Ong Kim Lea,
Viorel Chihaiab,
My-Phuong Pham-Ho*ac and
Do Ngoc Son*a
aHo Chi Minh City University of Technology, VNU-HCM, 268 Ly Thuong Kiet Street, District 10, Ho Chi Minh City, Vietnam. E-mail: phmphuong@hcmut.edu.vn; dnson@hcmut.edu.vn
bInstitute of Physical Chemistry “Ilie Murgulescu” of the Romanian Academy, Splaiul Independentei 202, Sector 6, 060021 Bucharest, Romania
cInstitute for Computational Science and Technology, Quang Trung Software City, SBI Building, Street No. 3, Tan Chanh Hiep Ward, District 12, Ho Chi Minh City, Vietnam
First published on 27th January 2020
MoS2 is one of the well-known transition metal dichalcogenides. The moderate bandgap of monolayer MoS2 is fascinating for the new generation of optoelectronic devices. Unfortunately, MoS2 is sensitive to gases in the environment causing its original electronic properties to be modified unexpectedly. This problem has been solved by coating MoS2 with polymers such as polyethyleneimine (PEI). Furthermore, the application of pressure is also an effective method to modify the physical properties of MoS2. However, the effects of polyethyleneimine and pressure on the electronic and optical properties of monolayer MoS2 remain unknown. Therefore, we elucidated this matter by using density functional theory calculations. The results showed that the adsorption of the PEI molecule significantly reduces the width of the direct bandgap of the monolayer MoS2 to 0.55 eV because of the occurrence of the new energy levels in the bandgap region due to the contribution of the N-2pz state of the PEI molecule. Remarkably, the transition from semiconductor to metal of the monolayer MoS2 and the MoS2/PEI system occurs at the tensile pressure of 24.95 and 21.79 GPa, respectively. The bandgap of these systems approaches 0 eV at the corresponding pressures. Importantly, new peaks in the optical spectrum of the clean MoS2 and MoS2/PEI appear in the ultraviolet region under compressive pressures and the infrared region under tensile strains.
Furthermore, control of the electronic and optical properties of the monolayer MoS2 by changing the crystal structure,10 layer thickness,4 and applying strain11 was achieved. The results showed that the homogeneous biaxial tensile strain of around 10% led to the semiconductor-to-metal transition in all semiconducting TMDs.12 Several theoretical and experimental studies showed that the bandgap of the monolayer MoS2 decreases as the tensile strain increases,12,13 while it is a parabola for compression.14–17 Fan et al. demonstrated that the bandgap at the K point changes from direct to indirect with compressive pressure.14 Generally, the influence on the electronic structure properties of monolayer MoS2 depends on the type of applied strain. Experimentally, it is difficult to exert a homogeneous hydrostatic strain on a 2D material sample and can only generate a biaxial compressive strain up to 0.2% with elaborate setups.18 In a recent study, the monolayer MoS2 was investigated under extreme hydrostatic pressures of up to 30 GPa. Although a higher pressure is not experimentally achievable at present, a metallization at the compressive pressure of 68 GPa was predicted by theoretical calculations.19 In addition to the ability to tune the electronic properties of the monolayer MoS2, the applied strain was found to modify the photoluminescence (PL) intensity. The PL maximum peak of the monolayer MoS2 exhibited a blue shift20,21 at the rate of approximately 20 meV GPa−1. At the pressure of 25 GPa, both real and imaginary parts of the dielectric function were shifted to red, and the peak height increases correspondence with an enhancement of the optical absorption.21 These results demonstrated that one can adjust the electronic and optical properties of MoS2 by applying pressures. Interestingly, the monolayer MoS2 phototransistor exhibited a better photo-responsivity compared to the graphene-based devices.22
The effects of organic molecules adsorption such as TCNQ, TCNE, TTF, BV, F4TCNQ, and benzene on the electronic and optical properties of the monolayer MoS2 have been studied in the literature.23–26 A reduction of the bandgap was found due to the existence of the flat molecular levels of the organic molecules in the bandgap region of the monolayer MoS2. The theoretical study showed that the adsorption of TCNQ significantly enhances the peak height of the optical absorption of the monolayer MoS2 in the infrared region of the spectrum.23,24 Du et al. experimentally studied the adsorption of the PEI on multilayer MoS2 field-effect transistors.8 The experimental results demonstrated that a reduction of 2.6 times in sheet resistance and 1.2 times in contact resistance have been achieved. The enhancement of electrical properties was reflected in an improvement of 70% for ON-current and 50% for extrinsic field-effect mobility. Our previous study showed that the adsorption of the PEI molecule enhances the luminescence ability of the multi-layer MoS2 FET from 0.14 to 4.41 AW−1.9 However, the influences of both pressure and the PEI adsorption on the electronic and optical properties of MoS2 remain unknown. Therefore, in the present work, we are going to clarify this matter by using the density functional theory calculations.
