Pd-Induced Electronic Activation and Strain-Tunable Adsorption-Coupled Electronic Modulation in Janus ZrSSe Monolayers

Abstract

Pd-decorated Janus ZrSSe monolayers provide a promising platform for adsorption-coupled electronic modulation in two-dimensional materials. Using first-principles density functional theory, we systematically investigate the structural stability, electronic properties, and adsorbate-induced electronic response of Pd-modified Janus ZrSSe. The results show that Pd is most stably anchored at the hollow site on the S-terminated surface, with a formation energy of $-1.45$ eV, while substitutional incorporation remains energetically unfavorable even after HSE06 refinement. Compared with pristine ZrSSe, Pd decoration markedly strengthens the interaction with adsorbates, leading to strong chemisorption for CO ($-1.026$ eV) and C₂H₂ ($-0.748$ eV), whereas H₂ remains comparatively weakly bound ($-0.258$ eV). Electronic-structure analysis reveals that CO induces the most pronounced perturbation because of strong orbital hybridization between Pd 4d states and C/O 2p states, resulting in the largest band-edge modulation among the three adsorbates. More importantly, biaxial strain provides an effective external degree of freedom for continuously tuning the electronic structure: tensile strain widens the band gap, whereas compressive strain systematically narrows it and ultimately drives a semiconductor-to-metal transition at sufficiently large compression. These findings establish Pd-decorated Janus ZrSSe as a strain-tunable electronic material in which adsorption, orbital hybridization, and band-edge evolution are intimately coupled, offering fundamental insights into controllable electronic modulation in polar two-dimensional systems.

1. Introduction

Adsorption-induced perturbations at solid surfaces are a central topic in surface and condensed-matter physics because they can reshape local bonding, redistribute charge, and generate interfacial dipoles, thereby modifying the electronic structure and transport characteristics of low-dimensional materials. Two-dimensional (2D) systems are particularly attractive in this context because their atomic-scale thickness maximizes surface exposure and makes their carrier density and band-edge states highly sensitive to external perturbations. In contrast to graphene, semiconducting 2D transition metal dichalcogenides (TMDs) provide intrinsic band gaps and chemically tunable electronic structures, making them versatile platforms for studying how adsorption affects local orbital hybridization and near-Fermi-level states [1,2,3].

Janus TMD monolayers offer an especially appealing symmetry-broken platform for such studies. By replacing one of the two equivalent chalcogen layers in $M{X}_2$ with a different chalcogen species, an $MXY$ structure is formed that lacks out-of-plane mirror symmetry [1,4]. Experimental advances have demonstrated that such atomically thin polar crystals can be stabilized [5]. The built-in out-of-plane dipole in Janus monolayers gives rise to chemically inequivalent terminations, modified work functions, and enhanced electromechanical responses [6,7,8]. These characteristics make Janus systems highly attractive for studies of adsorption-coupled electronic phenomena, since the intrinsic polarity can steer charge transfer, adsorption energetics, and band-edge alignment in an asymmetric manner [9,10].

A complementary strategy for activating 2D electronic materials is transition-metal decoration, which introduces localized d states that can hybridize with adsorbate orbitals. This effect is often rationalized within the Hammer–Nørskov d-band framework, where the position and filling of the d band govern the strength of adsorbate–surface coupling [11,12,13]. Among noble metals, Pd is a prototypical activator because it combines strong chemical affinity with versatile catalytic behavior [14]. When introduced on 2D semiconductors, Pd can create localized active electronic centers that strongly perturb adsorption-induced charge redistribution and band-edge states. Representative experiments have shown that Pd decoration can markedly enhance adsorption-induced electronic response in MoS₂-based heterostructures [15]. In parallel, strain engineering provides a continuous external handle for modifying bond lengths, orbital overlap, and symmetry, and has become a general route for tuning band structures and phase stability in 2D materials [16,17,18,19,20]. In Janus crystals, strain can couple to the intrinsic dipole and piezoelectric polarization, offering an additional degree of freedom for regulating adsorption-coupled electronic modulation.

