Design of Biosensors Based Transition-Metal Dichalcogenide for DNA-base Detection: A First-Principles Density Functional Theory Study

Document Type: Research Paper

Authors

Nanotechnology Research Institute, Faculty of Chemical Engineering, Babol University of Technology, Babol, Iran

Abstract

The main function purpose of nanobiosensors is to sense a biologically specific material and the kind of sensing platform and doping engineering has been an emerging topic and plays an important role in monolayer molybdenum disulfide (mMoS2). In this paper, we theoretically reveal the electronic structures of mMoS2 doped by 3d transition metals. Furthermore, adsorption of nucleic acid [Adenine (A), Cytosine (C), Guanine (G), Thymine (T) and Uracil (U)] on monolayers of modified transition-metal dichalcogenides (mTMDs) such as molybdenum disulfide (MoS2) and tungsten disulfide (WS2) as a sensing platform compared using first-principles density functional theory (DFT). We have found that all nucleobases were physisorbed on TMDs due to van der Waals (vdW) interaction and demonstrated that MoS2 is better than other TMDs and may also be used to detect nucleic acid sequence for medical science. Interestingly, the adsorption energies, band structures, electronic and magnetic properties of the transition metal (TM) atom adsorbed mMoS2 efficiently modified by absorbing different TM atoms which related to their number of d electrons. The order of binding energy of the nucleobases with MoS2 and WS2 are G > A > T > C > U using DFT D3 method.

Keywords


INTRODUCTION

The role of nucleic acid is vital in life science and its sequence detection is very important in the field of diagnosis of disease, forensic sciences, and genomics systems. Electrochemical biosensors have many advantages, including simplicity, high selectivity, and sensitivity and low-cost in nucleic acid sequence detection [1]. Nucleic acid molecules sequences are identified based on calorimetric and fluorescent assays [2], using nanomaterials such as carbon nanotube [3], quantum dots [4] and graphene [5]. The unusual properties of layered graphene and its successful applications in electronics, sensors, and energy devices have inspired and renewed interest in other monolayer materials [6].

To this end, alternative 2D materials [7] were developed in electrochemical sensors based on exceptional physical and electrochemical properties. Because of the poor ON/OFF current ratio in graphene [8], a semiconducting analog of graphene, particularly, MoS2 has attracted huge attention in the last few years. TMDs constitute a class of 2D materials with intrinsic band gaps of 1-2 eV used for various applications and increasingly becoming a popular material for biosensing applications. A significant number of publications have emerged about incorporation of TMDs into biosensors in recent years [9-12].

As already discussed in detail, high surface-to-volume ratio and layered structure of this materials can accommodate a large capacity of chemical biospecies. TMD materials have been widely used as effective assay methods for environmental monitoring, food safety, and clinical diagnosis in the past few decades due to their simplicity, low-cost, high sensitivity, and selectivity [13, 14].

MoS2 is one of the well-known metal dichalcogenides used in nanoelectronics, catalysis process, optoelectronics, and energy harvesting [15-18]. MoS2 is a semiconducting material with a hexagonal arrangement of S atoms on the surface and Mo atoms in the middle surfaces of S atoms. Mo and S atoms are covalently bonded with a typical distance of about 2.40 Å and the S-Mo-S angle of 80.6º.26 [19, 20]. Also, WS2 nanosheets are applied in solid lubricants, catalysis, tips in the scanning microscopy, and nucleic acid detection as nano-quenchers. Electronic and mechanical properties of WS2 have been studied theoretically by several groups [21-23].

TMDs (MoS2 and WS2) are used in electronics, catalysis and vapor molecule detection as a sensing platform [24-26]. With distinctive performances, mMoS2 has been a potential candidate as the complement or substitute to graphene [27]. Many methods were developed to explore the applications of mMoS2 for nanoscale electronics and spintronic devices and modulate the magnetic properties of mMoS2, such as defect, chemical doping, elastic strain, alloy, and chemical adsorption [28-32].

Our results present and discuss the relationships between the sensing performances and the special nanostructures within the MoS2 nanocomposites. It is well-known that adsorption of TM atoms is a promising and effective method for modulating the electronic and magnetic properties of materials which demonstrated a great potential in many materials [33]. Based on our results, variation in adsorption energy obtained from different TM atoms adsorbed on mMoS2 can effectively modulate the band structures and the magnetic properties of the mMoS2 due to their number of d electron.

