
and examination are the other important research fields of ferring, and stacking of monolayers (route II). The main achieve-
t-TMDs. Unfortunately, there is no such review paper deliberating ments in the preparation of t-TMDs are summarized in Table 1.
the synthesis and the properties of t-TMDs. This Review endeav- The following parts will provide more discussions about the syn-
ors to provide an update-to-date overview of the recent progress in thesis routes in a detailed manner as well as point out their
the synthesis of t-TMD bilayers and the study of the associated respective pros and cons.
properties. The detailed synthesis methods are presented and
discussed. The characterization techniques, including PL spectros- 2.1. Stacking
copy, Raman spectroscopy, and transmission electron microscopy
(TEM), are also introduced. In addition, the intriguing physical The quality of the twisted bilayer largely relies on the monolayer
characteristics of t-TMDs such as interlayer excitons and obtained during the first stage of the procedure, which is used in

Table 1. Summary of the preparation of the t-TMD bilayers.
Materials Transfer method Annealing conditions Twist angle produced Ref.
Stacking MoS2 Dry transfer 400 C in Ar atm for 3 h θ ¼ 17.7 , 51.5 , and 84.7 [32]
MoSe2 Wet transfer 300 C in Ar (95%)/H2 (5%) θ ranging from 33.3 to 60.0 [37]
at atm pressure for 2 h
W-doped MoSe2 Wet transfer 300 C in Ar (95%)/H2 (5%) θ ranging from 40.8 to 60.0 [37]
at atm pressure for 2 h
WSe2/WS2 heterostructure Wet transfer 350 C in Ar flow Eight different twist angles [36]
(90 sccm, 10 Torr) for 3 h
Materials Precursors Growth conditions Twist angle produced Ref.
One-step CVD MoS2 20 mg MoO3, 7 mg S Sit at 105 C with 500 sccm N2 θ ranging from 0 to 60.0 [34]
for 1 h; ramp to 700 C at
15 C min1 and sit for
5–10 min with 10–15 sccm N2;
cool down naturally with
500 sccm N2
MoSe2 MoO3, Se powder TSe ¼ 300 C, T ¼ 700 C, θ ¼ 7 , 21 , and 25 [33]
65 sccm Ar and 5 sccm H2,
t ¼ 10 min
WS2 WO3 powder, 200 sccm pure Ar for 30 min θ ¼ 0 , 13 , 30 , 41 , 60 , and 83 [35]
0.1 g S powder before heating; 20 sccm Ar for
ramping to 1100 C at the rate
of 20 C min1; growing for
20 min
the subsequent stacking process. Traditionally, ultrathin or orientations.[37] Briefly, the MoSe2 monolayer was first grown
monolayer TMDs can be produced by a chemical or mechanical on a SiO2/Si substrate using the CVD method. The chip was
exfoliation method. However, the procedure can be troublesome cut into two pieces, and one of them was spin coated with poly-
as it is difficult to maintain uniformity throughout the sample. methyl methacrylate (PMMA). Etching of the silicon substrate
Moreover, the flakes produced are generally small in size.[38] To later took place by soaking the coated chip into a KOH solution,
address that, researchers have found that CVD is able to produce to obtain a freestanding PMMA/MoSe2 film. The film was then
large-area domains as the experiment parameters in CVD can be rinsed in a beaker filled with deionized (DI) water to remove the
finely controlled, thereby tuning the sample morphology. There KOH residue. Subsequently, the bilayer MoSe2 was fabricated by
have been many review articles describing the methodology of stacking the rinsed film onto the second half of the chip with the
CVD to synthesize high-quality and large-area TMD mono- originally grown monolayer MoSe2. Prior to the final annealing
layer.[39–43] The important roles played by various control param- process, the stacked chip was soaked in an acetone solution to
eters have been respectively discussed, such as reaction remove PMMA. The t-MoSe2 bilayer sample was eventually
temperature, reaction time, ramp rate, and carrier gas flow rate, obtained after annealing at 300 C in an argon (Ar) and hydrogen
providing useful insights into the controllable CVD synthesis of (H2) environment at atmospheric pressure for 2 h. The prepared
2D materials with desirable quality. With that being said, CVD t-MoSe2 possesses a broad range of twisted angles, from 33.2 to
experiment parameters play a vital role in determining the qual- 60 , as shown in Figure 2a. The same procedure has also been
ity of the material. The same strategy is applied in the first step of applied to prepare twisted-bilayer W-doped MoSe2 samples, with
the stacking method to produce t-TMDs. Apart from the CVD twisted angles ranging from 40.8 to 60 (Figure 2b). A similar
method, recent progress on the improvement of the exfoliation transfer process has also been used by Huang et al. to prepare a
technique, i.e., the Au-assisted mechanical exfoliation method, t-MoS2 bilayer.[32] Notably, both the top and bottom layer sub-
has opened up more possibilities for constructing staked t- strates were coated with PMMA. It was believed that the presence
TMCs. More than 40 types of single-crystalline monolayers have of PMMA in the bottom layer substrate may prevent the mono-
been obtained with the aforementioned exfoliation method, as layer from being peeled off and degraded during assembly in
reported by Huang et al.[44] When the monolayer is successfully water. However, it was hard to thoroughly remove the PMMA
grown or exfoliated on a substrate, it is then subjected to the sub- sandwiched in between by the conventional acetone method,
sequent transferring process. There have been many attempts to degrading the quality of the as-prepared twisted-bilayer samples.
produce t-TMDs by manual transferring or stacking of the mono- Apart from the twisted-bilayer homostructures, the wet trans-
layers. The stacking process can generally be divided into two fer method has also been applied to assist the study of twisted-
types, i.e., wet transfer and dry transfer. bilayer heterostructures. In 2016, Wang et al. reported the
Using the wet transfer method, Puretzky et al. obtained fabrication of WSe2/WS2 bilayer heterostructures with various
twisted MoSe2 (t-MoSe2) bilayers with multiple stacking twist angles through artificially stacking CVD-grown WS2 and

Figure 2. t-TMD bilayers obtained by two-step stacking or one-step CVD methods. a) OM images of bilayer MoSe2 with the twist angle of 60.0 , 59.6 ,
58.6 , 56.9 , 55.3 , 51.3 , 42.6 , and 33.2 (from top to bottom), respectively. b) OM images of bilayer W-doped MoSe2 with the twist angle of 60.0 ,
59.5 , 58.5 , 56.7 , 50.9 , 49.0 , 47.3 , and 40.8 (from top to bottom), respectively. Scale bar: 20 μm. a,b) Reproduced with permission.[37] Copyright
2016, American Chemical Society. c) OM image of WSe2 monolayer, WS2 monolayer, and the overlapping WSe2/WS2 bilayer on a SiO2/Si substrate.
d) OM image of the bilayer MoS2 with the twist angle of 51.5 . The red and blue triangles show the area of interest, which refers to the bottom and top
layers, respectively. c,d) Reproduced with permission.[32] Copyright 2014, American Chemical Society. e,f ) AFM images of a twisted-bilayer WSe2/WS2 and
the corresponding sectional analysis before and after annealing, respectively. e,f ) Reproduced with permission.[36] Copyright 2016, American Chemical
Society. g–i) OM images of CVD-grown bilayer MoS2 with twist angle 0 , 15 , and 60 , respectively. Scale bar: 10 μm. g–i) Reproduced with permission.[34]
Copyright 2014, Nature Research. j–o) OM images of CVD-grown twisted WS2 bilayers with twist angles 0 , 13 , 30 , 41 , 60 , and 83 , respectively. Scale
bar: 10 μm. j–o) Reproduced with permission.[35] Copyright 2015, Wiley-VCH.
WSe2 monolayers.[36] A typical constructed WS2/WSe2 bilayer procedures are involved: 1) After the substrate with grown
with a twist angle of 25.3 is shown in Figure 2c. As the twisted WS2 was spin coated with PMMA, it was subjected to a curing
bilayer was heterogeneous with distinct materials, the two process at 100 C for 15 min to stabilize the monolayer flakes and
materials were grown separately on different substrates using 2) after the WSe2/WS2 layers were stacked, the assembly went
different precursors and synthesis conditions. The rest of the through a preheating step prior to the final annealing process
procedures are rather similar, except that two additional to ensure the PMMA was thoroughly removed.