Adsorption energy was calculated to understand the binding strength of the PEI fragment on MoS2 through the formula:
Ea = Esub+PEI − (Esub + EPEI), | (1) |
The pressure was estimated from the energy cost over the change of volume by:19
(2) |
Optical properties were determined through the frequency-dependent complex dielectric function, ε(ω) = ε1(ω) + iε2(ω)21,38 which is the measure of the light absorption and emission of material. The imaginary part of the dielectric constant ε2(ω) was computed by using the following expression:38
(3) |
(4) |
To study the influences of the pressure, we varied the lattice constant with the constraint a = b of the clean MoS2 and the MoS2/PEI system and explored the differences of two systems before and after changing the lattice constant. We considered the following values of the lattice constant: 2.84, 3.00, 3.18, 3.32, 3.48, and 3.64 Å. Two first values correspond to compressive strains and three last values correspond to the tensile strains.
The MoS2/PEI system was prepared by loading the PEI fragment on the optimized monolayer MoS2. The MoS2/PEI system was allowed to fully relax for every step of the calculations. The PEI fragment was initially placed parallel to the surface of the monolayer MoS2 with a vertical distance of around 3.5 Å. Since the stability of the PEI adsorption correlates to the number of nitrogen atoms associated with the surface of the substrate, the stable geometry of the MoS2/PEI system was obtained with a flat structure of the PEI fragment on the surface of the monolayer MoS2. We considered several possible adsorption sites of the PEI molecule on the surface of MoS2, see Fig. 2. After performing the geometric optimization, we obtained the optimized structures of MoS2/PEI with the total energies of −741.39, −741.38, −741.43, and −741.41 eV for those in Fig. 2a–d, respectively. Based on these energies, we found the most favorable adsorption configuration of the PEI molecule as the N atom with three alkyl groups, denoted as Ny, positioned on the top of the S atom (Fig. 2c). The average vertical distance from the lowest nitrogen atoms, the lowest carbon atoms and the lowest hydrogen atoms of PEI to the surface of the substrate are 3.63, 3.76, and 2.37 Å, respectively.
We now perform the structural optimization for each variation of the lattice constant for the most favorable adsorption configuration of the PEI molecule on the MoS2 substrate. The adsorption energy of the PEI fragment on the substrate and the distances from this molecule to the substrate are listed in Table 1. We found that the distances increase slightly with the increase of the lattice constant.
a (Å) | Ea (eV) | dN–S (Å) | dC–S (Å) | dH–S (Å) |
---|---|---|---|---|
2.84 | 0.139 | 3.53 | 3.63 | 2.26 |
3.00 | 0.063 | 3.57 | 3.72 | 2.30 |
3.18 | −0.007 | 3.63 | 3.76 | 2.37 |
3.32 | −0.085 | 3.68 | 3.81 | 2.42 |
3.48 | −0.315 | 3.73 | 3.88 | 2.46 |
3.64 | −0.479 | 3.83 | 3.93 | 2.51 |
Furthermore, Fig. 3 shows that the adsorption energy becomes more negative; and hence, the favorable order of the PEI adsorption increases with the increase of the lattice constant. For the compression, with the lattice constant a < 3.18 Å, the adsorption energy is positive, which indicates that the adsorption of the PEI molecule is unstable on the surface of the monolayer MoS2. In contrast, for the cases of without strain (a = 3.18 Å) and tensile strains (a > 3.18 Å), the adsorption energy is negative. This result implies that the adsorption of the PEI molecule is stable. We found that the higher the lattice constant, the stronger the PEI molecule adsorbs. However, the interaction between the PEI molecule and the monolayer MoS2 is weak and the PEI molecule physisorbed on the MoS2. Furthermore, the phonon vibrational energy of the stretching mode of PEI is 0.43 eV. This value is comparable to the adsorption energy of PEI; therefore, the adsorption of PEI is thermodynamically unstable except for applying a high enough tensile pressure to the MoS2/PEI system.