Within this context, Janus ZrSSe has emerged as a stable group-IV TMD-derived monolayer with intrinsic polarity and electronically tunable behavior. Previous first-principles studies have shown that ZrSSe is mechanically and dynamically stable and may exhibit enhanced thermoelectric performance compared with ZrS₂ [21]. Its electronic and optical properties are also sensitive to biaxial strain, including a possible strain-driven semiconductor–metal transition [8]. More recently, gas adsorption on pristine and metal-modified Janus ZrSSe has begun to be explored in the context of adsorption-induced electronic modulation, for example through NO adsorption and transition-metal-modified selectivity studies [22,23]. Nevertheless, a systematic understanding of how Pd decoration and biaxial strain jointly regulate adsorption energetics, charge redistribution, orbital hybridization, and band-edge evolution in Janus ZrSSe remains limited.The deliberate selection of H₂, CO, and C₂H₂ as target analytes in this work is strongly motivated by the critical practical demand for early safety warning and fault diagnosis in lithium-ion batteries. During the thermal runaway process of LIBs, these evolved gases serve as key characteristic indicators at different failure stages. Moreover, owing to their high flammability, toxicity, and explosion risk, they can readily trigger secondary catastrophic fires. Driven by this important practical requirement, the present work systematically investigates the related adsorption and electronic-response behaviors [24,25].

In this work, we employ first-principles density functional theory to investigate the adsorption and electronic response of CO, H₂, and C₂H₂ on pristine and Pd-decorated Janus ZrSSe monolayers. These molecules serve as representative adsorbates for probing how chemically distinct species modulate the local electronic structure of a Pd-activated Janus surface [26,27,28]. We first determine the energetically preferred Pd incorporation motifs and clarify the stability difference between substitutional and adatom-decorated configurations. We then analyze how Pd-induced local d states reshape adsorption behavior through charge redistribution, differential charge density, and orbital hybridization. Finally, by applying biaxial strain, we elucidate how mechanical control tunes the band gap and can even induce metallization of the host material. This work establishes Pd-decorated Janus ZrSSe as a strain-tunable electronic material in which adsorption, local electronic activation, and band-edge modulation are intimately coupled.

2. Materials and Methods

2.1. First-Principles Framework

All density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP) [29,30]. The electron-ion interactions were described by the projector augmented wave (PAW) method [31,32], and the valence electron wavefunctions were expanded in a plane-wave basis with a cutoff energy of 500 eV. The generalized gradient approximation (GGA) parameterized by Perdew, Burke, and Ernzerhof (PBE) was employed for the exchange-correlation functional [33]. To strictly account for the long-range van der Waals interactions involved in gas adsorption, the DFT-D3 dispersion correction scheme was applied in all calculations [34]. The Brillouin zone integration was sampled using a $3\times 3\times 1$ Monkhorst–Pack grid for geometric optimizations and a denser $6\times 6\times 1$ grid for static electronic structure analysis [35]. Convergence criteria were set to ${10}^{-5}$ eV for energy and 0.01 eV/Å for the Hellmann–Feynman forces. Furthermore, to more accurately describe the localized d electrons associated with Pd incorporation, the HSE06 hybrid functional was employed to refine the formation energies of all considered Pd configurations, including both substitutional and adatom-decorated structures [36]. The HSE06 results were used specifically for formation-energy evaluation, whereas the electronic properties, including band structures and band gaps, were calculated at the PBE level for consistency with geometry optimization and to maintain a uniform basis for comparison.

2.2. Electronic Response Descriptors and Adsorption Kinetics

The Pd–ZrSSe monolayer was modeled using a $3\times 3\times 1$ supercell with a vacuum layer of 20 Å along the z-direction to eliminate spurious periodic interactions. The structural stability was evaluated based on the binding energy, calculated as the energy difference between the Pd-decorated system and the sum of the pristine monolayer and isolated Pd atom.