Instead of wet transfer, another research group has proposed transfer methods provide new opportunities for the preparation
the use of a dry transfer method to synthesize t-MoS2 (Figure 1 of large-area high-quality t-TMDs.
and 2d).[32] The beginning of the dry transfer process was similar
to the wet transfer method, from growing of monolayer MoS2,
2.2. One-Step CVD
cutting into two pieces until obtaining the free-floating
PMMA/MoS2 top layer. Next, the PMMA/MoS2 layer was put
CVD is one of the most important and popular techniques to
onto a polydimethylsiloxane (PDMS) elastomer that was attached
produce high-quality inorganic nanomaterials. To date, numer-
to a glass slide, followed by baking at 130 C to soften the PMMA.
ous 2D TMDs, as well as their heterostructures, have been con-
The assembly was then placed under a microscope with the MoS2
structed by the increasingly matured CVD methods.[19,25,39–43] It
layer facing down, and above the substrate with the bottom layer
is thus a promising method to obtain t-TMD bilayers without the
facing up. With the aid of a microscope, the top layer was
involvement of sample transfer.
adjusted and aligned with respect to the bottom layer. The glass
Using MoO3 and S powder as precursors, Liu and colleagues
slide was subsequently lowered down until touching the bottom
demonstrated the growth of bilayer MoS2 on mica, fused silica as
layer, and the PDMS was slowly lifted, leaving the top layer in
well as a SiO2/Si substrate through ambient pressure chemical
contact with the bottom layer. The stacked MoS2 bilayer was
vapor deposition (APCVD).[34] Unlike the conventionally used Ar
eventually annealed at 400 C in an Ar atmosphere for 3 h to
strengthen the assembly. Notably, PMMA was not introduced gas, ultrahigh-purity nitrogen (N2) gas was used as the carrier gas
between the top and bottom layers at all in the dry transfer to deliver the S particles to the center of the furnace to react with
process. Without interference from the foreign substance, the MoO3. It is worth noting that the core idea of tuning the CVD
interlayer coupling between the two layers is therefore strong. parameter is to reduce the nucleation rate during the initial
Compared with the wet transfer method, removal of the growth stage to render the vertical layer-by-layer growth mode
PMMA residual in the dry transfer method was achieved only more preferable. It was observed that the ratio of MoS2 bilayer
by annealing without the acetone-soaking process, which can to monolayer is positively correlated with the growth time. At the
help minimize the degradation and oxidation caused by the liq- reaction temperature of 700 C, bilayers with yield up to 30% was
uid. In addition, with the support of a glass slide and the aid of a achieved after 10 min growth. Optical images of typical grown t-
microscope, the stacking process of the dry transfer method has MoS2 bilayers with angles of 0 , 15 , and 60 are shown in
higher controllability. Nevertheless, in the case of good sample Figure 2g–i. Direct CVD synthesis of t-MoS2 was also achieved
coverage on the substrate, the wet transfer method should be an by Lin et al. in 2018.[47]
efficient method as t-TMDs with a broad range of twist angles can Alternatively, Bachmatiuk and co-workers also reported the
be obtained without involving the microscope alignment process. CVD fabrication of t-MoSe2 bilayer on SiOx substrate with a
Regardless of wet or dry transfer, proper annealing is of great purpose-built horizontal-furnace reactor.[33] In addition to Ar
importance to the interlayer coupling of the stacked-bilayer gas, H2 gas was also introduced into the chamber when the
samples. It can help to remove adsorbates and polymer residues reaction was taking place. H2 plays an important role in the
during the transferring process, as well as enhancing the vdW experiment because it acts as a reducing agent to reduce the
coupling between the top and bottom layers. Wang et al. partic- precursors, thereby promoting the chemical reaction.[48] Both
ularly examined the atomic force microscopy (AFM) images of a monolayer and a t-MoSe2 bilayer were yielded using
the fabricated twisted WSe2/WS2 bilayers.[36] They found that this route.
the step height before and after annealing was 1.2 and In another report by Zheng et al., the one-step CVD method
0.8 nm, respectively, indicating the establishment of vdW cou- was also used to synthesize a twisted WS2 (t-WS2) bilayer on
pling between two decoupled layers during the annealing pro- quartz plates with twist angles of 0 , 13 , 30 , 41 , 60 , and
cess, as shown in Figure 2e,f. Moreover, without annealing, 83 , as shown in Figure 2j–o.[35] The procedures are similar to
the sample quality is compromised, which affects relevant char- the aforementioned ones, except that a piece of magnet was
acterization results. For example, no low-frequency (LF) Raman applied to push the sulfur precursor to the edge of the furnace
lines of as-prepared twisted-bilayer samples can be observed when the designated reaction temperature was reached so that
before annealing.[37] the S powder melted and supplied S vapor to the reaction center.
In addition to the commonly used etching-assisted wet/dry Moreover, the authors found that temperature and nucleation
transfer method, some dry peel-off methods have also been play vital roles in material synthesis. Bilayer t-WS2 could only
developed to fulfill the transferring purpose in recent years. be observed at the high growth temperature of 1100 C. At a
For instance, Ma et al. reported a capillary-force-assisted clean lower growth temperature of 850 C, only conventional AA
stamp technique to pick up and transfer 2D materials.[45] A thin and AB stacking bilayers could be obtained. It was speculated that
layer of vapor (e.g., water) is the key to enhance the adhesion at a high temperature, the top layer tends to grow along with its
energy between 2D materials and PDMS, thereby facilitating original nucleus orientation and overcome the angular mismatch
the pick-up process. Once the vapor evaporates, the adhesion with the bottom layer.
energy decreases, and 2D materials can be easily released on Considering the predominant effect of the twist angle on the
to the targeted substrate. This method holds great promise for properties of TMDs, substantial research effort has been devoted
the on-chip transfer of materials. Another peel-off method was by the community to achieve the precise control of the twist
developed by Tao et al. in 2018.[46] A water-soluble polyvinyl alco- angle. Up to now, it is still a great challenge to control the twist
hol (PVA) thin film was used for sample pickup, and could then angle of CVD-grown bilayer TMDs due to the complex CVD envi-
be easily removed by a floating dissolution process. These novel ronment. Nevertheless, with the intensive study, more and more

useful information has been unveiled, paving the way for control- certain experiments where such controllability is very much
lable synthesis in the future. For example, Mandyam et al. stud- demanded, a more delicate preparation method is required.
ied the twist angle distribution of around 100 CVD-grown bilayer When Cao et al. dig into the exotic physical properties of
WSe2 flakes.[49] They found that the majority showed a twist (TBG, the “magic” angle of 1.1 needs to be accurately
angle of 0 /60 (84%), whereas the rest showed the twist angles achieved.[13,14,51] The technique adopted was “tear and stack”,
of 30 (12%) and 15 (4%). This distribution of twist angles which enables subdegree control of the twist angle.[52,53] Its sche-
agrees well with the calculated stacking energy density, with matic is displayed in Figure 3a. Poly (bisphenol-A carbonate)
the 0/60 showing the lowest values, followed by 30 and 15 (PC)/PDMS stacked on a glass slide is first used to pick to a piece
twisted structures. Zhang et al. have further investigated the tem- of hexagonal boron nitride (h-BN) at 90 C. The h-BN is then
perature effect on the thermodynamically stable 0 /60 configu- brought into contact with one-half of a graphene flake. When lift-
rations and observed that AA stacking (0 ) of bilayer MoS2 ing the glass, the graphene tears into two halves, where one half
is favorable at a lower temperature (i.e., 750 C), whereas AB is bonded to h-BN due to vdW interaction, and the other is held to
stacking (60 ) is generally obtained at a higher temperature the substrate. Next, the substrate is rotated by the desired angle.
(i.e. 800 C).[4] Very recently, the direct growth of continuously Eventually, the TBG is obtained after lowering down the glass to
twisted superstructures of WS2 and WSe2 has been reported pick up the remaining graphene.
by Zhao et al. through the introduction of non-Euclidean surfaces A typical TBG-based device structure is shown in Figure 3b.
to the growth process. This work contributes greatly to the pre- The enlarged view clearly shows two graphene layers twisted rel-
cise control of twist angles and will be discussed in detail in the ative to each other. Unconventional superconductivity (Figure 3c)
next section.[50] and correlated insulator states at half filling (Figure 3d) were
observed in “magic” angle TBG, indicating the great importance
of the controllable synthesis of clean and small-twist-angle
2.3. Discussion
bilayers. In addition to the magic TBG, the “tear-and-stack” tech-
nique has also been successfully applied to obtain WSe2 bilayers
Both stacking and one-step CVD methods have demonstrated
with twist angles ranging from 0 to 17 .[54] Especially, a small
their ability to produce t-TMDs, and each group of researchers
twist angle of around 0 has been attained (Figure 3e). Very
has slightly different methods. For the stacking method, the
recently, MoSe2/MoSe2 bilayers with near 0 twist angle have
procedure is rather tedious as it involves a transfer step, which
been prepared by Sung et al. with this method as well
is a bit challenging and time-consuming. However, one benefit
(Figure 3f ), which demonstrate broken mirror/inversion sym-
of this stacking method is that it is easier to produce twisted-
metry.[29] The effect of crystal symmetry on TMD excitons was
heterostructure bilayers because the top and bottom layers are
investigated and revealed. This twist angle control should be gen-
initially separated. In addition, with the help of a microscope,
eralizable to more 2D systems, which may exhibit more intrigu-
the stacking process can be fairly controlled to achieve desirable
ing physics behaviors.
twist angles. It is noteworthy that the final annealing process Apart from the “tear-and-stack” technique, another method
plays a significant role in the preparation of t-TMDs because with atomic precision was reported by Gao et al. in 2019.[55]
it directly affects the coupling relationship between the two A scanning tunneling microscope (STM) tip was adopted to
layers, by removing adsorbates and PMMA residues during manipulate the folding and unfolding of graphene nanoislands,
the transferring process. Various annealing conditions have been as shown in Figure 3g. TBG with controllably tunable twist
explored by different groups to eliminate the PMMA residue. For angles was successfully prepared. The obtained twist angles
example, some researchers involved a PMMA curing step prior to are analyzed and summarized in Figure 3h. This STM origami
the etching process, or a prebaking step before the final anneal- method holds great promise for the construction of atomically
ing process, whereas other experimenters did not. precise nanostructures with quantum properties and shall be
Compared with the stacking method, the one-step CVD extended to the preparation of t-TMDs.
method is relatively faster as the sample transferring and stamp- It is worth mentioning that the latest work reported by Zhao
ing processes are no longer required. The CVD methods to pro- et al. provides new insights into the fabrication of t-TMDs.[50]
duce monolayer, bilayer, or even multilayer materials are rather Continuously twisted superstructures of TMDs including
similar. It is the control parameters that depict the final product. WS2 and WSe2 can be obtained by their proposed method, which
Therefore, synthesis parameters used for synthesizing t-TMD is shown in Figure 3i. The key idea is to introduce non-Euclidean
bilayers have to be extremely precise. In general, a relatively high surfaces into the CVD growth process, which can be achieved
reaction temperature is required because randomly twisted by drop casting nanoparticles as protrusions on flat Si/SiO2
bilayers are thermodynamically unfavorable, and the interlayer substrates. The combined screw-dislocation spirals and non-
bonding requires higher activation energy to form. In addition, Euclidean surfaces together result in supertwisted spiral TMDs,
the reaction time is also critical to promote the formation of as shown in Figure 3j–m.
bilayer TMDs. For instance, in the case of CVD-grown t-MoS2
bilayers, the ratio of MoS2 bilayer to monolayer was found to
increase with the reaction time.[34] 3. Characterization of 2D t-TMDs
Although the two methods discussed previously can meet
most research needs of t- TMDs, it is still a bit challenging to The properties of t-TMDs can be characterized by various kinds
produce a clean interface and controlled twist angles (especially of instruments, and PL and Raman spectroscopy are the two
a small twist angle) simultaneously. Thus, when performing most commonly used characterization tools. PL spectroscopy