Fig. 3 The adsorption energy of the PEI fragment on the monolayer MoS2 as a function of the lattice constant. |
The structural information was shown in Table 2 for MoS2 and Table 3 for PEI with the change of lattice constant. We found that the average bond length between S1–S2, Mo–Mo, Mo–S, and the average bond angles Mo1–S1–Mo2, S1–Mo2–S2 increase, while the thickness of the MoS2 slab shown by S1–S3 and also the angle S1–Mo1–S3 decrease with the increase of lattice constant. Although the structural parameters for the MoS2 monolayer have been modified, the MoS2 monolayer under pressure remains the 2H phase similar to that of without strain. We can confirm this information by Fig. 4, where the top view of the MoS2 shows the honeycomb structure and the side views exhibit a similar structure for all considering lattice constants of 2.84, 3.18, and 3.64 Å. Table 3 shows the structural information of PEI that the bond lengths of N–C and C–C, and the bond angles of C–C–N and C–N–C somehow increase with the increase of the lattice constant.
a (Å) | 2.84 | 3.00 | 3.18 | 3.32 | 3.48 | 3.64 | |
Average bond length (Å) | S1–S2 | 2.85 | 3.06 | 3.18 | 3.32 | 3.48 | 3.57 |
S1–S3 | 3.38 | 3.26 | 3.13 | 3.05 | 2.95 | 2.88 | |
Mo1–Mo2 | 2.84 | 3.00 | 3.19 | 3.31 | 3.46 | 3.66 | |
Mo–S | 2.37 | 2.38 | 2.41 | 2.45 | 2.49 | 2.53 | |
Average bond angle (°) | Mo1–S1–Mo2 | 74.17 | 78.21 | 82.30 | 85.28 | 88.60 | 91.29 |
S1–Mo2–S2 | 74.32 | 78.28 | 82.30 | 85.08 | 88.54 | 90.89 | |
S1–Mo1–S3 | 91.66 | 87.65 | 80.54 | 79.42 | 73.07 | 70.98 |
a (Å) | 2.84 | 3.00 | 3.18 | 3.32 | 3.48 | 3.64 | |
Bond length (Å) | Ny–C5 | 1.425 | 1.437 | 1.467 | 1.494 | 1.525 | 1.519 |
Ny–C4 | 1.427 | 1.438 | 1.462 | 1.484 | 1.501 | 1.486 | |
Ny–C7 | 1.443 | 1.451 | 1.471 | 1.488 | 1.498 | 1.483 | |
N1–C1 | 1.457 | 1.459 | 1.465 | 1.494 | 1.468 | 1.455 | |
N2–C2 | 1.450 | 1.444 | 1.461 | 1.472 | 1.486 | 1.474 | |
N2–C3 | 1.436 | 1.451 | 1.463 | 1.478 | 1.488 | 1.469 | |
N3–C6 | 1.459 | 1.461 | 1.464 | 1.470 | 1.469 | 1.459 | |
N4–C8 | 1.458 | 1.460 | 1.463 | 1.466 | 1.461 | 1.448 | |
C7–C8 | 1.509 | 1.519 | 1.543 | 1.570 | 1.615 | 1.635 | |
C3–C4 | 1.512 | 1.524 | 1.538 | 1.556 | 1.591 | 1.599 | |
C5–C6 | 1.508 | 1.514 | 1.537 | 1.558 | 1.588 | 1.578 | |
Bond angle (°) | C6C5Ny | 111.32 | 113.10 | 116.54 | 119.13 | 122.96 | 123.74 |
C4C3N2 | 110.41 | 112.37 | 115.27 | 117.23 | 118.62 | 118.63 | |
C8C7Ny | 109.56 | 110.63 | 113.54 | 115.63 | 119.09 | 119.27 | |
C3C4Ny | 109.81 | 111.11 | 115.02 | 117.74 | 121.79 | 122.44 | |
C3N2C2 | 109.57 | 112.15 | 116.40 | 119.52 | 123.78 | 126.79 | |
C5NyC4 | 113.45 | 113.52 | 114.45 | 115.11 | 116.75 | 116.77 | |
C7NyC4 | 112.10 | 113.19 | 113.98 | 114.69 | 116.59 | 118.82 | |
C7NyC5 | 111.90 | 110.93 | 111.71 | 112.22 | 113.97 | 111.86 |
Fig. 4 The upper and lower panels are the top and side views of the MoS2 slab with the change of lattice constant, respectively. |
Fig. 6 shows that, at zero pressure, the monolayer MoS2 has a direct bandgap of 1.68 eV at the K point in the Brillouin zone, which is in good agreement with the previous publications.4,32,41 The total electronic density of states shows that the valence band maximum (VBM) and the conduction band minimum (CBM) are contributed mainly by Mo-4d states and weakly by S-3p states. The adsorption of the PEI molecule makes the energy levels of the MoS2/PEI system lower than that of the isolated monolayer MoS2, simultaneously creates new flat energy levels in the bandgap region which significantly decreases the direct bandgap of