To characterize the adsorption-induced electronic response of the system, the electrical conductivity ($\sigma$) was estimated from the semiconductor band gap ($E_{\mathrm{g}}$) using the Arrhenius-type relation [37]:

$$\sigma =\lambda ·exp\left(-{\displaystyle \frac{E_{\mathrm{g}}}{2k_{\mathrm{B}}T}}\right)$$

(1)

where $\lambda$ is a constant prefactor, $k_{\mathrm{B}}$ is the Boltzmann constant, and T is the absolute temperature. Accordingly, an adsorption-induced relative conductivity change (S) is defined as $S=|({\sigma }_{\mathrm{gas}}^{-1}-{\sigma }_{\mathrm{pure}}^{-1})|/{\sigma }_{\mathrm{pure}}^{-1}$, where ${\sigma }_{\mathrm{pure}}$ and ${\sigma }_{\mathrm{gas}}$ denote the conductivity of the monolayer before and after gas adsorption, respectively.

The kinetic reversibility of adsorption is reflected by the recovery time ($\tau$), which represents the characteristic time required for gas desorption. This parameter was estimated using the transition state theory expression [38,39]:

$$\tau =A^{-1}·exp\left({\displaystyle \frac{|E_{\mathrm{ad}}|}{k_{\mathrm{B}}T}}\right)$$

(2)

where $E_{\mathrm{ad}}$ is the adsorption energy of the gas molecule, and A is the attempt frequency, assumed to be ${10}^{12}$ s⁻¹ following standard literature values for 2D materials [40]. Biaxial strain ($\epsilon$) was simulated by scaling the in-plane lattice constants, defined as $\epsilon =(a-a_0)/a_0$, where a and $a_0$ are the strained and equilibrium lattice constants, respectively.

3. Results and Discussion

3.1. Pristine Janus ZrSSe: Structural and Electronic Baseline

We begin by establishing the reference geometries for the adsorbates and the pristine Janus substrate. The optimized gas-phase molecules (Figure 1) exhibit equilibrium bond lengths of 1.143 Å for C–O in CO, 0.750 Å for H–H in H₂, and 1.207 Å (C≡C)/1.070 Å (C–H) in C₂H₂, consistent with their typical molecular characteristics.

The Janus ZrSSe monolayer (Figure 2) adopts a three-layer X–Zr–Y configuration with S and Se occupying the two inequivalent chalcogen planes. Such out-of-plane compositional asymmetry breaks mirror symmetry and is known to generate an intrinsic vertical polarity (built-in dipole) in Janus transition-metal dichalcogenides, which can amplify the electronic response to external perturbations such as adsorption and strain [41]. After structural relaxation, the lattice constant of ZrSSe is determined to be 3.70 Å, the Zr–S and Zr–Se bond lengths are 2.56 Å and 2.70 Å, respectively, in good agreement with prior first-principles reports on ZrSSe [8,21].

The pristine ZrSSe is predicted to be a narrow-gap semiconductor. At the PBE level used for geometry and baseline electronic trends, the band gap is 0.619 eV (Figure 3), close to previously reported values (∼0.60 eV) [8,21]. The projected density of states (PDOS) indicates that the band edges are dominated by Zr-4d states with appreciable hybridization with chalcogen p orbitals (S-2p and Se-3p), providing a clear orbital picture for how local perturbations (e.g., metal decoration or molecular adsorption) can modulate the near-gap electronic structure.

3.2. Weak Adsorption and Limited Electronic Perturbation on Pristine ZrSSe

In this work, adsorption behaviors are classified according to the adsorption energy criterion: $|E_{\mathrm{ad}}|<0.5$ eV is considered physisorption, whereas $|E_{\mathrm{ad}}|>0.5$ eV indicates chemisorption. Figure 4 summarizes the most stable adsorption configurations of CO, H₂, and C₂H₂ on pristine ZrSSe. The adsorption energies are modest, i.e., $E_{\mathrm{ad}}=-0.130$ eV (CO), $-0.068$ eV (H₂), and $-0.216$ eV (C₂H₂), with relatively large molecule–surface separations of 3.545, 3.103, and 3.441 Å, respectively (

Table 1. Adsorption properties of different gas molecules on ZrSSe.