Figure 3. The emergence of more sophisticated techniques for controllably preparing bilayers with small twisted angles. a) Schematic illustration of
the “tear-and-stack” technique. Reproduced with permission.[45] Copyright 2017, American Chemical Society. b) Schematic of a typical TBG-based
device. c) Four-probe resistance (Rxx) of the “magic” angle TBG-based device measured at a narrow carrier density range versus temperature, indi-
cating the unconventional superconductivity. b,c) Reproduced with permission.[13] Copyright 2018, Springer Nature. d) The conductance (G) of the
“magic” angle TBG-based device measured at T ¼ 0.3 K, revealing the half-filling insulating states. Reproduced with permission.[14] Copyright 2018,
Springer Nature. e) Optical image of the t-WSe2 bilayers with near 0 twist angle, obtained by the “tear-and-stack” technique. Reproduced with per-
mission.[54] Copyright 2020, American Physical Society. f ) Dark-field TEM image of a MoSe2/MoSe2 bilayer with near 0 twist angle. The dark and gray
regions correspond to the alternating domains with rhombohedral stacking symmetry. Reproduced with permission.[29] Copyright 2020, Springer
Nature. g) Schematic graphic of the STM origami process of folding and unfolding graphene nanoislands. h) Summary of the twist angles obtained.
g,h) Reproduced with permission.[55] Copyright 2020, American Association for the Advancement of Science. i) Schematic illustration of the formation
process of supertwisted spirals of layered materials. j) AFM image of aligned WS2 spirals grown on flat SiO2/Si substrate. k–m) AFM images of
supertwisted WS2 spirals grown around the dropcasted particles on SiO2/Si substrate. i–m) Reproduced with permission.[50] Copyright 2020,
American Association for the Advancement of Science.

can be used to study the optical properties of the material after 3.1. PL Spectroscopy
absorption of photons from a graph of intensity versus wave-
length because the interlayer coupling between the bilayers The interlayer coupling can modify the electronic band structure
changes the band structure and therefore alters the PL emission. of TMD bilayers significantly, and PL spectroscopy is a useful
Raman spectroscopy makes use of the molecule mechanical tool to unveil such changes. Zheng et al. have studied the PL
vibration to provide information on the change of phonon vibra- spectra of t-WS2 bilayers with different twist angles, as shown
tions.[56] Apart from them, optical microscopy (OM), AFM, TEM, in Figure 4a.[35] It has been observed that randomly twisted
density functional theory (DFT) calculations, etc., have also been WS2 (13 , 30 , 41 , and 83 ) has much higher PL intensity than
used to assist in the identification of twisted bilayers. the 0 (AA stacking) and 60 (AB stacking) twisted ones. Apart
Figure 4. PL study of the electronic coupling evolution in t-TMD bilayers. a) PL spectra of monolayer and bilayer WS2 with the twist angle of 0 , 13 , 30 ,
41 , 60 , and 83 . b) Plot of PL peaks’ positions of the curves in (a) obtained by Lorentz fitting with respect to the twist angle. a,b) Reproduced with
permission.[35] Copyright 2015, Wiley-VCH. c) PL spectra of monolayer and bilayer MoS2 with the twist angle of 0 , 15 , and 60 . d) Dependence of PL
peak energies of 44 MoS2 bilayers with respect to different twist angles. c,d) Reproduced with permission.[34] Copyright 2014, Nature Research.
e) Mapping of the PL integrated intensity of a twisted-bilayer MoS2 flake obtained by stacking. f ) PL spectrum of the MoS2 bilayer with the twist angle
of 84.7 . g–i) Twist angle dependence of the A trion to the A exciton PL intensity ratio (g), trion binding energy (h), and difference of PL peak energy
between B and A excitons (EB EA) (i). e–i) Reproduced with permission.[32] Copyright 2014, American Chemical Society. j,k) PL spectra of twisted bilayer
WSe2/WS2 obtained by stacking on SiO2/Si substrate before and after annealing. j–k) Reproduced with permission.[36] Copyright 2016, American
Chemical Society.