the monolayer MoS2 to 0.55 eV. The new energy levels appear in the region of the valence band of MoS2/PEI implying that the monolayer MoS2 became an n-type semiconductor upon the adsorption of the PEI molecule. The new flat energy levels mainly come from the contribution of the N-2p lone pair. Notably, at non-zero pressures, the bandgap of MoS2 and MoS2/PEI becomes narrower than that at zero pressure and gradually approaches 0 eV at some specific values of the pressure. When MoS2 and MoS2/PEI still show the semiconductor property, the significant difference between the electronic properties of two systems is that the bandgap of MoS2 changes from direct to indirect upon the variation of pressure, while MoS2/PEI remains direct bandgap with a smaller value due to the flat characteristics of the new energy levels but at a different position of the k-point mesh.
Fig. 6 The band structure and total density of states of MoS2 (upper panel) and MoS2/PEI (lower panel) with the variation of pressure. |
The pressure correspondence to the variation of the lattice constant of the MoS2 and MoS2/PEI systems was calculated by using the expression (2). The detailed values of the bandgap versus the pressure and the lattice constant were listed in Table 4. We found that the pressure varies from −56.18 to 41.14 GPa and −50.48 to 31.62 GPa for the lattice constant from 2.84 to 3.64 Å for the MoS2 and MoS2/PEI systems, respectively. The negative and positive signs represent the compressive and tensile strains in that order. We see that the pressure scale for the isolated MoS2 is similar to that for the MoS2/PEI system at the same value of the lattice constant.
a (Å) | MoS2 | MoS2/PEI | ||
---|---|---|---|---|
P (GPa) | Eg (eV) | P (GPa) | Eg (eV) | |
2.84 | −56.18 | 1.05 | −50.48 | 0.53 |
3.00 | −32.44 | 1.55 | −38.51 | 0.81 |
3.04 | −27.65 | 1.72 | −22.44 | 0.83 |
3.12 | −10.69 | 1.80 | −11.94 | 0.82 |
3.16 | −9.46 | 1.75 | −8.12 | 0.69 |
3.18 | 0 | 1.68 | 0 | 0.55 |
3.20 | 1.53 | 1.61 | 0.95 | 0.46 |
3.24 | 5.04 | 1.29 | 5.74 | 0.02 |
3.32 | 10.13 | 0.81 | 10.25 | 0.01 |
3.48 | 24.95 | 0 | 21.79 | 0 |
3.64 | 41.14 | 0 | 31.62 | 0 |
The bandgap versus the pressure is visualized in Fig. 7. We found that the behavior of the bandgap for the isolated MoS2 is similar to that for MoS2/PEI, which is a parabola with a maximum value of 1.80 and 0.83 eV at the pressure of −10.69 and −22.44 GPa for the isolated MoS2 and MoS2/PEI systems, respectively. Furthermore, when the tensile pressure approaches 24.95 and 21.79 GPa, the bandgap approaches 0 eV for these systems in that order. This result implies that the monolayer MoS2 and MoS2/PEI systems transit from semiconductor to metal at the obtained values of the tensile pressure. The transition was found to be a little easier for the MoS2/PEI than the monolayer MoS2. The parabolic behavior of the bandgap is in good agreement with the experiment for the isolated MoS2.16
The amount of the charge exchange between the monolayer MoS2 and the PEI fragment was presented by the Bader point charge in Table 5. The plus and minus signs denote the charge gain and loss, respectively. Particularly, the hydrogen and carbon atoms donate while the nitrogen atoms gain charge. The total outcome is that the PEI molecule donates and the MoS2 gains charge. The charge exchange was found to inversely proportional to the bandgap. The higher the charge exchange, the narrower the bandgap becomes. This result supports the nature of the n-type doping to the substrate by the PEI polymer.