System	Config.	${\mathit{E}}_{\mathbf{ad}}$ (eV)	${\mathit{d}}_{\mathbf{ad}}$ (Å)	${\mathit{Q}}_{\mathbf{CT}}$ (e)	Type
CO/ZrSSe	Figure 4a	−0.130	3.545	−0.070	Physical
H2/ZrSSe	Figure 4b	−0.068	3.103	−0.083	Physical
C2H2/ZrSSe	Figure 4c	−0.216	3.441	−0.089	Physical

). Consistently, Bader analysis yields only small net charge transfer ($|Q_{\mathrm{CT}}| \lesssim 0.09e$), indicating that the interaction is dominated by weak electrostatic and van der Waals contributions rather than appreciable chemical bonding. Subsequently, we calculated the total density of states and electronic band structures of the three gas adsorption systems, as shown in Figure 5, and compared them with the pristine ZrSSe system. The overall TDOS distributions of the three adsorption systems are similar to that of pristine ZrSSe, indicating that the introduction of gas molecules does not cause significant changes to the electronic density of states of the system. Further analysis of the band structures reveals that ZrSSe maintains its semiconducting nature after the adsorption of CO, H₂, and C₂H₂, with band gaps of 0.620 eV, 0.620 eV, and 0.622 eV, respectively, which are very close to the band gap value of pristine ZrSSe. These results suggest that pristine ZrSSe exhibits only limited electronic perturbation upon adsorption of these small molecules.

3.3. Energetic Preference and Stability of Pd Decoration: Emergence of a Localized Active Center

To introduce a well-defined adsorption center while preserving the two-dimensional character, we examined Pd incorporation in both substitutional and adatom-decorated forms on the inequivalent S and Se sides. The formation/binding energetics (

Table 2. Formation energies ($E_{\mathrm{form}}$) of Pd–ZrSSe monolayer at different sites.

Doping Type	Surface	Site	${\mathit{E}}_{\mathbf{form}}$ (PBE, eV)	${\mathit{E}}_{\mathbf{form}}$ (HSE06, eV)
Substitutional	S-surface	—	2.44	1.17
	Se-surface	—	1.95	0.61
Adsorption	S-surface	TS	1.74	0.42
		B	0.48	−0.70
		H	−0.26	−1.45
	Se-surface	TSe	1.96	0.62
		B	−0.27	−0.69
		H	−0.005	−1.16

) show that substitutional configurations are energetically unfavorable at both the PBE and HSE06 levels. In particular, the HSE06 formation energies for Pd substitution are 1.17 eV on the S surface and 0.61 eV on the Se surface, confirming that substitution is consistently disfavored relative to Pd adatom decoration. Although the formation energy of 0.61 eV is higher than the surface adsorption energy, it is not prohibitively high. Considering the typical chemical vapor deposition growth temperature of approximately 800 K, the thermal fluctuation energy suggests that a minority of substitutional defects could equilibrium-exist or freeze-in during the high-temperature growth phase. Nevertheless, under standard gas sensing operating conditions (298 K to 498 K), the surface adsorption configurations remain energetically far more favorable and thermodynamically stable, thus acting as the predominant modification mechanism governing the long-term sensing performance. Therefore, while minority substitutional configurations might exist as secondary defects from the growth stage, the Pd atom still prefers to undergo physical or chemical adsorption on the surface of the two-dimensional material rather than substituting its intrinsic atoms as the primary sensing active sites. In addition, adatom decoration can be stabilized, especially at high-symmetry hollow sites.

Among all considered adatom sites (Figure 6), Pd at the hollow site on the S side is the most stable structure. Importantly, when the energetics are refined using the hybrid functional for improved treatment of localized d electrons, the corresponding formation energy becomes $-1.45$ eV (Table 2), indicating a strong thermodynamic driving force toward Pd anchoring on the S-terminated face. The optimized geometry (Figure 7) reveals a short Pd–S bond length of 2.353 Å, consistent with the formation of a chemically bound Pd–S coordination environment.