from that, the 0 and 60 stacked bilayers show two dominating EB EA reaches the minimum. Again, changes shown in the
peaks, i.e., peak A and peak I, corresponding to direct and indirect aforementioned curves suggest the variations of the interlayer
transitions, respectively. This is consistent with previous reports coupling of MoS2 bilayers with different twist angles.
that t-WS2 with 0 and 60 twist angles are indirect-bandgap mate- Last but not least, Wang et al. adopted PL spectroscopy to study
rials.[57,58] In contrast, there is no indirect transition peak I in the the electronic interlayer coupling evolution in the twisted heter-
spectra of randomly twisted bilayers. However, it was found that ostructure bilayers, i.e., WSe2/WS2.[36] In their work, they partic-
the spectra of randomly twisted bilayers have a minor peak AI, ularly examined the PL spectra of the twisted bilayers before and
which is believed to be the interlayer excitonic transition. To better after annealing to study the effect of annealing on the PL emis-
illustrate the changes of PL spectra, a plot of PL peaks’ positions sion (Figure 4j,k). It was found that after annealing, the PL inten-
obtained by Lorentz fitting with respect to the twist angle is shown sity of WS2 and WSe2 significantly reduced by a factor of 37 and
in Figure 4b. The indirect-bandgap energy (peak I) is an indicator 18 in the heterojunction area, respectively, which is strong evi-
of the interlayer electronic coupling strength: the lower the indi- dence suggesting the formation of interlayer coupling between
rect bandgap energy, the stronger the coupling strength.[34] Ab ini- two layers.
tio calculation has later been applied to verify that the weakened
interlayer coupling between randomly twisted bilayers results
from the enlarged interlayer distance. 3.2. Raman Spectroscopy
A similar twist-angle-dependent trend has also been observed
in CVD-grown MoS2 bilayers by Liu et al.[34] Figure 4c displays Raman spectroscopy is one of the most useful techniques for
the PL spectra of monolayer as well as bilayer MoS2 with twist studying the evolution of mechanical couplings in TMD bilayers
angles of 0 , 15 , and 60 . For all the bilayers, the PL intensity because phonon vibrations in 2D TMDs are very sensitive to
decreases significantly compared to the intensity of monolayer interlayer coupling.[59] In the study reported by Huang et al.,
PL. In addition, they found that a minor peak II with lower Raman spectroscopy was adopted to study the properties of a
energy resulting from an indirect bandgap appeared in the t-MoS2 bilayer obtained by the stacking method, with the twist
twisted bilayer PL spectra. To further clarify the interlayer cou- angles of 17.72 , 51.5 , and 84.7 .[32] The corresponding
pling between the bilayers, they did the same measurement for Raman spectra are shown in Figure 5a. Two prominent peaks
plenty of samples with various twist angles and then plotted the can be identified in the range 360–440 cm1, corresponding to
transition energies of peak I and II (Figure 4d). As can be seen, the in-plane E2g and out-of-plane A1g phonon modes, respec-
peak I shows consistent energy for all the different twist angles. tively. However, no obvious angle dependence of the Raman sig-
However, for peak II, it was observed that the 0 twisted-bilayer nal can be observed in this work. Later, Liu and colleagues
MoS2 and 60 twisted-bilayer MoS2 have the lowest energy reported a more comprehensive Raman study of CVD-grown
among the rest of the randomly twisted bilayers. Surprisingly, MoS2 with various twist angles. They first studied the Raman
the trend is the same as that in the aforementioned Figure 4b, spectra of monolayer and bilayer MoS2 with twist angles of
whereby the respective minor peaks for both materials have the 0 , 15 , and 60 (Figure 5b). They found that the E2g peaks of
lowest energy for the 0 and 60 stacked bilayers. This can also be all the bilayers exhibit a constant redshift, whereas the A1g peak
explained by the stacking configuration of the bilayers. Bilayers shows a twist-angle-dependent blueshift. Therefore, the separa-
with 0 and 60 angles have the most symmetric configuration; tion between two peaks (ωA–ωE) serves as an effective indicator of
therefore, they are more stable. As a result, 0 and 60 twisted the mechanical interlayer coupling: the larger the separation, the
bilayers have the smallest indirect bandgap and the strongest stronger the mechanical coupling. To further reveal the twist
interlayer coupling between the layers. angle dependence, Raman spectra of 44 MoS2 bilayers with dif-
The PL spectra of a t-MoS2 bilayer obtained by the stacking ferent twist angles were collected and analyzed, whose results are
method have also been studied.[32] Similarly to CVD-grown plotted in Figure 5c. It can be clearly identified that the AA or AB
MoS2 bilayers, the t-MoS2 bilayer has much lower PL intensity stacked bilayer possesses the strongest coupling, which agrees
when compared to the monolayer, as indicated by the PL map- well with the result revealed by PL spectroscopy. Moreover,
ping result shown in Figure 4e. Moreover, it was found that the bilayers with other random twist angles possess weaker but con-
PL profile of bilayers varies with different twist angles. Figure 4f stant coupling, consistent with the result obtained by Huang
provides the typical PL spectrum of a t-MoS2 bilayer with a twist et al. that bilayer MoS2 with the twist angle of 17.72 , 51.5 ,
angle of 84.7 . As shown in Figure 4g, the PL intensity ratio of and 84.7 possess rather similar Raman spectra.
the A trion to the A exciton demonstrates an oscillatory behavior Apart from t-MoS2, Raman spectroscopy has also been applied
as a function of the twist angle. The curve is approximately sym- to investigate the interlayer coupling of t-WS2 bilayers.[35] Raman
metrical with respect to the twist angle of 60 , which is consistent spectra of the WS2 bilayers with random twist angles ranging
with the symmetry of the D3h crystal structure of the monolayer from 0 to 83 are shown in Figure 5d. As expected, some char-
MoS2. A similar oscillatory and periodic trend was also observed acteristic differences indicate different phonon vibrations in the
in the PL emission energy difference between the A and A bilayers arising from the interlayer coupling effect. For example,
peaks, which indicates the binding energy of the trions the longitudinal acoustic phonon (2LA) peaks of the randomly
(Figure 4h). Moreover, the difference between the PL peak twisted bilayer WS2 have shifted to a higher wavenumber com-
energy of B and A excitons, EB EA, exhibits similar but opposite pared to the highly symmetrically configured 0 and 60 bilayers.
periodic and oscillatory behavior, as shown in Figure 4i. EB EA Moreover, their A1g peaks are generally redshifted and broad-
is inversely correlated to the interlayer coupling. When θ ¼ 0 ened, suggesting the weaker interlayer mechanical coupling of
and 60 , the interlayer coupling is the strongest, whereas randomly twisted bilayer WS2 than the more stable AA or AB

