Atomically Thin Heterostructures based on Single-Layer Tungsten ...
Supplemental Information
Atomically Thin Heterostructures based on Single-Layer Tungsten Diselenide and Graphene Yu-Chuan Lin,1 Chih-Yuan S. Chang,2 Ram Krishna Ghosh,3 Jie Li,3 Hui Zhu,4 Rafik Addou,4 Bogdan Diaconescu,5 Taisuke Ohta,5 Xin Peng,4 Ning Lu,4 Moon J. Kim,4 Jeremy T. Robinson,6 Robert M, Wallace,4 Theresa S. Mayer,3 Suman Datta,3 Lain-Jong Li, 2,7* and Joshua A. Robinson 1* 1 Department of Materials Science and Engineering and Center for 2-Dimensional and Layered Materials, The Pennsylvania State University, University Park, Pennsylvania, 16802, United States 2 Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan 3 Department of Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania, 16802, United States 4 Department of Materials Science and Engineering, The University of Texas at Dallas, Richardson, Texas 75080 United States 5 Sandia National Laboratories, Albuquerque, New Mexico, 87185, Untied States 6 Naval Research Laboratory, Washington D.C., 20375, United States 7 Physical Science and Engineering Division, King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia. *E-mail: [email protected] and [email protected]
The following pages include supporting figures and experimental data.
1. Growth of Epitaxial Graphene:1 Epitaxial graphene is grown on diced SiC wafers via sublimation of silicon from 6H-SiC (0001) at 1725 oC for 20 min under 200 Torr argon (Ar) background pressure after hydrogen (H2) etching at 1500 oC for 15 min under 700 Torr 10% H2/Ar mixtures. Typical SiC surface morphology is shown in Fig. S1.
Figure S1 Epitaxial graphene morphology: (a) Growth results in a uniform SiC morphology with SiC (0001) forming the terrace surface and (1-10n) plane forming the step edge. (b) The average terrace width and step height is 3 to 5 µm and 10 nm to 20 nm, respectively.
2. Growth & Properties of WSe2 Layers on Epitaxial Graphene:2 Tungsten tri-oxide (WO3) powder (0.3 g) was placed in a ceramic crucible located in the center of the furnace. The Se powder is placed in a separate ceramic boat up-stream of the WO3, and maintained at 270 °C during the reaction. Graphene substrates were subsequently placed down-stream of the crucible loaded with WO3, and Se and WO3 vapors were transported to the targeting sapphire substrates by an Ar/H2 (80:20) forming gas at 1 Torr of chamber pressure. The center of the hot zone was held at 925 °C for 30 min to 1 hour and the furnace was then naturally cooled to room temperature. The as-grown heterostructures are characterized using atomic force microscopy (AFM) (Fig. S2a-f) and Low energy electron microscopy (Fig. S2g-h), which verify significant fractions of aligned monolayer WSe2 domains are grown on the EG. Raman spectroscopy, transmission electron microscopy (TEM) (Fig. S3), X-ray photoelectron spectroscopy (XPS) (Table S1) were also used to verify the high quality nature of the system. A WITec CRM200 Confocal Raman microscope with a 488 nm laser wavelength is utilized for structural characterization. A BRUKER Dimension with a scan rate of 0.5 Hz was utilized for the AFM measurements. The TEM cross-sectional samples were made via utilizing a NanoLab dual-beam FIB/SEM system. Protective layers of SiO2 and Pt were deposited to protect the interesting region during focused ion beam milling. TEM imaging was performed using a JEOL 2100F operated at 200 kV. CAFM measurement was performed in BRUKER
Dimension. X-ray photoelectron spectroscopy (Table S1) was performed using a monochromatic Al-Kα source (E =1486.7 eV) and an Omicron Argus detector operating with pass energy of 15 eV. The spot size used during the acquisition is equal to 0.5 mm. Core-level spectra taken with 15 sweeps are analyzed with the spectral analysis software analyzer.4
Figure S2 As-grown epitaxial graphene (a) is utilized as the substrate for synthesis of WSe2 monocrystal triangles. Growths of 15 (b) and 30 (c) minutes indicate that surface coverage of the WSe2 is heavily dependent on growth time. At longer growth times, the 1-1.5 µm flakes coalesce to form larger crystalline flakes and films. (AFM image scale bar: 2 µm) (d) 60 min growth of WSe2, highlighted in blue, results in more than 75 % coverage over the graphene surface (SEM image scale: 3 µm). AFM image (e) and distribution of WSe2 island orientation (f) showing that the majority of WSe2 triangles orient at the same direction (0o to the reference) (image scale: 6 µm). (g) Auger electron spectroscopy (AES) image via LEEM acquired at 96.5 eV (Se) and 170.5 eV (W) electron kinetic energy. The brightness in the AES image is proportional with the elemental content of Se and W. The vast majority of the triangular WSe2 islands are atomically thin (i.e. orange circles) with sporadic thicker islands (green circles).