Dynamical stability was evaluated via vibrational analysis. The absence of imaginary frequencies (48.95–375.88 cm⁻¹ for Pd–ZrSSe) confirms that the decorated configuration corresponds to a true local minimum on the potential-energy surface. Therefore, Pd decoration on the S-side hollow site provides a physically well-justified and stable platform to study adsorption-induced electronic modulation on a Janus substrate.

3.4. Adsorption on Pd–ZrSSe: Bonding Characteristics, Charge Redistribution, and Band-Edge Modulation

3.4.1. Adsorption Energetics and Geometrical Fingerprints

With Pd providing a localized d-electron center, the adsorption behavior changes qualitatively (Figure 8 and

Table 3. Adsorption properties of different gas molecules on Pd–ZrSSe.

System	Config.	${\mathit{E}}_{\mathbf{ad}}$ (eV)	${\mathit{d}}_{\mathbf{ad}}$ (Å)	${\mathit{Q}}_{\mathbf{CT}}$ (e)	Type
CO/Pd–ZrSSe	Figure 8a	−1.026	1.979	−0.174	Chemical
H2/Pd–ZrSSe	Figure 8b	−0.258	1.991	−0.227	Physical
C2H2/Pd–ZrSSe	Figure 8c	−0.748	2.312	−0.033	Chemical

). CO binds most strongly with $E_{\mathrm{ad}}=-1.026$ eV and forms a short Pd–C bond of 1.979 Å in an end-on configuration. C₂H₂ also exhibits strong binding ($E_{\mathrm{ad}}=-0.748$ eV), adopting a nearly parallel geometry with Pd–C distances down to 2.312 Å, indicating the formation of a Pd–C₂H₂ $\pi$-complex-like interaction. In contrast, H₂ remains weakly bound ($E_{\mathrm{ad}}=-0.258$ eV) and does not show clear chemical bond formation, despite a short geometric proximity to Pd (1.991 Å). These trends establish the adsorption-strength hierarchy of CO > C₂H₂≫ H₂.

3.4.2. Charge Transfer Versus Chemical Bonding: Differential Charge Density and Bader Analysis

To quantify electron redistribution, we analyzed both Bader charge transfer and charge-density difference (CDD). Here, a negative $Q_{\mathrm{CT}}$ value denotes electron transfer from the Pd–ZrSSe substrate to the adsorbate; accordingly, CO, H₂, and C₂H₂ gain $0.174$, $0.227$, and $0.033e$ (Table 3), respectively, indicating that all three molecules behave as net electron acceptors [42,43]. The CDD isosurfaces in Figure 9 provide crucial additional insight: for CO and C₂H₂, electron accumulation is localized in the Pd–C bonding region, consistent with genuine bond formation and substantial orbital overlap. For H₂, however, the redistribution is comparatively diffuse with minimal accumulation directly between H₂ and Pd, suggesting that its larger apparent $Q_{\mathrm{CT}}$ mainly reflects polarization/charge rearrangement in the vicinity of the Pd center rather than a strong covalent interaction. It is worth emphasizing that the numerical value of $Q_{\mathrm{CT}}$ alone (e.g., $-0.227e$ for H₂) is insufficient and potentially misleading for characterizing the adsorption type. It is fundamentally crucial to emphasize that the numerical value of $Q_{\mathrm{CT}}$ from Bader population analysis should not be used alone for characterizing or classifying the adsorption type (physisorption vs. chemisorption). This limitation is particularly prominent at heavy-metal/light-gas interfaces, where the electronic density is highly diffuse and the mathematical zero-flux surfaces can partition the spatial grid with inherent ambiguity. Therefore, rather than relying on total atomic charges, our diagnostic approach strictly leads with the 3D charge density difference and thermodynamic analysis to determine the interaction nature. As visualized in the real-space CDD plots, the larger apparent $Q_{\mathrm{CT}}$ of H₂ compared to CO does not denote a localized chemical bond. Instead, it reflects an integrated charge transfer/rearrangement within specific bonding and diffuse molecular regions under intense substrate polarization. Due to the absence of core electrons in Hydrogen, the shared valence electrons are localized by the zero-flux grid predominantly into the boundary of the H atoms, yielding a higher apparent atomic sum. Consequently, guided by the CDD and orbital hybridization traits, CO is verified as localized chemisorption, whereas H₂ remains a physisorption state dominated by regional polarization. To accurately distinguish between physisorption and chemisorption, this quantitative charge analysis must be coupled with the visualization of spatial charge redistribution, underscoring the necessity of a multifaceted theoretical approach.