Figure 5. Raman and other characterizations of t-TMD bilayers. a) Raman spectra of stacked bilayer MoS2 with the twist angle of 17.72 , 51.5 , and 84.7 .
b) Raman spectra of CVD-grown monolayer and bilayer MoS2 with the twist angle of 0 , 15 , and 60 . c) Plot of the Raman peak separation between the
A1g and E2g modes (ωA–ωE) versus the twist angles. d) Raman spectra of bilayer WS2 with the twist angle of 0 , 13 , 30 , 41 , 60 , and 83 . e) Raman
spectra of twisted bilayer WSe2/WS2 obtained by stacking on a sapphire substrate. f ) Stokes and anti-Stokes Raman spectra of bilayer MoSe2 with twist
angle ranging from 33.2 to 60 . g) Plot of the twist-angle-dependent Raman shifts of the shear and two breathing modes of stacked bilayer MoSe2.
h) FFT-filtered TEM image of bilayer WSe2/WS2 and corresponding FFT pattern. i) Schematics of bilayer MoS2 with different twisted stackings. Mo and S
atoms are shown in green and yellow balls, respectively. j) Plot of calculated interlayer distance and indirect bandgap of bilayer WS2 with respect to twist
angles. a) Reproduced with permission.[32] Copyright 2014, American Chemical Society. b,c,i) Reproduced with permission.[34] Copyright 2014, Nature
Research. d,j) Reproduced with permission.[35] Copyright 2015, Wiley-VCH. e,h) Reproduced with permission.[36] Copyright 2016, American Chemical
Society. f,g) Reproduced with permission.[37] Copyright 2016, American Chemical Society.
stacked bilayers. This finding is in good agreement with the of bilayer heterostructures. For example, Wang et al. studied
result observed from MoS2. the Raman spectra of a WSe2/WS2 heterostructure obtained
In addition to bilayers made of the same material, Raman by the stacking method, as shown in Figure 5e.[36] They noted
spectroscopy can also be used to reveal the interlayer coupling that the A1g peak of WSe2 in the bilayer spectra was redshifted
with respect to its monolayer spectra. As WSe2 generally red- interlayer coupling, as discussed earlier (Figure 2e,f ). TEM is
shifts as the number of layers increase, the presence of interlayer another effective tool to study the structure of t-TMDs on a
coupling between WS2 and WSe2 could thus be confirmed. microscopic scale. For example, Figure 5h shows a fast
Moreover, a new peak at 309.4 cm1 appeared, which corre- Fourier transform (FFT)-filtered TEM image of bilayer WSe2/
sponds to a layer-number-sensitive vibrational mode of WS2. WS2.[36] Moiré fringes can be clearly observed, resulted from
This again shows the interlayer coupling effect between two the overlaying of lattice-mismatched or rotated 2D heterostruc-
stacked layers. tures.[65] The twist angle of 20 can be determined by measuring
An alternate effort has been devoted by Puretzky et al. to uti- the two hexagonal reciprocal lattices in the image.
lize LF Raman spectroscopy to determine the interlayer coupling To understand the origin of changes in properties, various
between t-MoSe2 bilayers.[37] Compared with the conventional research groups have performed DFT calculations to study the
Raman approach, LF Raman spectroscopy is more sensitive to coupling evolution of t-TMD bilayers. It has been found that
probing vdW interlayer coupling as it can detect the in-plane the interlayer distance of t-MoS2 bilayers has a strong variation
shear and out-of-plane breathing vibrations of the entire 2D layer with respect to various twist angles, as shown in Figure 5i.[34] The
moving as a whole unit.[37,56,60,61] It has already been successfully most stable AA- and AB-stacked MoS2 possesses the smallest
applied to study the effect of interlayer coupling on twisted mul- interlayer separation of 0.61 nm, and the separation of the four
tilayer graphene.[62,63] Figure 5f shows the Stokes and anti-Stokes twisted MoS2 is 0.65 nm. This can be understood physically
LF Raman spectra of t-MoSe2 bilayers with twist angles ranging through repulsive steric effects. For the twisted-bilayer MoS2,
from 33.2 to 60 . Notably, the acquired LF Raman spectra pos- the S atoms of the top layer are located near the S atoms of
sess a highly sensitive response to the interlayer coupling, and the bottom layer. Therefore, the resulting repulsion between
even a slight deviation of the twist angle may result in very dif- adjacent S atoms leads to a larger interlayer distance. On the con-
ferent Raman spectra. For instance, a slight twist angle has trary, in the energetically favorable AA and AB stackings, the
caused the breathing mode B1 to appear in the spectrum of S atoms of the top layer correspond to the trigonal vacancies
59.6 but not in the one of 60 . The intensity of the B1 peak of the S atoms of the bottom layer in a staggered manner, thereby
becomes much more pronounced as the twist angle further reducing the repulsion and minimizing the interlayer distance.
decreases to 56.9 . Another interesting feature is that the shear A similar result has also been reported on the CVD-grown t-WS2
mode peak S quickly disappears when the twist angle decreases bilayers (Figure 5j).[35] In addition, it has been found that the
to 55.3 . This could be attributed to the reduction of the in-plane twisted bilayer has a larger indirect bandgap than the one with
restoring force as stacking becomes unaligned with mismatched AA or AB stacking. The larger interlayer distance and indirect
periodicity. Apart from that, a new breathing peak B2 shows up bandgap together indicate the weaker interlayer couplings in
when the twist angle decreases to 55.3 , corresponding to the twisted bilayers.
mismatched alignments. For better visualization, the Raman
shifts of the shear and breathing modes with respect to the twist
angles are summarized and shown in Figure 5g. The same
4. Physical Properties of 2D t-TMDs
set of samples has also been investigated using high-frequency Moiré superlattices can be formed by vertically stacking two
(HF) Raman spectroscopy, and it was found that the spectra layers of materials with different lattice constants and/or with
have much fewer variations of frequency and intensity with a twist angle, as shown in Figure 5h. The corresponding
the twist angle. moiré potential landscapes enable the emergence of exotic phe-
Almost at the same time, Huang et al. reported the LF Raman nomena with a variety of building blocks, such as fractal quan-
study of t-MoS2 bilayers.[64] Similar findings have been demon- tum Hall effect and tunable Mott insulators in graphene/BN
strated that the LF modes are much more responsive to the inter- moiré superlattices[11,12,14,66,67] and unconventional supercon-
layer stacking and coupling in contrast to the HF Raman modes. ductivity in TBG.[13] Recent studies have not only given an insight
Especially, when the twist angle is near 0 and 60 , the frequency into the relationship between twist angles and the optoelectronic
and intensity variation of LF modes can reach 8 cm1 and five performance of heterojunctions, but also established t-TMDs as a
times, respectively. Therefore, it can be summarized that LF promising platform to investigate interlayer excitons and corre-
Raman spectroscopy is sometimes more effective in studying lated physics.[30,31,54,68–70]
interlayer stacking and coupling as it can provide more detailed
and comprehensive characterization.
4.1. Identification of Interlayer Excitons
3.3. Other Techniques Theory predicts that remarkable effects on optical excitations
could be resulted from the moiré potential in 2D valley semicon-
In addition to PL and Raman spectroscopy, OM, AFM, TEM, and ductors with the Coulomb bounded electrons and holes in differ-
DFT calculations have also been applied to identify and study the ent monolayers.[28,71,72] These interlayer excitons exhibit many
t-TMD bilayers. As shown in Figure 2, most twisted samples can appealing properties, such as valley-contrasting physics,[28,71–78]
be observed from the OM images directly due to the color con- long population and valley lifetimes,[73,79–81] high electrical tun-
trast. AFM is a very high-resolution type of scanning probe ability,[73,77,79,82] and strong many-body correlations.[73,79,80]
microscopy (SPM), which can probe the boundary between However, the signatures of these interlayer excitons had not been
the monolayer and bilayer. On top of that, the corresponding detected through experiments until Seyler et al. reported evi-
height profile can also be used to investigate the formation of dence of moiré-trapped interlayer valley excitons in MoSe2/WSe2
heterobilayers.[69] They observed PL peaks about two orders small twist angle and a large twist angle are shown in Figure 6b.
narrower than free interlayer excitons near the same energy The latter only shows a single resonance in the energy range of
at low temperatures (Figure 6a) indicating the presence of 1.6–1.8 eV from the WSe2 A exciton state. In contrast, the moiré