Assessment of the graphene’s layer thickness was carried out using LEEM-IV. A false color image of the LEEM-IV along a line is typically used for this purpose. One such example is shown in Fig 2-2 (b), along with the individual spectra representative for 1GL and 2GL in (c). The electron energy of the characteristic dips for 1GL and 2GL are highlighted by blue and red arrows, respectively. We suspect the slight difference in the energy of the dips in 2GL is due to the slight difference in the interaction between graphene and the underlying SiC substrate. The ‘narrow’ terraces located in between large terraces show multiple dips consistent the existence of many-layer EG.
Figure S2-2 (a) LEEM image acquired at the electron energy 1.6 eV above the vacuum level (reproduced from Fig. 1 (f)). (b) The false color image of the LEEM-IV along the green dash line in (a). Area of 1GL and 2GL are specified by the black arrows. The blue and red arrows specify the energy of the characteristic dips for 1GL and 2GL, respectively. (c) Representative LEEM-IV spectra of 1GL (blue) and 2GL (red) extracted from the same dataset.
Figure S3 (a) The cropped region in crossectional TEM image of WSe2/EG heterostructures acquires the intensity profile (b). The 1L WSe2 region (highlighted in (b)) is roughly estimated to be 0.645 nm in thickness.
Table S1: Summary of Results for X-ray Photoelectron Spectroscopy (“VT” = vapor transport) Reference
Se3d5/2
Se3d3/2
W4f7/2
W4f5/2
XPS
Sample
2
55.0
55.9
32.8
35.0
Non-mono
Selenization of
MgKα
WO3: 1L WSe2 on sapphire
5
54.9
55.8
32.7
34.9
Mono Al Kα1
Exfoliated geological crystal
6
7
55.5
32.9
54.5
35.0
32.3
Non-mono
Selenization of
MgKα
WO3
Synchrotron
Cleaved VT crystal
8,9
31.8
8,9
This work
32.8
54.9
55.8
32.6
34.0
34.9
34.8
Non-mono
Cleaved VT p
MgKα
type crystal
Non-mono
Cleaved VT n
MgKα
type crystal
Mono Al Kα1
Selenization of WO3: 1L WSe2 on epi-graphene
Finally, scanning tunneling microscopy was accomplished in ultra-high vacuum (UHV) under the same conditions described in Ref.3. The WSe2/EG sample was imaged using an Omicron variable temperature scanning tunneling microscope (STM) at room temperature and 10-10 mbar.
The STM images were obtained in the constant-current mode, with an etched tungsten tip.
3. Device Fabrication and other measured I-V curves: The hetero-junction device demonstrated in the main text (Figure 3) is fabricated with electron beam lithography and lift-off of evaporated metal contacts. The process flow is illustrated in Figure S4 (a-e). In the first step, the graphene contact is patterned and developed with electron beam (e-beam) lithography. Subsequently, metal contacts Ti/Au (10nm/40nm) are deposited with low pressure electron beam evaporation (10-7 Torr) after a oxygen plasma treatment to reduce the contact resistance (45s at 100 W, 50 sccm He, 150 sccm O2 at 500 mTorr).10 Then a layer of 30 nm Al2O3 is deposited conformally over the entire substrate with atomic layer deposition (ALD), which serves as a protection layer for subsequent processing steps and a passivation layer. ALD deposited Al2O3 capping layer has been reported as an effective film to substantially improve carrier mobility in 2D materials.11 In the second e-beam lithography step, a pattern of etch regions are defined (Figure S4 (c-d)), including an opening on the Ti/Au pads, and regions for the later WSe2 contacts. The Al2O3 capping layer on these regions are first removed with hydrofluoric acid (Figure S5 (c)), followed by oxygen plasma etching to remove the monolayer WSe2 and few layers of graphene (Figure S4 (d)). This step prevents shorting through the underlying graphene layer after depositing the WSe2 contacts. In the third e-beam step, the WSe2 contact pads and thin lines are defined, and the Al2O3 layer on the WSe2 triangular sheets is removed by hydrofluoric acid prior to the metal deposition. Then 50 nm thick Palladium (Pd) layer is deposited by electron beam evaporation at 10-7 Torr. The high work function Pd contacts with WSe2 have been reported to produce a smaller Schottky barrier, and many orders higher current density compared to Ti/Au contacts.12 Figure S4 (f) shows the optical microscope images of the finalized arrays of hetero-junction devices.