Vibrational analyses further support the stability of the adsorbed complexes, with no imaginary frequencies found for CO/Pd–ZrSSe (43.53–2062.26 cm⁻¹), H₂/Pd–ZrSSe (50.43–3713.41 cm⁻¹), and C₂H₂/Pd–ZrSSe (5.44–3392.13 cm⁻¹).

3.4.3. Orbital Hybridization and Band-Gap Response

The adsorption-induced electronic response is clarified by TDOS/PDOS and band structures (Figure 10). For CO adsorption, the TDOS exhibits an overall rightward shift relative to the bare Pd–ZrSSe reference, and the PDOS reveals pronounced hybridization between Pd-4d and C-2p/O-2p states in the range of approximately $-4$ to $-2$ eV. This hybridization is consistent with strong metal–molecule coupling at the Pd site and rationalizes the largest adsorption energy among the three molecules. For C₂H₂, moderate TDOS shifting and Pd-4d/C-2p hybridization (around $-5.5$ to $-4$ eV) indicate a substantial, but weaker, coupling than in CO. In contrast, H₂ adsorption produces only minor TDOS changes and negligible PDOS overlap between Pd-4d and H-1s, consistent with its physisorption-like interaction.

At the band-structure level, adsorption modulates the gap size: the calculated gaps for CO/Pd–ZrSSe, H₂/Pd–ZrSSe, and C₂H₂/Pd–ZrSSe are 0.681, 0.636, and 0.634 eV, respectively (Figure 10). Therefore, CO induces the most pronounced band-edge renormalization, while H₂ and C₂H₂ produce comparatively smaller changes. From a condensed-matter perspective, such adsorption-driven band-gap modulation provides a direct microscopic mechanism for the electronic response of the system: perturbing the near-gap states changes the thermal activation barrier for carriers and thus the conductivity.

For completeness, we estimated the conductivity change using the Arrhenius-type relation and obtained indicative relative conductivity variations of 528% (CO), 163% (H₂), and 153% (C₂H₂) [38]. Within this simplified model, the pronounced conductivity variation induced by CO is a direct consequence of its strongest Pd-mediated hybridization and the associated largest band-structure perturbation.

3.4.4. Desorption Kinetics as an Energetic Consequence of Binding Strength

The kinetic reversibility of adsorption is closely linked to the binding strength between the adsorbate and the Pd–ZrSSe surface. Using the transition-state-theory estimate [38], the recovery time exhibits a strong temperature dependence (Figure 11). CO desorption is slow at 298 K ($1.91\times {10}^5$ s) but accelerates rapidly at elevated temperatures (9.77 s at 398 K and 0.325 s at 448 K), reflecting its strong chemisorption. C₂H₂ shows a moderate room-temperature recovery time (4.43 s) and becomes very fast upon heating. By contrast, H₂ desorption is ultrafast over the entire considered temperature range, consistent with its weak interaction. These trends follow naturally from the adsorption energy hierarchy and reinforce the physical picture established above: Pd introduces a strong d-state adsorption center whose coupling is molecule dependent.

3.5. Biaxial Strain as a Control Knob: Charge Transfer Tuning and Gap Engineering

Janus monolayers are intrinsically susceptible to strain due to their broken out-of-plane symmetry and internal polarity [41,44]. We therefore investigated biaxial strain as a generic and experimentally accessible parameter to tune the electronic response (Figure 12). The strain dependence of charge transfer and band gap is summarized in Figure 13.