Furthermore, understanding biomolecular interactions between the organic molecules and surfaces is important for designing better sensors with a higher selectivity [34, 35]. Hence, many theoretical studies have been carried out to find out nucleic acid interactions with 2D surfaces of MoS2 for sensing the biomolecules. In this work, we investigated the adsorption energy, stable geometry, magnetic and electronic properties of TM atoms adsorbed on mMoS2.

COMPUTATION METHODS

DFT calculations were performed via Xplorer - OpenMX code simulation package [36]. We used non-local van der Waals density functional (DFT-D3) at the GGA level of theory with the revPBE exchange-correlation functional for the calculations to describe the interactions between bases and substrates [37-39]. The core electrons represented by norm-conserving Morrison, Bylander and Kleinman pseudo-potentials [40]. The kinetic energy cutoff was set on 250 Ryd for integration of charge density in real space.

In order to cut the interaction between the supercell images, a 30 Å spacing was used in the direction perpendicular to the substrate. For all structures, geometry optimization was performed with convergence thresholds of 10-4 eV for energy and 0.01 eV/Å for Hellmann-Feynman force without any symmetry constraints [41]. Calculations were performed with 12×12×1 lateral supercells of MoS2 (72 atoms) and WS2 (108 atoms) [42]. Brillouin zone integrations were performed with a 3×3×1 Monkhorst-Pack k-point mesh to calculate total energies and densities of states. Binding energies of the bases with substrates (MoS2 or WS2) were calculated using the equation:

EBE = ESub + Bas - (ESub + EBas)

Where, ESub + Bas, ESub, and EBas are total energies of the complex, substrate, and bases [43]. The metal atoms were first placed in three high symmetry positions on the surface of mMoS2 (TMo, TS, and H sites) and then relaxed.

RESULTS AND DISCUSSION

Adsorption of nucleobases on to mMoS2 and mWS2 surfaces

In this section, we have placed bases on top of MoS2 and WS2 as the starting geometries for the calculations; several initial configurations were constructed. All of the starting configurations were relaxed and approached the surface of the MoS2/WS2 sheet with parallel and perpendicular orientation with respect to the molecular axis of MoS2/WS2. Furthermore, full structural optimization was carried out to evaluate the relative stability of the relaxed configurations. The least vertical distance and energy configuration, for each system, were chosen among the various possibilities. Force minimized structures of MoS2-base and WS2-base complexes are shown in Fig. 1 and Fig. 2, respectively.

The calculated vertical separations between mTMDs and nucleobases are in the ranges 2.90 - 3.36 Å and 2.96-3.30 Å for Mos2 and WS2, respectively. It is clear that the difference in the vertical distances between the TMDs is very small and bases do not change the surface geometrical parameters such as M-S bond distances and S-M-S angles significantly indicating the physical adsorption (through vdW interactions) of nucleobases on the mMoSsurface. In TMDs, heteroatoms like N and O can incline the molecules slightly on MoS2 and WS2. In addition to the heteroatoms, the hydrogen atoms in the tail (-NH2 and -CH3 group) of nucleobase interacts with MoS2 and WS2 surfaces. For instance, in MoS2-A complex, the H atom of CH3 group in adenine is in close contact with S atom of the surface with a typical distance of about 2.70 Å. In the case of MoS2-C, H of a methyl group and O of cytosine are relatively closer to substrates with distances of about 2.69 Å and 3.30 Å, respectively. For MoS2-G, H of NH2 and O of guanine interact stronger than the other atoms and the vertical distance for this close contact is about 3.19 Å. In the case of MoS2-T, H atoms of two methyl groups of thymine have a close contact about 3.09 Å with surface, and for MoS2-U, similar to other complexes, H of CH3 of uracil is in closer contact, about 3.12 Å, than other atoms [44-48].

Furthermore, from the energy values, it can be seen that for all the nucleobase molecules, the binding energies for the most stable systems lie within the range of -0.734 eV to -0.816 eV for MoS2 and -0.760 eV to -0.836 eV for WS2. The order of binding affinity of nucleobases with MoS2 and WS2 is G > A > T > C > U, which is similar to earlier studies. However, among all bases, G has the highest binding affinity and it may be recognized by substrates similar to silicon carbide or polyaniline. Our calculated geometries match with earlier studies. The structure and energetics of MoS2/WS2 and nucleobases are discussed in Table 1.