trapped excitons. The trapped interlayer excitons show strong superlattice formed in the small angle-twisted heterostructures
circular valley polarization, which has the same helicity for a brings about three prominent peaks (labeled I to III), which have
given twist angle, suggesting that the trapping potential is a comparable oscillator strength in this range, corresponding to
result of the moiré landscape preserving the threefold rotational distinct moiré exciton states. Moreover, its gate dependencies
symmetry. are different from that of the A exciton in the WSe2 monolayer
Jin et al. also provided insight into the emergent moiré exciton and the WSe2/WS2 heterostructure with a large twist angle
states observed in closely aligned WSe2/WS2 heterostructure.[30] (Figure 6c). These unusual dependencies that vary for different
Reflection contrast spectra of a WSe2/WS2 heterostructure with a peaks cannot be explained by any established electron–electron
Figure 6. Investigation of interlayer excitons in t-TMDs. a) PL spectrum of MoSe2/WSe2 heterobilayers with the twist angle of 2 at excitation powers of
10 μW (dark red) and 20 nW (blue). The inset shows the Lorentzian fit to a representative PL peak indicating a linewidth of about 100 μeV (20 nW
excitation power). Reproduced with permission.[69] Copyright 2019, Springer Nature. b) Reflection contrast spectra of a WSe2/WS2 heterostructure with
a small twist angle (light blue, top) and one with a large twist angle (black, bottom). c) Detailed reflection contrast spectra in the range of 1.6–1.9 eV
(the WSe2 A exciton) on the electron-doping side. Units of the electron concentration: cm2. b,c) Reproduced with permission.[30] Copyright 2019,
Springer Nature. d) Anomalous diffusion of interlayer excitons arising from the interplay between the moiré potentials and strong many-body interactions.
Twist-angle-dependent width σ2t – σ20 with respect to the pump–probe delay times with N0 ¼ 6.0 1012 cm2 at 295 K, indicating a faster transport in
the 60 heterobilayer. e) Interlayer exciton transport of the 60 heterobilayer at different exciton densities at 295 K. f ) Temperature-dependent interlayer
exciton transport of the 60 heterobilayer with N0 ¼ 4.1 1012 cm2. d–f ) Reproduced with permission.[31] Copyright 2020, Springer Nature. g) DOCP
of the t-WSe2 bilayers with twist angles of 180 , 0 , 2 , 17 , respectively (from top to bottom). h) The doping dependence of
DOCP. Switching from large values (>80%) in the n-doped regime to almost zero (<5%) in the p-doped regime is displayed. g,h) Reproduced with
permission.[54] Copyright 2020, American Physical Society.
interactions. Instead, they can be understood through an empir- momentum space, t-WSe2 is characterized by slow depolariza-
ical theory that introduces periodic moiré exciton potential in the tion and long valley lifetimes.[28,88–90] The doping dependence
strong-coupling regime[83] because the interlayer electron– of DOCP is shown in Figure 6h. Although the interlayer exciton
exciton interaction can be sufficiently enhanced by the moiré DOCP exceeds 80% in the n-doped regime, it can be switched to
superlattice. The multiple flat bands with substantially reduced almost zero (<5%) in the p-doped regime. This study demon-
bandwidth generated in this model also provide a promising plat- strates the possibility to build tunable chiral photonics and elec-
form to explore exotic states, such as a topological exciton insu- trically switchable spin–valley-based devices with t-TMD bilayers
lator and a strongly correlated exciton Hubbard model.[30] and opens up avenues for applications in quantum information
Apart from the WSe2/WS2 heterostructure, twist-angle- storage and processing.[54]
dependent interlayer exciton emission was observed in the
bilayer WS2/MoS2 heterostructure as well.[84] A noticeable inter- 4.3. Correlated Electronic Phases
layer exciton peak at around 1.5 eV can be observed in the PL
spectra of coherent WS2/MoS2, whereas no peak was found in t-TMDs exhibit different electronic properties from their non-
the artificially stacked random WS2/MoS2. The interlayer transi- twisted counterparts as well. Recent studies have shown that
tion is responsible for the emission, which is suppressed in the the charge mobility in 30 twisted MoS2 bilayers is higher than
random stacks due to nonradiative recombinations by momen- the 0 structures, manifesting as a larger on/off ratio of devices
tum mismatch or interfacial impurity centers. fabricated by 30 twisted bilayers.[85] Interlayer decoupling by
More recently, precise control of the interlayer twist angle in incommensurate structure or smaller interlayer resistance
large-scale MoS2 homostructures has been realized by Liao et al. may cause this phenomenon.[85,91]
with the stacking method.[85] The PL results demonstrate that the Similarly to TBG, t-TMDs also provide an ideal platform for
energies of indirect interlayer excitons can be tuned by precisely band engineering. Flat bands whose bandwidths vary continu-
controlling the twist angles. This work introduces a pathway to ously with the inverse moiré wavelengths and van Hove singu-
industrial applications of t-TMD photonics. larities (vHSs) of the moiré Brillouin zone that may support the
emergent electronic phases have been predicted to exist.[15,92–95]
4.2. Diffusion of Interlayer Exciton Regarding this, Wang et al. have specially looked into the t-WSe2
bilayers.[70] A schematic diagram of the band structure of t-WSe2
To further understand how the moiré landscapes modulate the is shown in Figure 7a. Carrier correlations are strong enough
properties of t-TMDs, several research works focused on exciton within these low-energy flat bands to induce a correlation-driven
motions in moiré superlattices. Using transient absorption insulation state near half-filling (Figure 7b). They also observed a
microscopy (TAM) combined with first-principles calculations, potential superconducting transition at low temperature away
from half-filling (Figure 7c), though the superconductor gap
Yuan et al. investigated interlayer exciton dynamics and transport
seems to be weakly formed. A phase diagram for the t-WSe2
in WS2/WSe2 heterobilayers with different twist angles (0 and
bilayer with respect to angle and displacement field is shown
60 ).[31] First-principle calculations indicate that moiré potentials
in Figure 7d. According to single-particle band-structure calcula-
of 0 are much deeper than those of 60 and ΔEK–Q (the mean
tions (Figure 7e), its qualitative correlation with the density of
difference of the spatial variation of the energy difference
states (DOS) at half-filling indicates that the DOS at half-filling
between the K–Q and the K–K transitions over the entire
plays a key role in stabilizing the entire measured angle range.
moiré pattern) is larger for 0 than for 60 . Therefore, the inter-
Due to its widely tunable electronic structure, the t-WSe2 system
layer excitons in 60 samples are more mobile accordingly
with engineered flat bands can be generalized as a model system
(Figure 6d). The migration speed of interlayer excitons is also
for exploring other emerging electronic phases, such as exciton
faster at higher densities, as shown in Figure 6e, which can
condensates, spin liquids, and magnetic ordering.[70]
be explained by the net repulsive Coulomb interaction between
the interlayer excitons attributed to electron–hole separation and
the aligned electric dipole moments.[73,86,87] Moreover, as the 5. Conclusion
temperature decreases, a reduced migration speed of exciton
is observed (Figure 6f ), which is of opposite tendency to intra- In this review, a comprehensive overview of the state-of-the-art
layer excitons. These results serve as a guideline for the investi- research activities in the preparation of t-TMDs and their intrigu-
gation of exciton and spin transport in vdW heterostructures and ing properties is provided. The reported synthesis conditions of
shine light on designing quantum photonics communication twisted-bilayer MoS2, MoSe2, WS2, W-doped MoSe2, and WSe2/
devices.[31] WS2 heterostructures are summarized and discussed. Two major
The valley dynamics of t-TMDs have been proved to be mag- synthetic routes, i.e., the two-step stacking method and one-step
nificently tunable via twist angles and electrostatic doping by CVD method, are introduced and compared. The stacking
Scuri et al.[54] They investigated twisted WSe2 (t-WSe2) bilayers method is advantageous in controlling the twist angle, whereas
with a twist angle of 0 to 17 . It was found that in contrast to the an additional annealing step is required to eliminate the polymer
almost zero degree of circular polarization (DOCP) of a neutral residues during the transferring process. On the contrary, the
interlayer exciton in natural bilayers (180 ), as high as 60% can be CVD method offers opportunities to directly grow much cleaner