For charge transfer (Figure 13a), H₂ maintains the largest $|Q_{\mathrm{CT}}|$ across the full strain range, and its interaction is strongly strain dependent: compressive strain enhances the electron-accepting character (from $-0.227e$ at 0% to ∼$-0.265e$ at $-4$%), whereas tensile strain reduces it (down to ∼$-0.147e$ at $+6$%). In comparison, CO and C₂H₂ exhibit smaller-amplitude, non-monotonic variations of $Q_{\mathrm{CT}}$, indicating a more intricate interplay between geometric deformation and Pd–molecule hybridization.

More strikingly, the band gap displays an almost linear dependence on biaxial strain for all adsorption systems (Figure 13b): compressive strain narrows the gap, while tensile strain widens it. Under sufficiently strong compression, the gap approaches closure, signaling a strain-driven semiconductor-to-metal transition. Specifically, as derived from our quantitative threshold analysis, the precise critical strain value for metallization in the Pd-decorated Janus ZrSSe system is approximately $-6\%$ under biaxial compression, where the band gap completely vanishes. Physically, this behavior can be understood as strain-modulated orbital overlap and level alignment (Zr-d, chalcogen-p, and Pd-d contributions) near the band edges; increasing overlap under compression promotes band broadening and gap closure, while tensile strain reduces overlap and opens the gap. Compared with the prior work on pristine ZrSSe monolayers [8], where strain simply alters the intrinsic atomic distances, our system introduces a novel strain-modulated interfacial orbital hybridization mechanism. Under compression, the spatial overlap between the localized d-orbitals of the Pd adatom and the neighboring chalcogen p-orbitals is strongly enhanced, heavily regulating the gas adsorption strength and charge transfer pathways. Consequently, biaxial strain serves as an effective tuning knob (with an ideal window within $\pm 3\%$) to simultaneously optimize the gas sensitivity and recovery time, effectively breaking the conventional trade-off in 2D gas sensors. Therefore, beyond molecule-specific adsorption effects, Pd–ZrSSe provides a clear example of a Janus 2D platform where the electronic ground state itself (semiconducting versus metallic) can be tuned continuously by mechanical deformation.

Overall, the strain results demonstrate that (i) charge redistribution at the Pd active site is mechanically tunable, and (ii) the band-gap response can be engineered over a wide range up to gap closure. These are general condensed-matter features of Pd–ZrSSe that are expected to be relevant to a broad range of contexts where controllable adsorption-induced electronic modulation is desired.

4. Conclusions

In summary, first-principles calculations demonstrate that Pd decoration provides an effective route for electronically activating the Janus ZrSSe monolayer while preserving its mechanically tunable character. The Pd adatom is most stably anchored at the hollow site on the S-terminated surface, forming a robust Pd–S coordination environment that serves as a localized active electronic center. In contrast to pristine ZrSSe, which exhibits only weak interaction with CO, H₂, and C₂H₂, Pd decoration markedly strengthens the adsorption of CO and C₂H₂, whereas H₂ remains comparatively weakly bound. This molecule-dependent behavior originates from the different degrees of hybridization between Pd 4d states and the molecular orbitals of the adsorbates.

The electronic response follows the same trend. CO induces the largest perturbation of the near-Fermi electronic structure and the most pronounced band-gap modulation because of its strongest Pd-mediated coupling, while H₂ and C₂H₂ lead to comparatively smaller changes. The corresponding recovery-time behavior is consistent with the adsorption-strength hierarchy, further confirming that the interaction between the Pd active center and the adsorbates is strongly molecule dependent. In addition, biaxial strain provides an efficient external control parameter for tuning both charge redistribution and band structure: tensile strain widens the band gap, whereas compressive strain narrows it and can ultimately trigger a semiconductor-to-metal transition under sufficiently large compression.

Overall, these results establish Pd-decorated Janus ZrSSe as a strain-tunable 2D electronic material in which adsorption, local electronic activation, orbital hybridization, and band-edge evolution are intimately coupled. This work provides a clear microscopic picture of controllable adsorption-induced electronic modulation in polar two-dimensional systems and offers a useful framework for the design of electronically responsive Janus materials.