Based on the results, the highest difference between the binding energy values for individual nucleobase molecules interacting with MoS2 and WS2 was calculated to be 0.082 eV and 0.076 eV, respectively. Consequently, the MoS2 sheet does seem to be a proper choice to distinguish different the nucleobase molecules based on the obtained energy profile. In this regards, in the design of biosensors for detection of nucleic acid, we focus on mMoS2 as a sensing platform.

First-principles study of transition and alkali metal atoms adsorption on mMoS2

Researchers demonstrated that DNA-MoS2 strong binding can be achieved by introducing metal atoms, such as transition and alkali metal atoms on to the mMoS2 surface [49]. So, the binding characteristics and adsorption behavior of transition and alkali metal atoms on mMoS2 were studied using the DFT calculations. We considered the possible adsorption sites for metal atoms approaching the mMoSsurface. The selected sites are the top site of the S atom (Ts), the top site above the Mo atom (TMo) and the hollow site above the center of the hexagon or pentagon ring (H). The configurations of the metal atoms adsorption systems are shown in Fig. 3.

To find the most favorable adsorption configurations, the optimized mMoS2 surface was employed for metal atoms adsorption. After full structural relaxation of the selected configurations, the binding energy for these three sites is calculated. Table 2 shows the extranuclear electrons arrangement (EEA), adsorption energies (Eads), vertical distance and charge transfer (CT) analysis of the metal atoms (based on Hirschfeld) adsorbed on to mMoS2 at the most stable adsorption sites.

It can be seen that the TMo site is the most favorable surface adsorption site for the metal. Also, in all alkali cases, negligible charge transfer occurred between the involving systems while there is obvious electron transfer from transition metal atoms to the mMoS2. Based on Hirschfeld charge analysis, we found that the number of electrons transferred reduces gradually from Fe to Ni.

Furthermore, the results show that the interactions between the transition metal (TM) atoms and the nearest S atoms can be easily qualified as the covalent interactions, and such TM atoms have been chemisorbed on the surface. Consequently, the TM-mMoS2 sheet seems to be a proper choice to distinguish the base molecules based on the obtained adsorption energies. The variation in adsorption energies with different TM atoms may be related to their d electrons. The Ni with 8 d electrons has the largest adsorption energy, while the adsorption energies of Cu with full filled d orbital are small. The overall variation trend of the adsorption energies for different TM atoms with the number of d electron is consistent with the analysis of the bond length of TM-S (Mo), and hence, we focus on TM-mMoS2 as a sensing platform in biosensors.

Magnetic and electronic properties of TM-mMoS2

In addition, the magnetic properties of mMoS2 can be modulated by the embedded transition metals. The ground state of mMoS2 is nonmagnetic, while the transition metal-embedded mMoS2 show total spin magnetic moments of 2.21 µB, 1.01 µB, 0.00 µB, and 1.00 µB for Fe, Co, Ni, and Cu, respectively. To gain a deeper understanding of the interaction between the embedded metal atoms and the defect center of mMoS2, among all transition metal elements, we embedded Fe to Cu (except Ni (µB=0)) in the mMoS2 and calculated the density of state (DOS) curves of Fe-MoS2, Co-MoS2, and Cu - MoS2 systems, as displayed in Fig. 4.

Fig. 4a-c shows the DOS of Fe-MoS2, Co-MoS2 and Cu-MoS2 systems. A distinct feature is that asymmetric DOS peaks appear near the Fermi level, which is significantly different from the perfect mMoS2.

In a good agreement with experiments and earlier theories, pristine mMoS2 shows a band gap of 1.68 eV. Changes in the band gap occur when transition metals were adsorbed on MoS2 surfaces. According to the results, one may conclude from the analysis of geometries and energetics of the above-mentioned systems that Ni-MoS2 can act as a potential candidate to sense the nucleobases in DNA sequencing and other biomolecules.

With further increase in the number of the adsorbed Ni atoms on one side of mMoS2, the binding energies gradually decrease, as shown in Fig. 6.