reached in twisted bilayers upon illumination with circularly t-TMD bilayers; however, the twist angle produced is random and
polarized light (Figure 6g). In addition, due to nondegenerate unpredictable. Establishing the correlation between CVD synthe-
spin at K and Q points and indirection in both real and sis parameters and the twist angles of as-grown TMD bilayers
Figure 7. Study of correlated electronic phases of t-TMDs. a) Schematic illustration of the hybridized band structure of t-WSe2 bilayers in the K valley, spin
up (left), and K 0 valley, spin down (right). b) Resistance as a function of carrier density of bilayer WSe2 with six different configurations, measured at
T ¼ 1.8 K. c) Temperature-dependent resistance of bilayer WSe2 with a twist angle of 5.1 at densities near half-filling. At temperatures below 35 K, the
insulating response is identified at half-filling (blue). At temperatures below 3 K, zero resistance in the vicinity of half-filling (red and green) is observed.
d) Phase diagram of the t-WSe2 with respect to displacement field and twist angle, revealing the observed metallic and insulating regions at half-filling.
White circles indicate the metal-to-insulator transitions. e) Calculated DOS at half-filling as a function of the displacement field and twist angle.
Reproduced with permission.[70] Copyright 2020, Springer Nature.
would be a valuable yet challenging future research direction. Science Foundation of China (Grant No. 51771224, and 61888102) and
Notably, the success of synthesizing the “magic” angle TBG the National Key R&D Program of China (Grant No. 2016YFA0202301
by the “tear-and-stack” technique provides a promising direction and 2018FYA0305800).
for the controllable synthesis and study of t-TMDs, especially at
small twisted angles. The introduction of non-Euclidean surfaces
into the CVD growth process to prepare supertwisted TMDs Conflict of Interest
offers new possibilities to construct and study t-TMDs. The authors declare no conflict of interest.
In addition, as the most commonly used characterization
techniques to probe the interlayer coupling between the twisted
bilayers, Raman spectroscopy and PL spectroscopy are particu-
larly elaborated on. Several intriguing physical properties Keywords
identified in t-TMDs, including moiré superlattices, interlayer chemical vapor deposition, stacking method, transition metal
excitons, and correlated electronic phases, are summarized. dichalcogenides, twisted bilayers, 2D
Along with the more in-depth exploration of the twisted multi-
layer TMDs, more novel properties and new phenomena may Received: December 30, 2020
Revised: February 9, 2021
arise from this intriguing structure. To realize practical optoelec-
Published online: March 10, 2021
tronic applications, more research efforts are required to address
issues related to twist-angle-controlled high-quality synthesis and
comprehensive characterization. [1] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano,
Nat. Nanotechnol. 2012, 7, 699.
[2] Y. Chen, Z. Fan, Z. Zhang, W. Niu, C. Li, N. Yang, B. Chen, H. Zhang,
Acknowledgements Chem. Rev. 2018, 118, 6409.
[3] M. Chhowalla, H. S. Shin, G. Eda, L.-J. Li, K. P. Loh, H. Zhang, Nat.
This work was supported by National Research Foundation Singapore
program NRF-CRP21-2018-0007 and NRF-CRP22-2019-0007, Singapore Chem. 2013, 5, 263.
Ministry of Education via AcRF Tier 3 Programme “Geometrical [4] X. Zhang, H. Nan, S. Xiao, X. Wan, X. Gu, A. Du, Z. Ni, K. K. Ostrikov,
Quantum Materials” (MOE2018-T3-1-002), AcRF Tier 2 (MOE2016-T2- Nat. Commun. 2019, 10, 598.
1-131) and AcRF Tier 1 RG4/17 and RG7/18. This research was also sup- [5] S. Kim, A. Konar, W.-S. Hwang, J. H. Lee, J. Lee, J. Yang, C. Jung,
ported by A*STAR under its AME IRG Grant (Project No. 19283074). The H. Kim, J.-B. Yoo, J.-Y. Choi, Nat. Commun. 2012, 3, 1011.
authors also gratefully acknowledge the support from the National Nature [6] C. Long, Y. Dai, Z.-R. Gong, H. Jin, Phys. Rev. B 2019, 99, 115316.
[7] A. Luican, G. Li, A. Reina, J. Kong, R. Nair, K. S. Novoselov, [37] A. A. Puretzky, L. Liang, X. Li, K. Xiao, B. G. Sumpter, V. Meunier,
A. K. Geim, E. Andrei, Phys. Rev. Lett. 2011, 106, 126802. D. B. Geohegan, ACS Nano 2016, 10, 2736.
[8] G. Li, A. Luican, J. L. Dos Santos, A. C. Neto, A. Reina, J. Kong, [38] S. A. Han, R. Bhatia, S.-W. Kim, Nano Convergence 2015, 2, 17.
E. Andrei, Nat. Phys. 2010, 6, 109. [39] Y. Shi, H. Li, L.-J. Li, Chem. Soc. Rev. 2015, 44, 2744.
[9] M. Yankowitz, J. Xue, D. Cormode, J. D. Sanchez-Yamagishi, [40] Q. Ji, Y. Zhang, Y. Zhang, Z. Liu, Chem. Soc. Rev. 2015, 44, 2587.
K. Watanabe, T. Taniguchi, P. Jarillo-Herrero, P. Jacquod, [41] Z. Cai, B. Liu, X. Zou, H.-M. Cheng, Chem. Rev. 2018, 118, 6091.
B. J. LeRoy, Nat. Phys. 2012, 8, 382. [42] S. Y. Kim, J. Kwak, C. V. Ciobanu, S. Y. Kwon, Adv. Mater. 2019, 31,
[10] K. S. Novoselov, E. McCann, S. Morozov, V. I. Fal’ko, M. Katsnelson, 1804939.
U. Zeitler, D. Jiang, F. Schedin, A. Geim, Nat. Phys. 2006, 2, 177. [43] H. Li, Y. Li, A. Aljarb, Y. Shi, L.-J. Li, Chem. Rev. 2017, 118, 6134.
[11] C. R. Dean, L. Wang, P. Maher, C. Forsythe, F. Ghahari, Y. Gao, [44] Y. Huang, Y.-H. Pan, R. Yang, L.-H. Bao, L. Meng, H.-L. Luo, Y.-Q. Cai,
J. Katoch, M. Ishigami, P. Moon, M. Koshino, Nature 2013, 497, 598. G.-D. Liu, W.-J. Zhao, Z. Zhou, Nat. Commun. 2020, 11, 2453.
[12] L. Ponomarenko, R. Gorbachev, G. Yu, D. Elias, R. Jalil, A. Patel, [45] X. Ma, Q. Liu, D. Xu, Y. Zhu, S. Kim, Y. Cui, L. Zhong, M. Liu, Nano
A. Mishchenko, A. Mayorov, C. Woods, J. Wallbank, Nature 2013, Lett. 2017, 17, 6961.
497, 594. [46] L. Tao, H. Li, Y. Gao, Z. Chen, L. Wang, Y. Deng, J. Zhang, J. B. Xu,
[13] Y. Cao, V. Fatemi, S. Fang, K. Watanabe, T. Taniguchi, E. Kaxiras, Adv. Mater. Technol. 2018, 3, 1700282.
P. Jarillo-Herrero, Nature 2018, 556, 43. [47] M.-L. Lin, Q.-H. Tan, J.-B. Wu, X.-S. Chen, J.-H. Wang, Y.-H. Pan,
[14] Y. Cao, V. Fatemi, A. Demir, S. Fang, S. L. Tomarken, J. Y. Luo, X. Zhang, X. Cong, J. Zhang, W. Ji, ACS Nano 2018, 12, 8770.
J. D. Sanchez-Yamagishi, K. Watanabe, T. Taniguchi, E. Kaxiras, [48] J. C. Shaw, H. Zhou, Y. Chen, N. O. Weiss, Y. Liu, Y. Huang, X. Duan,
Nature 2018, 556, 80. Nano Res. 2014, 7, 511.
[15] F. Wu, T. Lovorn, E. Tutuc, I. Martin, A. MacDonald, Phys. Rev. Lett. [49] S. V. Mandyam, M.-Q. Zhao, P. Masih Das, Q. Zhang, C. C. Price,
2019, 122, 086402. Z. Gao, V. B. Shenoy, M. Drndic, A. T. C. Johnson, ACS Nano
[16] Z. Shi, X. Wang, Y. Sun, Y. Li, L. Zhang, Semicond. Sci. Technol. 2018, 2019, 13, 10490.
33, 093001. [50] Y. Zhao, C. Zhang, D. D. Kohler, J. M. Scheeler, J. C. Wright,
[17] M. Yankowitz, S. Chen, H. Polshyn, Y. Zhang, K. Watanabe, P. M. Voyles, S. Jin, Science 2020, 370, 442.
T. Taniguchi, D. Graf, A. F. Young, C. R. Dean, Science 2019, 363, 1059. [51] A. Uri, S. Grover, Y. Cao, J. Crosse, K. Bagani, D. Rodan-Legrain,
[18] K. Kim, A. DaSilva, S. Huang, B. Fallahazad, S. Larentis, T. Taniguchi, Y. Myasoedov, K. Watanabe, T. Taniguchi, P. Moon, Nature 2020,
K. Watanabe, B. J. LeRoy, A. H. MacDonald, E. Tutuc, Proc. Natl. Acad. 581, 47.
Sci. 2017, 114, 3364. [52] Y. Cao, J. Luo, V. Fatemi, S. Fang, J. Sanchez-Yamagishi, K. Watanabe,
[19] J. Zhou, J. Lin, X. Huang, Y. Zhou, Y. Chen, J. Xia, H. Wang, Y. Xie, T. Taniguchi, E. Kaxiras, P. Jarillo-Herrero, Phys. Rev. Lett. 2016, 117,
H. Yu, J. Lei, Nature 2018, 556, 355. 116804.
[20] X. Huang, Z. Zeng, H. Zhang, Chem. Soc. Rev. 2013, 42, 1934. [53] K. Kim, M. Yankowitz, B. Fallahazad, S. Kang, H. C. Movva, S. Huang,
[21] X. Duan, C. Wang, A. Pan, R. Yu, X. Duan, Chem. Soc. Rev. 2015, S. Larentis, C. M. Corbet, T. Taniguchi, K. Watanabe, Nano Lett. 2016,
44, 8859. 16, 1989.
[22] S. Manzeli, D. Ovchinnikov, D. Pasquier, O. V. Yazyev, A. Kis, Nat. [54] G. Scuri, T. I. Andersen, Y. Zhou, D. S. Wild, J. Sung, R. J. Gelly,
Rev. Mater. 2017, 2, 17033. D. Bérubé, H. Heo, L. Shao, A. Y. Joe, Phys. Rev. Lett. 2020, 124,
[23] D. Voiry, A. Mohite, M. Chhowalla, Chem. Soc. Rev. 2015, 44, 2702. 217403.
[24] K. Chen, X. Wan, J. Xu, Adv. Funct. Mater. 2017, 27, 1603884. [55] H. Chen, X.-L. Zhang, Y.-Y. Zhang, D. Wang, D.-L. Bao, Y. Que,
[25] J. Zhou, B. Tang, J. Lin, D. Lv, J. Shi, L. Sun, Q. Zeng, L. Niu, F. Liu, W. Xiao, S. Du, M. Ouyang, S. T. Pantelides, Science 2019, 365, 1036.
X. Wang, Adv. Funct. Mater. 2018, 28, 1801568. [56] X. Zhang, X.-F. Qiao, W. Shi, J.-B. Wu, D.-S. Jiang, P.-H. Tan, Chem.
[26] P. K. Sahoo, S. Memaran, Y. Xin, L. Balicas, H. R. Gutiérrez, Nature Soc. Rev. 2015, 44, 2757.
2018, 553, 63. [57] H. R. Gutiérrez, N. Perea-López, A. L. Elías, A. Berkdemir, B. Wang,
[27] A. M. Jones, H. Yu, J. S. Ross, P. Klement, N. J. Ghimire, J. Yan, R. Lv, F. López-Urías, V. H. Crespi, H. Terrones, M. Terrones, Nano
D. G. Mandrus, W. Yao, X. Xu, Nat. Phys. 2014, 10, 130. Lett. 2013, 13, 3447.
[28] F. Wu, T. Lovorn, A. MacDonald, Phys. Rev. B 2018, 97, 035306. [58] W. Zhao, Z. Ghorannevis, L. Chu, M. Toh, C. Kloc, P.-H. Tan, G. Eda,