Cross section view
Top view
(a)
(b)
(c)
(d)
(e)
Figure S4 Process flow for the fabrication of the hetero-junction devices on atomically thin heterostructures based on WSe2 and Graphene. (a) Patterning and depositing Ti/Au for graphene contacts; (b) Coating the entire substrate with ALD Al2O3; (c) Define the isolation region on graphene contact pads, and later WSe2 contact regions, and remove the Al2O3 layer; (d) Isolation etching with oxygen plasma to remove the WSe2 and graphene; (e) Patterning the WSe2 contacts (Pd). (f) The optical microscopy image of the arrays of devices.
4. Density Functional Theory (DFT) and Non-Equilibrium Green’s Function (NEGF) Transport Simulation: We
carry
out
DFT
simulations
by
using
Atomistix
Toolkit
(ATK),
(www.quantumwise.com).13 The effective potential is calculated by utilizing an interface supercell (Figure S5) of WSe2/graphene that is periodic in x and y direction and having a vacuum layer of 12 Å in the z direction. The vacuum layer is considered here to isolate the heterostructure from any interaction with its periodic image. To construct the WSe2-graphene interface, we merge the 3x3 supercell consisting of monolayer WSe2 with 4x4 super cell comprising of monolayer graphene. This allows us to reduce the lattice mismatch between the WSe2 and the graphene hexagonal unit cell. Before performing the calculations for determination of electronic properties (such as effective potential) we optimize the structure by using a quasi-Newton method until all the forces acting on the supercell atoms are smaller than 0.02 eV/Å. Post structure optimization, we obtain a van der Waals distance of ~3.53 Å between the WSe2 and the graphene. To construct the two terminal device, we then add Pd atomic layers cleaved along [111] plane to the supercell.
The electronic properties are estimated using the Perdew-Burke-Ernzerhof (PBE) functional under generalized gradient approximations (GGA). To incorporate the long range van der Waals correction arising from the inter-layer interaction within the GGA approximation, we have also included Grimme’s DFT-D2 functional with S6 = 0.75 under PBE functional. A double-ζ polarized basis set is used for the electron wave function. These electron wave functions compare well with converged plane wave basis sets whereas the core electrons are defined by norm-conserving pseudo-potentials. According to the dimension of this supercell we use a k-point sampling of 5x5x1 in the Brillouin zone. In addition, the tolerance parameter was set to 10−5 with maximum steps of 200, and a Pulay mixer algorithm was used as the iteration control parameter with mesh cut-off energy of 10 Hartree.
Figure S5 (a) Atomic structure of the WSe2-Graphene supercell. Energy band structure of (b) monolayer WSe2 and (c) monolayer graphene. The bandgap value of 1.61 eV is in good agreement with literature.14
Figure S6 The effective potential profile in the Pd [111]/WSe2/EG heterostructure supercell as calculated by density functional theory (DFT) along the out of plane direction demonstrates that, the finite barrier to electron transport arising from the WSe2/EG inter-layer gap is higher than the Pd/WSe2 Schottky junction barrier height.
Reference: (1)
Lin, Y.; Lu, N.; Perea-lopez, N.; Li, J.; Lin, Z.; Peng, X.; Lee, C. H.; Browning, P. N.; Bresnehan, M. S.; Calderin, L.; Kim, M. J.; Mayer, T. S.; Robinson, J. A. ACS Nano 2014, 8, 3715–3723.
(2)
Huang, J.-K.; Pu, J.; Hsu, C.-L.; Chiu, M.-H.; Juang, Z.-Y.; Chang, Y.-H.; Chang, W.-H.; Iwasa, Y.; Takenobu, T.; Li, L.-J. ACS Nano 2014, 8, 923–930.