Although the binding energies slightly depend on the detailed configurations of the adsorbed Ni atoms for some Ni adsorption concentrations, the decreasing trend of the binding energy still remains for the single Ni. Considering the different possibilities in adsorption configuration for different Ni concentrations, Ni atoms still favor adsorbing at the top sites of Mo atoms even if the Ni concentration is high. Furthermore, with the increase in the number of the adsorbed Ni atoms on the two sides of mMoS2, the binding energies decrease gradually. In addition, with increasing the Ni adsorption concentration, the difference between different adsorption configurations could be decreased due to the repulsive interactions between Ni ions.

CONCLUSION

Despite their extensive applications in energy storage and optoelectronic devices, the use of mMoS2 nanosheets for biosensing is still largely unexplored. In this study, adsorption characteristics of nucleic acid on TMDs (MoS2 and WS2) as a sensing platform were investigated using first-principles DFT-D3. We illustrated the key properties of 2D MoS2 offering unique opportunities for biosensing. Results revealed that nucleobases are physisorbed on MoS2 and WS2 with the order of interactions being G > A > T > C > U. In general, some new features occur in adsorption of transition metal (TM) atoms (from Fe to Cu) onto TMDs. The results show that the most stable adsorption site is TMo . However, almost all the TM atoms are chemically adsorbed on mMoS2. As these originate from the transition metal atoms at different energies, a selective detection is possible in the Ni-MoS2 cases. It is expected that many of new biosensing concepts are discovered and developed in the next decade to offer more opportunities for research activities in relevant areas and for industries via the rapid uptake of this material into biosensing systems [50-53].

ACKNOWLEDGEMENTS

Financial support of this Project by the Nanotechnology Research Institute of ChemicalEngineering in the Babol Noshirvani University of Technology is gratefully acknowledged.Authors acknowledge Professor M. Jahanshahi (Head of Nanotechnology Research Institute,Babol, Iran) for his help in the project.

CONFLICT OF INTEREST

The authors declared to no conflict of interest.

 

[1] Gopel W, Hesse, J., and Zemei, J.N. Sensors, A comprehensive Survey. Chemical and Biochemical Sensors. 2008; 2.

[2] Pumera M, Loo AH. Layered transition-metal dichalcogenides (MoS2 and WS2) for sensing and biosensing. TrAC Trends in Analytical Chemistry. 2014;61:49-53.

[3] Turner APF, Karube I, Wilson GS, Worsfold PJ. Biosensors: fundamentals and applications. Analytica Chimica Acta. 1987;201:363-4.

[4] Shruthi G, Amitha C, Mathew BB. Biosensors: a modern day achievement. J Instrument Technol. 2014; 2: 26-39.

[5] Ronkainen N, Okon S. Nanomaterial-Based Electrochemical Immunosensors for Clinically Significant Biomarkers. Materials. 2014;7(6):4669-709.

[6] Rao CNR, Gopalakrishnan K, Maitra U. Comparative Study of Potential Applications of Graphene, MoS2, and Other Two-Dimensional Materials in Energy Devices, Sensors, and Related Areas. ACS Applied Materials & Interfaces. 2015;7(15):7809-32.

[7] Mao K, Wu Z, Chen Y, Zhou X, Shen A, Hu J. A novel biosensor based on single-layer MoS2 nanosheets for detection of Ag+. Talanta. 2015;132:658-63.

[8] Su S, Chao J, Pan D, Wang L, Fan C. Electrochemical Sensors Using Two-Dimensional Layered Nanomaterials. Electroanalysis. 2015;27(5):1062-72.

[9] Shavanova K, Bakakina Y, Burkova I, Shtepliuk I, Viter R, Ubelis A, et al. Application of 2D Non-Graphene Materials and 2D Oxide Nanostructures for Biosensing Technology. Sensors. 2016;16(2):223.

[10] Vovusha H, Sanyal B. Adsorption of nucleobases on 2D transition-metal dichalcogenides and graphene sheet: a first principles density functional theory study. RSC Advances. 2015;5(83):67427-34.

[11] Li M-Y, Chen C-H, Shi Y, Li L-J. Heterostructures based on two-dimensional layered materials and their potential applications. Materials Today. 2016;19(6):322-35.

[12] Choi W, Choudhary N, Han GH, Park J, Akinwande D, Lee YH. Recent development of two-dimensional transition metal dichalcogenides and their applications. Materials Today. 2017;20(3):116-30.

[13] Han SA, Bhatia R, Kim S-W. Synthesis, properties and potential applications of two-dimensional transition metal dichalcogenides. Nano Convergence. 2015;2(1).

