MoSe₂ nanosheet as a lung cancer biosensor: A DFT study

Lung cancer (also called pulmonary carcinoma) is the third most common and the leading cause of cancer-related mortality in the world.¹ Early detection of lung carcinoma is of utmost importance, for it not only promotes the chance of treatment success but also prevents misdiagnosis, which, in turn, exposes the patient to unfitting or even harmful medication.² Some of the conventional diagnostic methods such as ultrasound, magnetic resonance imaging (MRI), positron emission tomography (PET), and computerized tomography (CT), in addition to having major drawbacks including complexity, high cost, low yield, and slowness, have presented low efficiency in the early detection of lung neoplasms as they are based on morphological and phenotypic properties of tumors.^3–5 Invasive methods such as bronchoscopy and needle biopsy, having their special difficulties, also suffer from similar insufficiencies in early lung cancer diagnosis.⁵ Due to the mentioned challenges, the development of cancer diagnosis through biomarker-level measurement has attracted much attention in recent years.^6–9 Biomarkers, as a sign of cancer disease, may exist in body fluids (such as blood, tears, saliva, urine, etc.) or even in the exhaled air of a cancer patient.^2,10–12 Although there are various techniques for determination and quantification of the biomarkers in gaseous or liquid samples, including gas chromatography–mass spectrometry (GC–MS),¹³ liquid chromatography–mass spectrometry (LC–MS),¹⁴ electronic nose,¹⁵ selected-ion flow-tube mass spectrometry (SIFT-MS),¹⁶ high field asymmetric waveform ion mobility spectrometry (FAIMS),¹⁷ fabrication of sensors for the detection of specific cancer biomarkers is of great importance, since the utilization of aforementioned methods is not possible for large-population screening due to their cost and complexity. There are plenty of biosensor types, including potentiometric, chemiresistive, photoluminescence, fluorescence, and electrochemical, which are designed and fabricated to detect biomaterials, including cancer biomarkers as a sign of cancer disease.^18–20 Some volatile organic compounds (VOCs) have been determined as cancer biomarkers whose volume in body fluids or exhaled air of cancerous patients differ significantly.^21–24 Cancer detection by determination of the VOC level, using biosensors, is completely non-invasive, cost-effective, and probably the simplest method in execution.

Two-dimensional (2D) nanomaterials, such as transition-metal dichalcogenides (TMDCs), transition-metal dioxides (TMDOs), hexagonal boron nitride (hBN), and graphitic C₃ N₄(g-C₃ N₄), are currently extensively investigated for applications in electronics, optoelectronics, energy storage, energy conversion, sensors, etc.^7,25,26 They have also found various biomedical applications including but not limited to bioimaging, tissue engineering, cancer therapy, drug delivery, cancer theranostics, and biosensing.^27–30 Considering biomedical applications, there are serious concerns about the application of 2D materials such as their biodegradability, biocompatibility, and toxicity, in addition to concerns about the stability, preparation complexity, water solubility, and scalability in some of them.^25,27,31 Fortunately, in the case of sensory application, we do not need to worry about some of the mentioned concerns such as toxicity and biocompatibility unless the fabricated sensor is planned to be implanted.

TMDCs have been tremendously investigated as sensing materials due to their appropriate properties like suitable bandgaps, high surface area, and high density of electronic states.^25,32,33 MoSe₂ is a member of the TMDC family, whose atoms bond with each other by covalent bonds and form layers of Se–Mo–Se, which stack on one another with van der Waals interaction to form a 2D structure with an interlayer spacing of ∼6.5 Å.^34,35 Due to some beneficial features such as high carrier mobility,³⁴ high surface electroactivity,³⁶ direct bandgap, and diversity of fabrication methods,³⁷ MoSe₂ has the potential for application as the sensing material for various targets in different methods, among which chemiresistive sensors, due to their low cost, fast response, and high sensitivity, are very interesting for real-world applications.^38–42 In addition, the known anticancer properties of selenium make it a candidate for theranostic applications if its sensory properties would be approved.⁴³ A chemiresistive sensor principally is a material whose electrical resistivity is related to the amount of a target chemical in its environment. The theoretical study of the adsorption of various molecules to specific sensing materials is useful for proposing possible sensing materials for specific targets as well as investigating the sensing mechanism of experimentally introduced sensors.⁴⁴ In particular, many DFT-based studies of electronic features of various materials have been conducted recently to evaluate their applicability in both the detection and treatment of cancer.^45–47 

In this work, we have studied the possibility of 2D-MoSe₂ application as a chemiresistive biosensor for lung cancer detection. Three of the most popular VOCs reported as the cancer biomarkers, including hexanal (C₆H₁₂O), nonanal (C₉H₁₈O), and p-cresol (C₇H₈O), have been selected to evaluate their adsorption influence on electronic properties of 2D-MoSe₂.^23,48 To the best of our knowledge, it is the first time that the sensitivity of MoSe₂ to these biomarkers is studied.