(3)
Wallace, R. M. ECS Trans. 2014, 64, 109–116.
(4)
Herrera-Gómez, A.; Hegedus, A.; Meissner, P. L. Appl. Phys. Lett. 2002, 81, 1014.
(5)
McDonnell, S.; Azcatl, A.; Addou, R.; Gong, C.; Battaglia, C.; Chuang, S.; Cho, K.; Javey, A.; Wallace, R. M. ACS Nano 2014, 8, 6265–6272.
(6)
Salitra, G.; Hodes, G.; Klein, E.; Tenne, R. Thin Solid Films 1994, 245, 180–185.
(7)
Schellenberger, A.; Schlaf, R.; Mayer, T.; Holub-Krappe, E.; Pettenkofer, C.; Jaegermann, W.; Ditzinger, U. A.; Neddermeyer, H. Surf. Sci. 1991, 241, L25–L29.
(8)
Lang, O.; Tomm, Y.; Schlaf, R.; Pettenkofer, C.; Jaegermann, W. J. Appl. Phys. 1994, 75, 7814–7820.
(9)
Henrion, O.; Jaegermann, W. Ber. Bunsen-Ges. Phys. Chem. 1998, 102, 96–102.
(10) Robinson, J. A.; Labella, M.; Zhu, M.; Hollander, M.; Kasarda, R.; Hughes, Z.; Trumbull, K.; Cavalero, R.; Snyder, D. Appl. Phys. Lett. 2011, 98, 053103. (11)
Das, S.; Appenzeller, J. Nano Lett. 2013, 13, 3396–3402.
(12)
Fang, H.; Chuang, S.; Chang, T. C.; Takei, K.; Takahashi, T.; Javey, A. Nano Lett. 2012, 12, 3788–3792.
(13)
Brandbyge, M.; Mozos, J.-L.; Ordejón, P.; Taylor, J.; Stokbro, K. Phys. Rev. B 2002, 65, 165401.
(14)
Ghosh, R. K.; Mahapatra, S. IEEE J. Electron Devices Soc. 2013, 1, 175–180.
Atomically Thin Heterostructures based on Single-Layer Tungsten Diselenide and Graphene Yu-Chuan Lin,1 Chih-Yuan S. Chang,2 Ram Krishna Ghosh,3 Jie Li,3 Hui Zhu,4 Rafik Addou,4 Bogdan Diaconescu,5 Taisuke Ohta,5 Xin Peng,4 Ning Lu,4 Moon J. Kim,4 Jeremy T. Robinson,6 Robert M, Wallace,4 Theresa S. Mayer,3 Suman Datta,3 Lain-Jong Li, 2,7* and Joshua A. Robinson 1* 1 Department of Materials Science and Engineering and Center for 2-Dimensional and Layered Materials, The Pennsylvania State University, University Park, Pennsylvania, 16802, United States 2 Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 10617, Taiwan 3 Department of Electrical Engineering, The Pennsylvania State University, University Park, Pennsylvania, 16802, United States 4 Department of Materials Science and Engineering, The University of Texas at Dallas, Richardson, Texas 75080 United States 5 Sandia National Laboratories, Albuquerque, New Mexico, 87185, Untied States 6 Naval Research Laboratory, Washington D.C., 20375, United States 7 Physical Science and Engineering Division, King Abdullah University of Science and Technology, Thuwal 23955-6900, Saudi Arabia. *E-mail: [email protected] and [email protected]
The following pages include supporting figures and experimental data.
1. Growth of Epitaxial Graphene:1 Epitaxial graphene is grown on diced SiC wafers via sublimation of silicon from 6H-SiC (0001) at 1725 oC for 20 min under 200 Torr argon (Ar) background pressure after hydrogen (H2) etching at 1500 oC for 15 min under 700 Torr 10% H2/Ar mixtures. Typical SiC surface morphology is shown in Fig. S1.
Figure S1 Epitaxial graphene morphology: (a) Growth results in a uniform SiC morphology with SiC (0001) forming the terrace surface and (1-10n) plane forming the step edge. (b) The average terrace width and step height is 3 to 5 µm and 10 nm to 20 nm, respectively.