P (GPa) | MoS2/PEI | Eg (eV) | F1 | F2 | F3 | F | ||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
23H | 8C | 5N | 25Mo | 50S | PEI | MoS2 | ||||||
−50.48 | −1.62 | −3.21 | 4.79 | −40.27 | 40.32 | −0.05 | 0.05 | 0.53 | −28.97 | −35.66 | 54.67 | −9.96 |
−38.51 | −1.91 | −3.09 | 4.96 | −43.45 | 43.48 | −0.04 | 0.04 | 0.81 | −36.07 | −36.08 | 60.41 | −11.74 |
0 | −1.94 | −3.11 | 5.01 | −43.66 | 43.70 | −0.04 | 0.04 | 0.55 | −35.84 | −36.14 | 60.32 | −11.67 |
10.25 | −1.95 | −3.10 | 4.92 | −43.27 | 43.40 | −0.13 | 0.13 | 0.01 | −34.95 | −35.30 | 57.99 | −12.25 |
21.79 | −2.52 | −2.97 | 5.07 | −42.85 | 43.27 | −0.42 | 0.42 | 0 | −44.24 | −33.16 | 58.80 | −18.60 |
31.62 | −2.76 | −3.06 | 5.24 | −42.01 | 42.59 | −0.58 | 0.58 | 0 | −46.76 | −33.21 | 58.26 | −21.71 |
In common sense, a simple adsorbate will approach the substrate surface as its adsorption energy becomes more negative. However, for the complicated structure as the PEI molecule, as shown in Table 1, the distances from the nearest N, C, and H atoms of PEI to the surface of the MoS2 monolayer exhibit that the PEI molecule is slightly moving away from the surface as the lattice constant increases with the more negative adsorption energy. This observation seems to be contradicted; however, it can be explained in terms of the attractive electrostatic force between the PEI and the MoS2, which increases with the increase of the lattice constant. As shown in Table 5, the attractive force was found for the pairs of H–S and C–S and the repulsive force for the pairs of N–S atoms. The N–S repulsive force is strong enough to win the H–S and C–S attractive forces; and hence, it somehow lifts the whole PEI molecule away from the MoS2 surface. However, the total electrostatic force F remains the increasing attractive property with an increase in the pressure. The increasing attractive property of the total force is attributed mainly to the increase of the attractive interaction of the H atoms with the S atoms.
Fig. 8 presents the total DOS and the orbital projected DOS of the monolayer MoS2 and MoS2/PEI systems. At zero pressure, the total DOS exhibits a high symmetry of spin-up and spin-down components implying the non-magnetic properties of both MoS2 and MoS2/PEI systems. For the negative pressures, the projected DOS of the monolayer MoS2 shows that the contribution to the VBM and CBM is attributed to the dx2−y2 orbital of the Mo atoms and the px state of the S atoms. For the pressure ≥ 0, the dz2 orbital becomes dominant for the energy range near the Fermi level. The adsorption of the PEI molecule contributes to the occurrence of the new states, which comes from the contribution of the pz states of the N atoms. We can see more details in Fig. 9, the charge clouds of the Ny nitrogen atoms clearly exhibit the shape of the pz orbitals. The charge donation and accumulation distribute on the PEI fragment and the surface of the MoS2, respectively. This result is in good agreement with the behavior of the Bader point charge.