Theoretical calculations have been carried out using the Perdew–Burke–Ernzerhof (PBE) functional within the generalized gradient approximation, which is implemented in the Quantum ESPRESSO package.^49,50 Ultrasoft pseudopotentials with electronic orbitals [Kr] 5s2 4d4, [Ar] 4s2 4p4 3d10, [He] 2s2 2p4, [He] 2s2 2p2, and 1s1 in valence are used for Mo, Se, O, C, and H atoms, respectively. Marzari–Vanderbilt smearing with a value of 0.01 Ry is set to guarantee accuracy.⁵¹ The energy cutoff for electron charge wave function and electron density is set to 85 and 500 Ry, respectively. A Monkhorst–Pack k-point mesh of 6 × 6 × 1 was used for sampling the Brillouin zone in the geometry optimization of MoSe₂.⁵² The mesh grid increased to 12 × 12 × 1 for electronic structure calculation. The self-consistent field convergence limit was set to 10⁻⁸ Ry. The atomic positions are fully relaxed until the convergence of forces acting on ions reaches 0.001 Ry/au. The semiempirical Grimme's DFT-D3 method is used to consider long-range dispersion correlations.⁵³ The DFT-D3 method is used a lot in order to include the effect of non-local correlations. Although previous versions of the Grimme method (DFT-D and DFT-D2) are very simple, the C6 and C8 parameters in the D3 version depend on the coordination number of each atom, estimated using the distances to other nearby atoms. This method is fully flexible and system-dependent⁵⁴ and makes DFT-D3 more accurate than the earlier versions as its mean relative deviation accuracy is almost 4.7%.⁵⁵ To investigate the adsorption of hexanal (C₆H₁₂O) and p-cresol (C₇H₈O) on a MoSe₂ sheet, a 6 × 4 × 1 supercell of the MoSe₂ unit cell with a vacuum of 15 Å along the z-axis created, which have been proven to be large enough to prevent the interaction between adjacent molecules. A 7 × 4 × 1 supercell is also used for nonanal (C₉H₁₈O). A 3 × 6 × 1 k-point mesh grid is considered for the geometry optimization of combined systems. The Bader atoms-in-molecule (BAIM) technique was carried out to study the charge transfer between the adsorbed VOC molecules and the MoSe₂ monolayer.

The band structure and density of states (DOS) calculation results of the relaxed MoSe₂ monolayer are illustrated in Fig. 1. The obtained band structure and DOS patterns, as well as the calculated direct bandgap (E_g= 1.523 eV), are similar to the ones previously reported in the literature.⁵⁸ Moreover, it is figured out that the monolayer of MoSe₂ exhibits p-type semiconducting behavior as its band structure and DOS pattern show that the fermi level is closer to the valance band. This finding is in agreement with the results of previous theoretical and experimental studies.^38,42,59,60

To investigate the effect of the biomarkers' adsorption on the electronic properties of the MoSe₂ monolayer, bandgap alteration, and charge transfer were studied (see Table I). The bandgap of the monolayer lowered to 1.5208 eV due to the adsorption of Hexanal, while the adsorption of p-cresol and nonanal decreased the bandgap to 1.5201 eV. On the other hand, our calculations show −0.0388, −0.0007, and −0.0506 e charge transfer for the adsorption of hexanal, p-cresol, and nonanal, respectively. The value of charge transfer is obtained as the difference between total Bader charges on atoms of molecules before and after adsorption (Q = Q_before − Q_after). In each case, the negative sign indicates the charge transfer from the monolayer to the VOC molecule. It should be noticed that the decrease of E_g and electron donation to the adsorbed molecule both indicate that the MoSe₂ monolayer would be more conductive upon exposure to the biomarkers. This is because the MoSe₂ monolayer acts as a p-type semiconducting sensor as mentioned previously. It is well known that p-type semiconductors get more conductive when they lose negative charge.⁶³ 

A biosensor can be evaluated by two parameters. The first one is its success in the adsorption of the target molecule, and the second is the significance of the variations of its electrical parameters (e.g., resistance). As expressed earlier, the calculated adsorption energies show the appropriate behavior of the proposed sensor in the adsorption of the considered biomarkers. However, the evaluation of the proposed sensor's performance in terms of the significance of electrical parameters alteration needs more study.

The calculated sensing characteristics of the proposed 2D-MoSe₂ monolayer have been demonstrated in Fig. 3.

The adsorption and sensing performance of the MoSe₂ monolayer upon three lung cancer VOC biomarkers were theoretically investigated. The adsorption energy was calculated at −0.518, −0.577, and −0.699 eV for hexanal, p-cresol, and nonanal, respectively, which reveal the physisorption of biomarkers. The sensing mechanism of the monolayer was also investigated by studying the bandgap alteration and charge transfer at exposure to the biomarkers. Our finding, in agreement with the previous reports, confirms the p-type sensing behavior of the MoSe₂ monolayer. The sensing performance of the proposed sensor was also evaluated. To this end, the response of the proposed sensor was estimated by taking into account the effect of bandgap on the conductivity of the semiconductor. Moreover, the recovery time of the proposed sensor was estimated using the van't-Hoff–Arrhenius expression. The proposed sensor exhibits an acceptable response to the biomarkers and desirable recovery times in the desorption process. Therefore, based on our findings in this study, we propose that the MoSe₂ monolayer can be applied as a biosensor for the early detection of lung cancer, and we hope that in the future works, prominent biosensing properties will be achieved by making some modifications to this material.

The authors have no conflicts to disclose.

Ali Mosahebfard: Supervision (lead); Writing – original draft (lead); Writing – review & editing (equal). Mohammad Moaddeli: Software (lead); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.