2. Growth & Properties of WSe2 Layers on Epitaxial Graphene:2 Tungsten tri-oxide (WO3) powder (0.3 g) was placed in a ceramic crucible located in the center of the furnace. The Se powder is placed in a separate ceramic boat up-stream of the WO3, and maintained at 270 °C during the reaction. Graphene substrates were subsequently placed down-stream of the crucible loaded with WO3, and Se and WO3 vapors were transported to the targeting sapphire substrates by an Ar/H2 (80:20) forming gas at 1 Torr of chamber pressure. The center of the hot zone was held at 925 °C for 30 min to 1 hour and the furnace was then naturally cooled to room temperature. The as-grown heterostructures are characterized using atomic force microscopy (AFM) (Fig. S2a-f) and Low energy electron microscopy (Fig. S2g-h), which verify significant fractions of aligned monolayer WSe2 domains are grown on the EG. Raman spectroscopy, transmission electron microscopy (TEM) (Fig. S3), X-ray photoelectron spectroscopy (XPS) (Table S1) were also used to verify the high quality nature of the system. A WITec CRM200 Confocal Raman microscope with a 488 nm laser wavelength is utilized for structural characterization. A BRUKER Dimension with a scan rate of 0.5 Hz was utilized for the AFM measurements. The TEM cross-sectional samples were made via utilizing a NanoLab dual-beam FIB/SEM system. Protective layers of SiO2 and Pt were deposited to protect the interesting region during focused ion beam milling. TEM imaging was performed using a JEOL 2100F operated at 200 kV. CAFM measurement was performed in BRUKER
Dimension. X-ray photoelectron spectroscopy (Table S1) was performed using a monochromatic Al-Kα source (E =1486.7 eV) and an Omicron Argus detector operating with pass energy of 15 eV. The spot size used during the acquisition is equal to 0.5 mm. Core-level spectra taken with 15 sweeps are analyzed with the spectral analysis software analyzer.4
Figure S2 As-grown epitaxial graphene (a) is utilized as the substrate for synthesis of WSe2 monocrystal triangles. Growths of 15 (b) and 30 (c) minutes indicate that surface coverage of the WSe2 is heavily dependent on growth time. At longer growth times, the 1-1.5 µm flakes coalesce to form larger crystalline flakes and films. (AFM image scale bar: 2 µm) (d) 60 min growth of WSe2, highlighted in blue, results in more than 75 % coverage over the graphene surface (SEM image scale: 3 µm). AFM image (e) and distribution of WSe2 island orientation (f) showing that the majority of WSe2 triangles orient at the same direction (0o to the reference) (image scale: 6 µm). (g) Auger electron spectroscopy (AES) image via LEEM acquired at 96.5 eV (Se) and 170.5 eV (W) electron kinetic energy. The brightness in the AES image is proportional with the elemental content of Se and W. The vast majority of the triangular WSe2 islands are atomically thin (i.e. orange circles) with sporadic thicker islands (green circles).
Assessment of the graphene’s layer thickness was carried out using LEEM-IV. A false color image of the LEEM-IV along a line is typically used for this purpose. One such example is shown in Fig 2-2 (b), along with the individual spectra representative for 1GL and 2GL in (c). The electron energy of the characteristic dips for 1GL and 2GL are highlighted by blue and red arrows, respectively. We suspect the slight difference in the energy of the dips in 2GL is due to the slight difference in the interaction between graphene and the underlying SiC substrate. The ‘narrow’ terraces located in between large terraces show multiple dips consistent the existence of many-layer EG.
Figure S2-2 (a) LEEM image acquired at the electron energy 1.6 eV above the vacuum level (reproduced from Fig. 1 (f)). (b) The false color image of the LEEM-IV along the green dash line in (a). Area of 1GL and 2GL are specified by the black arrows. The blue and red arrows specify the energy of the characteristic dips for 1GL and 2GL, respectively. (c) Representative LEEM-IV spectra of 1GL (blue) and 2GL (red) extracted from the same dataset.
Figure S3 (a) The cropped region in crossectional TEM image of WSe2/EG heterostructures acquires the intensity profile (b). The 1L WSe2 region (highlighted in (b)) is roughly estimated to be 0.645 nm in thickness.