[14] Zhang Y, Zheng B, Zhu C, Zhang X, Tan C, Li H, et al. Single-Layer Transition Metal Dichalcogenide Nanosheet-Based Nanosensors for Rapid, Sensitive, and Multiplexed Detection of DNA. Advanced Materials. 2014;27(5):935-9.

[15] Junager N, Kongsted J, Astakhova K. Revealing Nucleic Acid Mutations Using Förster Resonance Energy Transfer-Based Probes. Sensors. 2016;16(8):1173.

[16] Tanaka S-i. Adsorption of DNA molecule and DNA Patterning on Si substrate. AIP Conference Proceedings: AIP; 2004.

[17] Wen Y-N, Xia M-G, Zhang S-L. Size effect on the magnetic and electronic properties of the monolayer lateral hetero-junction WS2-MoS2 nanoribbon. Applied Surface Science. 2016;371:376-82.

[18] Ketabi S, Hashemianzadeh SM, MoghimiWaskasi M. Study of DNA base-Li doped SiC nanotubes in aqueous solutions: a computer simulation study. Journal of Molecular Modeling. 2013;19(4):1605-15.

[19] Goswami N, Giri A, Pal SK. MoS2 Nanocrystals Confined in a DNA Matrix Exhibiting Energy Transfer. Langmuir. 2013;29(36):11471-8.

[20] Cho Y, Min SK, Yun J, Kim WY, Tkatchenko A, Kim KS. Noncovalent Interactions of DNA Bases with Naphthalene and Graphene. Journal of Chemical Theory and Computation. 2013;9(4):2090-6.

[21] Farimani AB, Min K, Aluru NR. DNA Base Detection Using a Single-Layer MoS2. ACS Nano. 2014;8(8):7914-22.

[22] Dolatabadi JEN, Mashinchian O, Ayoubi B, Jamali AA, Mobed A, Losic D, et al. Optical and electrochemical DNA nanobiosensors. TrAC Trends in Analytical Chemistry. 2011;30(3):459-72.

[23] Thomas S, Rajan AC, Rezapour MR, Kim KS. In Search of a Two-Dimensional Material for DNA Sequencing. The Journal of Physical Chemistry C. 2014;118(20):10855-8.

[24] Alinezhad H, Ganji MD, Soleymani E, Tajbakhsh M. A comprehensive theoretical investigation about the bio-functionalization capability of single walled CNT, BNNT and SiCNT using DNA/RNA nucleobases. Applied Surface Science. 2017;422:56-72.

[25] Wang Y, Wang B, Huang R, Gao B, Kong F, Zhang Q. First-principles study of transition-metal atoms adsorption on MoS2 monolayer. Physica E: Low-dimensional Systems and Nanostructures. 2014;63:276-82.

[26] Yue Q, Chang S, Qin S, Li J. Functionalization of monolayer MoS2 by substitutional doping: A first-principles study. Physics Letters A. 2013;377(19-20):1362-7.

[27] Ma D, Ju W, Li T, Zhang X, He C, Ma B, et al. The adsorption of CO and NO on the MoS2 monolayer doped with Au, Pt, Pd, or Ni: A first-principles study. Applied Surface Science. 2016;383:98-105.

[28] Zhu J, Zhang H, Tong Y, Zhao L, Zhang Y, Qiu Y, et al. First-principles investigations of metal (V, Nb, Ta)-doped monolayer MoS2: Structural stability, electronic properties and adsorption of gas molecules. Applied Surface Science. 2017;419:522-30.

[29] Ma D, Ju W, Li T, Yang G, He C, Ma B, et al. Formaldehyde molecule adsorption on the doped monolayer MoS 2 : A first-principles study. Applied Surface Science. 2016;371:180-8.

[30] Petrova NV, Yakovkin IN, Zeze DA. Metallization and stiffness of the Li-intercalated MoS2 bilayer. Applied Surface Science. 2015;353:333-7.

[31] Feng J, Liu K, Bulushev RD, Khlybov S, Dumcenco D, Kis A, et al. Identification of single nucleotides in MoS2 nanopores. Nature Nanotechnology. 2015;10(12):1070-6.