[29] J. Sung, Y. Zhou, G. Scuri, V. Zólyomi, T. I. Andersen, H. Yoo, ACS Nano 2013, 7, 791.
D. S. Wild, A. Y. Joe, R. J. Gelly, H. Heo, S. J. Magorrian, D. Bérubé, [59] A. Berkdemir, H. R. Gutiérrez, A. R. Botello-Méndez, N. Perea-López,
A. M. M. Valdivia, T. Taniguchi, K. Watanabe, M. D. Lukin, P. Kim, A. L. Elías, C.-I. Chia, B. Wang, V. H. Crespi, F. López-Urías,
V. I. Fal’ko, H. Park, Nat. Nanotechnol. 2020, 15, 750. J.-C. Charlier, Sci. Rep. 2013, 3, 1755.
[30] C. Jin, E. C. Regan, A. Yan, M. I. B. Utama, D. Wang, S. Zhao, Y. Qin, [60] G. R. Bhimanapati, Z. Lin, V. Meunier, Y. Jung, J. Cha, S. Das, D. Xiao,
S. Yang, Z. Zheng, S. Shi, Nature 2019, 567, 76. Y. Son, M. S. Strano, V. R. Cooper, ACS Nano 2015, 9, 11509.
[31] L. Yuan, B. Zheng, J. Kunstmann, T. Brumme, A. B. Kuc, C. Ma, [61] H. Zeng, B. Zhu, K. Liu, J. Fan, X. Cui, Q. Zhang, Phys. Rev. B 2012, 86,
S. Deng, D. Blach, A. Pan, L. Huang, Nat. Mater. 2020, 19, 671. 241301.
[32] S. Huang, X. Ling, L. Liang, J. Kong, H. Terrones, V. Meunier, [62] P. Tan, W. Han, W. Zhao, Z. Wu, K. Chang, H. Wang, Y. Wang,
M. S. Dresselhaus, Nano Lett. 2014, 14, 5500. N. Bonini, N. Marzari, N. Pugno, Nat. Mater. 2012, 11, 294.
[33] A. Bachmatiuk, R. Abelin, H. T. Quang, B. Trzebicka, J. Eckert, [63] J.-B. Wu, X. Zhang, M. Ijäs, W.-P. Han, X.-F. Qiao, X.-L. Li, D.-S. Jiang,
M. H. Rummeli, Nanotechnology 2014, 25, 365603. A. C. Ferrari, P.-H. Tan, Nat. Commun. 2014, 5, 5309.
[34] K. Liu, L. Zhang, T. Cao, C. Jin, D. Qiu, Q. Zhou, A. Zettl, P. Yang, [64] S. Huang, L. Liang, X. Ling, A. A. Puretzky, D. B. Geohegan,
S. G. Louie, F. Wang, Nat. Commun. 2014, 5, 4966. B. G. Sumpter, J. Kong, V. Meunier, M. S. Dresselhaus, Nano Lett.
[35] S. Zheng, L. Sun, X. Zhou, F. Liu, Z. Liu, Z. Shen, H. J. Fan, Adv. Opt. 2016, 16, 1435.
Mater. 2015, 3, 1600. [65] J. Kang, J. Li, S.-S. Li, J.-B. Xia, L.-W. Wang, Nano Lett. 2013, 13, 5485.
[36] K. Wang, B. Huang, M. Tian, F. Ceballos, M.-W. Lin, M. Mahjouri- [66] B. Hunt, J. D. Sanchez-Yamagishi, A. F. Young, M. Yankowitz,
Samani, A. Boulesbaa, A. A. Puretzky, C. M. Rouleau, M. Yoon, B. J. LeRoy, K. Watanabe, T. Taniguchi, P. Moon, M. Koshino,
ACS Nano 2016, 10, 6612. P. Jarillo-Herrero, Science 2013, 340, 1427.
[67] G. Chen, L. Jiang, S. Wu, B. Lyu, H. Li, B. L. Chittari, K. Watanabe, [80] P. Nagler, G. Plechinger, M. V. Ballottin, A. Mitioglu, S. Meier,
T. Taniguchi, Z. Shi, J. Jung, Nat. Phys. 2019, 15, 237. N. Paradiso, C. Strunk, A. Chernikov, P. C. Christianen,
[68] W. Choi, I. Akhtar, M. A. Rehman, M. Kim, D. Kang, J. Jung, C. Schüller, 2D Mater. 2017, 4, 025112.
Y. Myung, J. Kim, H. Cheong, Y. Seo, ACS Appl. Mater. Interfaces [81] B. Miller, A. Steinhoff, B. Pano, J. Klein, F. Jahnke, A. Holleitner,
2018, 11, 2470. U. Wurstbauer, Nano Lett. 2017, 17, 5229.
[69] K. L. Seyler, P. Rivera, H. Yu, N. P. Wilson, E. L. Ray, D. G. Mandrus, [82] J. S. Ross, P. Rivera, J. Schaibley, E. Lee-Wong, H. Yu, T. Taniguchi,
J. Yan, W. Yao, X. Xu, Nature 2019, 567, 66. K. Watanabe, J. Yan, D. Mandrus, D. Cobden, Nano Lett. 2017,
[70] L. Wang, E.-M. Shih, A. Ghiotto, L. Xian, D. A. Rhodes, 17, 638.
C. Tan, M. Claassen, D. M. Kennes, Y. Bai, B. Kim, Nat. Mater. [83] F. Wu, T. Lovorn, A. H. MacDonald, Phys. Rev. Lett. 2017, 118, 147401.
2020, 19, 861. [84] H. Heo, J. H. Sung, S. Cha, B.-G. Jang, J.-Y. Kim, G. Jin, D. Lee,
[71] H. Yu, Y. Wang, Q. Tong, X. Xu, W. Yao, Phys. Rev. Lett. 2015, 115, J.-H. Ahn, M.-J. Lee, J. H. Shim, Nat. Commun. 2015, 6, 7372.
187002. [85] M. Liao, Z. Wei, L. Du, Q. Wang, J. Tang, H. Yu, F. Wu, J. Zhao, X. Xu,
[72] H. Yu, G.-B. Liu, J. Tang, X. Xu, W. Yao, Sci. Adv. 2017, 3, e1701696. B. Han, Nat. Commun. 2020, 11, 2153.
[73] P. Rivera, K. L. Seyler, H. Yu, J. R. Schaibley, J. Yan, D. G. Mandrus, [86] L. A. Jauregui, A. Y. Joe, K. Pistunova, D. S. Wild, A. A. High, Y. Zhou,
W. Yao, X. Xu, Science 2016, 351, 688. G. Scuri, K. De Greve, A. Sushko, C.-H. Yu, Science 2019, 366, 870.
[74] P. Nagler, M. V. Ballottin, A. A. Mitioglu, F. Mooshammer, [87] D. Unuchek, A. Ciarrocchi, A. Avsar, Z. Sun, K. Watanabe,
N. Paradiso, C. Strunk, R. Huber, A. Chernikov, P. C. Christianen, T. Taniguchi, A. Kis, Nat. Nanotechnol. 2019, 14, 1104.
C. Schüller, Nat. Commun. 2017, 8, 1551. [88] A. Kormányos, V. Zólyomi, V. I. Fal’ko, G. Burkard, Phys. Rev. B 2018,
[75] C. Jiang, W. Xu, A. Rasmita, Z. Huang, K. Li, Q. Xiong, W.-b. Gao, Nat. 98, 035408.
Commun. 2018, 9, 753. [89] J. R. Schaibley, H. Yu, G. Clark, P. Rivera, J. S. Ross, K. L. Seyler,
[76] W.-T. Hsu, L.-S. Lu, P.-H. Wu, M.-H. Lee, P.-J. Chen, P.-Y. Wu, W. Yao, X. Xu, Nat. Rev. Mater. 2016, 1.
Y.-C. Chou, H.-T. Jeng, L.-J. Li, M.-W. Chu, Nat. Commun. 2018, 9, [90] P. Rivera, H. Yu, K. L. Seyler, N. P. Wilson, W. Yao, X. Xu, Nat.
1356. Nanotechnol. 2018, 13, 1004.
[77] A. Ciarrocchi, D. Unuchek, A. Avsar, K. Watanabe, T. Taniguchi, [91] K. Zhou, D. Wickramaratne, S. Ge, S. Su, A. De, R. K. Lake, Phys.
A. Kis, Nat. Photonics 2019, 13, 131. Chem. Chem. Phys. 2017, 19, 10406.
[78] K. Tran, G. Moody, F. Wu, X. Lu, J. Choi, K. Kim, A. Rai, D. A. Sanchez, [92] F. Wu, T. Lovorn, E. Tutuc, A. H. MacDonald, Phys. Rev. Lett. 2018,
J. Quan, A. Singh, Nature 2019, 567, 71. 121, 026402.
[79] P. Rivera, J. R. Schaibley, A. M. Jones, J. S. Ross, S. Wu, G. Aivazian, [93] M. H. Naik, M. Jain, Phys. Rev. Lett. 2018, 121, 266401.
P. Klement, K. Seyler, G. Clark, N. J. Ghimire, Nat. Commun. 2015, [94] D. A. Ruiz-Tijerina, V. I. Fal’ko, Phys. Rev. B 2019, 99, 125424.
6, 6242. [95] C. Schrade, L. Fu, Phys. Rev. B 2019, 100, 035413.
Bijun Tang received her B.E. degree from Nanyang Technological University (NTU), Singapore, with First
Class Honours (2017). She is currently pursuing a Ph.D. degree under the supervision of Prof. Zheng Liu
at the School of Materials Science and Engineering, NTU. Her research interests mainly focus on the
synthesis and exploration of novel 2D materials.
Haitao Yang is an associate professor at the Institute of Physics, Chinese Academy of Sciences. He
completed his Ph.D. at the Institute of Physics and then worked in Prof. Migaku Takahashi’s group as a
JSPS postdoctorate research fellow at Tohoku University. He also worked in Prof. Shouheng Sun’s group
as a visiting scholar (2016) at Brown University. His current research interests focus on the growth and
physical properties of low-dimensional materials.
Jiadong Zhou is a professor at the Beijing Institute of Technology. He received his B.S degree (2009) at
Shaanxi University of Science & Technology, M.S. degree (2013) at Shanghai Institute of Ceramics,
Chinese Academy of Sciences, and completed his Ph.D. (2017) at NTU, Singapore. He then worked as a
postdoctorate research fellow at the School of Materials Science and Engineering, NTU. His current
research interests focus on the synthesis, characterizations, and applications of novel 2D crystals.
Zheng Liu is an associate professor at the School of Materials Science and Engineering, NTU. He
received his B.S. degrees (2005) at Nankai University (China) and completed his Ph.D. at the National
Center for Nanoscience and Technology (NCNST), China. He then worked in Prof. Pulickel M. Ajayan
and Prof. Jun Lou’s groups as a joint postdoctorate research fellow (2010–2012) and research scientist
(2012–2013) at Rice University. His current research interests focus on the growth and applications of
2D materials with only single-atom thickness, via innovative methods and artificial intelligence
technology.

Small Structures - 2021 - Tang - Recent Advances in Synthesis and Study of 2D Twisted Transition Metal Dichalcogenide

Uploaded by

Copyright:

Available Formats

Small Structures - 2021 - Tang - Recent Advances in Synthesis and Study of 2D Twisted Transition Metal Dichalcogenide

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Small Structures - 2021 - Tang - Recent Advances in Synthesis and Study of 2D Twisted Transition Metal Dichalcogenide

Uploaded by

Copyright:

Available Formats

REVIEW

Recent Advances in Synthesis and Study of 2D Twisted

due to their potential applications in vari-

Small Struct. 2021, 2, 2000153 2000153 (1 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (2 of 17) © 2021 Wiley-VCH GmbH

Table 1. Summary of the preparation of the t-TMD bilayers.

Materials Transfer method Annealing conditions Twist angle produced Ref.

Materials Precursors Growth conditions Twist angle produced Ref.

Small Struct. 2021, 2, 2000153 2000153 (3 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (4 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (5 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (6 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (7 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (8 of 17) © 2021 Wiley-VCH GmbH

Small Struct. 2021, 2, 2000153 2000153 (9 of 17) © 2021 Wiley-VCH GmbH

You might also like