Fig. 9 The charge density difference of the MoS2/PEI system. Occupied and unoccupied states are presented in red and green, respectively. |
The parabolic behavior of the bandgap, as shown in Fig. 7, is attributed to the different responses of the dx2−y2 and dz2 orbitals of the MoS2 monolayer toward the pressure, see Fig. 8 (upper panel). At the negative pressures, the dx2−y2 orbital plays the main role, while at the higher pressures, the dz2 orbital gradually becomes dominant at the VBM and CBM. The contribution from these orbitals is balanced at the VBM and CBM; and hence, the bandgap is broadest for the pressure of −10.69 GPa. When the dz2 orbital approaches the Fermi level, the bandgap becomes narrower, as the tensile pressure increases. Furthermore, the study of HOMO and LUMO can support the understanding of the interaction between the MoS2 substrate and the PEI adsorbate; and hence, the parabolic behavior of the MoS2/PEI bandgap. We presented the HOMO and LUMO states of the isolated MoS2 and isolated PEI in Fig. 8 (lower panel). The values of the MoS2 LUMO and the PEI HOMO were also listed in Table 6. We found that the HOMO–LUMO gap varies with the parabolic behavior as that of the bandgap of the MoS2/PEI system (Fig. 7), which increases first and then decreases with the increase of the pressure from the negative to the positive value.
a (Å) | 2.84 | 3.00 | 3.12 | 3.16 | 3.18 | 3.32 | 3.48 | 3.64 | |
Isolated PEI | HOMO (eV) | −0.51 | −0.53 | −0.49 | −0.53 | −0.56 | −0.51 | 0.00 | 0.00 |
Isolated MoS2 | LUMO (eV) | 0.60 | 1.11 | 1.32 | 1.27 | 1.22 | 0.40 | 0.00 | 0.00 |
LUMO–HOMO (eV) | 1.11 | 1.64 | 1.81 | 1.80 | 1.78 | 0.91 | 0.00 | 0.00 |
P (GPa) | MoS2 | P (GPa) | MoS2/PEI | ||||||
---|---|---|---|---|---|---|---|---|---|
Eg (eV) | Ep (eV) | λ (nm) | ε2(ω) | Eg (eV) | Ep (eV) | λ (nm) | ε2(ω) | ||
−56.18 | 1.05 | 4.66 | 266 | 11.06 | −50.48 | 0.53 | 4.61 | 269 | 9.98 |
−32.44 | 1.55 | 2.60 | 477 | 11.58 | −38.51 | 0.81 | 2.76 | 449 | 11.10 |
0 | 1.68 | 2.73 | 453 | 14.88 | 0 | 0.55 | 2.72 | 456 | 14.91 |
10.13 | 0.81 | 2.39 | 521 | 13.36 | 10.25 | 0.01 | 2.35 | 527 | 13.38 |
24.95 | 0 | 2.00 | 621 | 13.18 | 21.79 | 0 | 2.01 | 618 | 12.86 |
41.14 | 0 | 1.42 | 873 | 12.06 | 31.62 | 0 | 1.41 | 877 | 11.12 |
As shown in Fig. 10a, at P = 0 GPa, the photoluminescence spectrum of the monolayer MoS2 exhibits the main peak at 453 nm within the range of visible light (400–800 nm). The result indicates that the MoS2 can absorb and emit the blue light, which is in agreement with the previous report.22,32 At non-zero pressures, the main PL peak shifts to the longer wavelength and simultaneously suppresses the peak height compared to that at zero pressure. Besides, for the positive pressure, the main peak tends to separate into two peaks. Furthermore, the intensity of the maximum PL is lower for the negative pressure than the positive pressure. We also found new peaks in the photoluminescence of the isolated MoS2 occurring in the ultraviolet region upon the compression and in the infrared region upon the tensile strain. The photoluminescence spectrum of the MoS2/PEI system in Fig. 10b remains almost unchanged compared to that of the isolated MoS2. This result implies that the benefit of the PEI coating is not only to protect the MoS2 from the adsorption of the unexpected gases but also remain the optical properties of the MoS2, which can be used for various optoelectronic applications.
This journal is © The Royal Society of Chemistry 2020 |