Table S1: Summary of Results for X-ray Photoelectron Spectroscopy (“VT” = vapor transport) Reference
Se3d5/2
Se3d3/2
W4f7/2
W4f5/2
XPS
Sample
2
55.0
55.9
32.8
35.0
Non-mono
Selenization of
MgKα
WO3: 1L WSe2 on sapphire
5
54.9
55.8
32.7
34.9
Mono Al Kα1
Exfoliated geological crystal
6
7
55.5
32.9
54.5
35.0
32.3
Non-mono
Selenization of
MgKα
WO3
Synchrotron
Cleaved VT crystal
8,9
31.8
8,9
This work
32.8
54.9
55.8
32.6
34.0
34.9
34.8
Non-mono
Cleaved VT p
MgKα
type crystal
Non-mono
Cleaved VT n
MgKα
type crystal
Mono Al Kα1
Selenization of WO3: 1L WSe2 on epi-graphene
Finally, scanning tunneling microscopy was accomplished in ultra-high vacuum (UHV) under the same conditions described in Ref.3. The WSe2/EG sample was imaged using an Omicron variable temperature scanning tunneling microscope (STM) at room temperature and 10-10 mbar.
The STM images were obtained in the constant-current mode, with an etched tungsten tip.
3. Device Fabrication and other measured I-V curves: The hetero-junction device demonstrated in the main text (Figure 3) is fabricated with electron beam lithography and lift-off of evaporated metal contacts. The process flow is illustrated in Figure S4 (a-e). In the first step, the graphene contact is patterned and developed with electron beam (e-beam) lithography. Subsequently, metal contacts Ti/Au (10nm/40nm) are deposited with low pressure electron beam evaporation (10-7 Torr) after a oxygen plasma treatment to reduce the contact resistance (45s at 100 W, 50 sccm He, 150 sccm O2 at 500 mTorr).10 Then a layer of 30 nm Al2O3 is deposited conformally over the entire substrate with atomic layer deposition (ALD), which serves as a protection layer for subsequent processing steps and a passivation layer. ALD deposited Al2O3 capping layer has been reported as an effective film to substantially improve carrier mobility in 2D materials.11 In the second e-beam lithography step, a pattern of etch regions are defined (Figure S4 (c-d)), including an opening on the Ti/Au pads, and regions for the later WSe2 contacts. The Al2O3 capping layer on these regions are first removed with hydrofluoric acid (Figure S5 (c)), followed by oxygen plasma etching to remove the monolayer WSe2 and few layers of graphene (Figure S4 (d)). This step prevents shorting through the underlying graphene layer after depositing the WSe2 contacts. In the third e-beam step, the WSe2 contact pads and thin lines are defined, and the Al2O3 layer on the WSe2 triangular sheets is removed by hydrofluoric acid prior to the metal deposition. Then 50 nm thick Palladium (Pd) layer is deposited by electron beam evaporation at 10-7 Torr. The high work function Pd contacts with WSe2 have been reported to produce a smaller Schottky barrier, and many orders higher current density compared to Ti/Au contacts.12 Figure S4 (f) shows the optical microscope images of the finalized arrays of hetero-junction devices.
Cross section view
Top view
(a)
(b)
(c)
(d)
(e)
Figure S4 Process flow for the fabrication of the hetero-junction devices on atomically thin heterostructures based on WSe2 and Graphene. (a) Patterning and depositing Ti/Au for graphene contacts; (b) Coating the entire substrate with ALD Al2O3; (c) Define the isolation region on graphene contact pads, and later WSe2 contact regions, and remove the Al2O3 layer; (d) Isolation etching with oxygen plasma to remove the WSe2 and graphene; (e) Patterning the WSe2 contacts (Pd). (f) The optical microscopy image of the arrays of devices.
4. Density Functional Theory (DFT) and Non-Equilibrium Green’s Function (NEGF) Transport Simulation: We
carry
out
DFT
simulations
by
using
Atomistix
Toolkit
(ATK),
(www.quantumwise.com).13 The effective potential is calculated by utilizing an interface supercell (Figure S5) of WSe2/graphene that is periodic in x and y direction and having a vacuum layer of 12 Å in the z direction. The vacuum layer is considered here to isolate the heterostructure from any interaction with its periodic image. To construct the WSe2-graphene interface, we merge the 3x3 supercell consisting of monolayer WSe2 with 4x4 super cell comprising of monolayer graphene. This allows us to reduce the lattice mismatch between the WSe2 and the graphene hexagonal unit cell. Before performing the calculations for determination of electronic properties (such as effective potential) we optimize the structure by using a quasi-Newton method until all the forces acting on the supercell atoms are smaller than 0.02 eV/Å. Post structure optimization, we obtain a van der Waals distance of ~3.53 Å between the WSe2 and the graphene. To construct the two terminal device, we then add Pd atomic layers cleaved along [111] plane to the supercell.