[32] Anu Prathap MU, Srivastava R, Satpati B. Simultaneous detection of guanine, adenine, thymine, and cytosine at polyaniline/MnO2 modified electrode. Electrochimica Acta. 2013;114:285-95.

[33] Fan D-H, Niu D-J, Huang K-J. Simultaneous determination of adenine and guanine in DNA based on carboxylic acid functionalized graphene. Sensors and Actuators B: Chemical. 2010.

[34] Larijani HT, Jahanshahi M, Ganji MD, Kiani MH. Computational studies on the interactions of glycine amino acid with graphene, h-BN and h-SiC monolayers. Physical Chemistry Chemical Physics. 2017;19(3):1896-908.

[35] Shokuhi Rad A, Zareyee D, Peyravi M, Jahanshahi M. Surface study of gallium- and aluminum- doped graphenes upon adsorption of cytosine: DFT calculations. Applied Surface Science. 2016;390:444-51.

[36] H. K. T. Ozaki JY, M.J. Han, N. Kobayashi, M. Ohfuti, F. Ishii, T. Ohwaki, H.Weng.

[37] Perdew JP, Burke K, Ernzerhof M. Generalized Gradient Approximation Made Simple. Physical Review Letters. 1996;77(18):3865-8.

[38] Zhang Y, Yang W. Comment on “Generalized Gradient Approximation Made Simple”. Physical Review Letters. 1998;80(4):890-.

[39] Grimme S, Antony J, Ehrlich S, Krieg H. A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. The Journal of Chemical Physics. 2010;132(15):154104.

[40] Morrison I, Bylander DM, Kleinman L. Nonlocal Hermitian norm-conserving Vanderbilt pseudopotential. Physical Review B. 1993;47(11):6728-31.

[41] Banerjee A, Adams N, Simons J, Shepard R. Search for stationary points on surfaces. The Journal of Physical Chemistry. 1985;89(1):52-7.

[42] Monkhorst HJ, Pack JD. Special points for Brillouin-zone integrations. Physical Review B. 1976;13(12):5188-92.

[43] Boys SF, Bernardi F. The calculation of small molecular interactions by the differences of separate total energies. Some procedures with reduced errors. Molecular Physics. 1970;19(4):553-66.

[44] Hirshfeld FL. Bonded-atom fragments for describing molecular charge densities. Theoretica Chimica Acta. 1977;44(2):129-38.

[45] Ganji MD, Mirnejad A, Najafi A. Theoretical investigation of methane adsorption onto boron nitride and carbon nanotubes. Science and Technology of Advanced Materials. 2010;11(4):045001.

[46] Ganji MD, Rezvani M. Boron nitride nanotube based nanosensor for acetone adsorption: a DFT simulation. Journal of Molecular Modeling. 2012;19(3):1259-65.

[47] Ahmadian N, Ganji MD, Laffafchy M. Theoretical investigation of nerve agent DMMP adsorption onto Stone–Wales defected single-walled carbon nanotube. Materials Chemistry and Physics. 2012;135(2-3):569-74.

[48] Tavassoli Larijani H, Darvish Ganji M, Jahanshahi M. Trends of amino acid adsorption onto graphene and graphene oxide surfaces: a dispersion corrected DFT study. RSC Advances. 2015;5(113):92843-57.

[49] Soltani A, Ahmadian N, Amirazami A, Masoodi A, Lemeski ET, Moradi AV. Theoretical investigation of OCN− adsorption onto boron nitride nanotubes. Applied Surface Science. 2012;261:262-7.

[50] Watson GW, Wells RPK, Willock DJ, Hutchings GJ. π adsorption of ethene on to the {111} surface of copper. Surface Science. 2000;459(1-2):93-103.

[51] Rankin RB, Sholl DS. Assessment of heterochiral and homochiral glycine adlayers on Cu(110) using density functional theory. Surface Science. 2004;548(1-3):301-8.

[52] Yang M-L, Zhu Y-A, Zhou X-G, Sui Z-J, Chen D. First-Principles Calculations of Propane Dehydrogenation over PtSn Catalysts. ACS Catalysis. 2012;2(6):1247-58.

[53] Sadeghi M, Jahanshahi M, Ghorbanzadeh M, Najafpour G. Adsorption of DNA/RNA nucleobases onto single-layer MoS2 and Li-Doped MoS2: A dispersion-corrected DFT study. Applied Surface Science. 2018;434:176-87.