The electronic properties are estimated using the Perdew-Burke-Ernzerhof (PBE) functional under generalized gradient approximations (GGA). To incorporate the long range van der Waals correction arising from the inter-layer interaction within the GGA approximation, we have also included Grimme’s DFT-D2 functional with S6 = 0.75 under PBE functional. A double-ζ polarized basis set is used for the electron wave function. These electron wave functions compare well with converged plane wave basis sets whereas the core electrons are defined by norm-conserving pseudo-potentials. According to the dimension of this supercell we use a k-point sampling of 5x5x1 in the Brillouin zone. In addition, the tolerance parameter was set to 10−5 with maximum steps of 200, and a Pulay mixer algorithm was used as the iteration control parameter with mesh cut-off energy of 10 Hartree.
Figure S5 (a) Atomic structure of the WSe2-Graphene supercell. Energy band structure of (b) monolayer WSe2 and (c) monolayer graphene. The bandgap value of 1.61 eV is in good agreement with literature.14
Figure S6 The effective potential profile in the Pd [111]/WSe2/EG heterostructure supercell as calculated by density functional theory (DFT) along the out of plane direction demonstrates that, the finite barrier to electron transport arising from the WSe2/EG inter-layer gap is higher than the Pd/WSe2 Schottky junction barrier height.
Reference: (1)
Lin, Y.; Lu, N.; Perea-lopez, N.; Li, J.; Lin, Z.; Peng, X.; Lee, C. H.; Browning, P. N.; Bresnehan, M. S.; Calderin, L.; Kim, M. J.; Mayer, T. S.; Robinson, J. A. ACS Nano 2014, 8, 3715–3723.
(2)
Huang, J.-K.; Pu, J.; Hsu, C.-L.; Chiu, M.-H.; Juang, Z.-Y.; Chang, Y.-H.; Chang, W.-H.; Iwasa, Y.; Takenobu, T.; Li, L.-J. ACS Nano 2014, 8, 923–930.
(3)
Wallace, R. M. ECS Trans. 2014, 64, 109–116.
(4)
Herrera-Gómez, A.; Hegedus, A.; Meissner, P. L. Appl. Phys. Lett. 2002, 81, 1014.
(5)
McDonnell, S.; Azcatl, A.; Addou, R.; Gong, C.; Battaglia, C.; Chuang, S.; Cho, K.; Javey, A.; Wallace, R. M. ACS Nano 2014, 8, 6265–6272.
(6)
Salitra, G.; Hodes, G.; Klein, E.; Tenne, R. Thin Solid Films 1994, 245, 180–185.
(7)
Schellenberger, A.; Schlaf, R.; Mayer, T.; Holub-Krappe, E.; Pettenkofer, C.; Jaegermann, W.; Ditzinger, U. A.; Neddermeyer, H. Surf. Sci. 1991, 241, L25–L29.
(8)
Lang, O.; Tomm, Y.; Schlaf, R.; Pettenkofer, C.; Jaegermann, W. J. Appl. Phys. 1994, 75, 7814–7820.
(9)
Henrion, O.; Jaegermann, W. Ber. Bunsen-Ges. Phys. Chem. 1998, 102, 96–102.
(10) Robinson, J. A.; Labella, M.; Zhu, M.; Hollander, M.; Kasarda, R.; Hughes, Z.; Trumbull, K.; Cavalero, R.; Snyder, D. Appl. Phys. Lett. 2011, 98, 053103. (11)
Das, S.; Appenzeller, J. Nano Lett. 2013, 13, 3396–3402.
(12)
Fang, H.; Chuang, S.; Chang, T. C.; Takei, K.; Takahashi, T.; Javey, A. Nano Lett. 2012, 12, 3788–3792.
(13)
Brandbyge, M.; Mozos, J.-L.; Ordejón, P.; Taylor, J.; Stokbro, K. Phys. Rev. B 2002, 65, 165401.
(14)
Ghosh, R. K.; Mahapatra, S. IEEE J. Electron Devices Soc. 2013, 1, 175–180.
