Lithiation of Epitaxial Monolayer Borophene Authors: Sherif Kamal, Insung Seo, Iulia Cojocariu, Matteo Jugovac, Andrea Locatelli, Yoshihiro Gohda, Marko Kralj, Tevfik Onur Menteş, Marin Petrović Published: 2026 ABSTRACT Functionalization of borophene offers a viable route to modify its properties and create advanced boron-based 2D derivatives. For tuning the electronic structure of borophene, an aspect of both fundamental and applicative importance, alkali metals are particularly attractive because of their efficient charge donation. In light of this, we study morphological and electronic modifications of the borophene monolayer on Ir(111) after Li deposition in an ultra-high vacuum. By combining micro-probe low-energy electron microscopy and diffraction with X-ray spectroscopy and comprehensive density functional theory calculations, we are able to disentangle the complex interaction between Li, B, and Ir atoms. It is found that adsorbed Li distorts the crystal lattice of borophene, especially after sample annealing, and that Li intercalation is unfavorable. Charge redistribution within the system is governed by the high chemical reactivity and efficient electron donation of Li atoms, which bind strongly to B and Ir atoms and disrupt the preset energy balance of the epitaxial borophene layer. Our study showcases the advantages and challenges of alkali-metal decoration of borophene and provides guidelines for future borophene functionalization strategies. Introduction The ultimate surface-to-volume ratio of two-dimensional materials (2DMs) renders molecular and atomic adsorption a powerful tool for modifying their intrinsic properties. Adsorption, often accompanied by intercalation, provides versatile opportunities for nanostructuring and manipulation of the electronic charge within the 2DM itself. Particular attention has been devoted to the adsorption and intercalation of alkali metals because of their simple structure, efficient charge donation, and relevance in energy storage applications. Hence, alkali metal decoration has been extensively studied for a range of 2DMs, including graphene, hexagonal boron nitride, phosphorene, and molybdenum disulfide, where high electron doping, work function reduction, the emergence of new electronic phases, and modification of many-body excitations have been observed. − In contrast, more recently discovered 2DMs still lack comprehensive structural and electronic characterization regarding their interaction with alkali atoms. One such material is borophene (Bo), which has been synthesized in monolayer and bilayer forms on various transition metal substrates. Although post-growth adsorption of simple gases (H, O), transition metals (Au), and more complex organic molecules on Bo polymorphs has been realized, − a controlled decoration with efficient charge donors such as alkali atoms is lacking to date. Such studies are experimentally challenging due to the complex synthesis and high chemical reactivity of single-layer borophene. − On the contrary, bulk and powder boron-alkali systems are more easily handled and have been investigated by now, especially Li-based compounds, owing to their technological relevance in rechargeable batteries. − From a theoretical perspective, several studies have shown that Li and Cs adsorption on freestanding Bo sheets results in cluster-free, thermodynamically stable atomic structures which are characterized by notable charge transfer and work function decrease. , Moreover, low diffusion barriers and high specific capacity were reported for Bo sheets decorated with various alkali and alkaline-earth atoms, highlighting their potential as anode materials. − Going beyond the monolayer limit, interactions of alkali and alkaline-earth atoms with bilayer Bo have been predicted to increase its superconducting critical temperature. , In this work, we present a spectro-microscopic insight into Li deposition on the Bo monolayer on Ir(111), supported by comprehensive density functional theory (DFT) calculations. Our results provide a fundamental understanding of the interaction of Li atoms with an epitaxial monolayer of B atoms, with implications for other Bo polymorphs and alkali-metal adsorbates. The observed modifications of the electronic properties and crystal structure, which exhibit complex behavior, offer guidance for the design of future experiments and enable a critical reassessment of the suitability of Bo for battery-related applications. Results Li Deposition at Room Temperature Borophene on Ir(111) (Bo/Ir) was synthesized via a dissolution-segregation mechanism of B atoms originating from a borazine precursor, as reported earlier (see Methods for details). , The Bo/Ir samples were fabricated with submonolayer Bo coverage, as shown in a representative bright-field (BF) low-energy electron microscopy (LEEM) micrograph in Figure a, which enabled simultaneous examination of Li interaction with Bo-covered and Bo-free regions of the sample. Throughout the manuscript, those regions are referred to as Bo and Ir regions, respectively, although their morphology, crystallography, and chemical composition evolve during the experiments. In Figure a, Ir terraces filled with a single-layer, single-domain Bo span across the field of view and exhibit a sharp (6 × 2) pattern in low-energy electron diffraction (LEED) with very little background intensity, see Figure e. Such LEEM and LEED signatures are characteristic of the high-quality, monolayer χ6 Bo polymorph on Ir(111). ,, (a–c) BF-LEEM micrographs (U st = 12 V) of a pristine, Li-decorated and subsequently annealed Bo/Ir sample, respectively. In all images, two distinct regions (Bo and Ir) can be identified, separated by a transition (Tr) zone in the annealed sample. (d) LEEM-IV data extracted from Bo, Ir, and Tr regions of the samples shown in panels a–c. (e–g) LEED patterns (E = 40 eV) originating from the Bo regions of samples shown in panels a–c. Upon sample annealing, spots corresponding to the Li- 12/7 superstructure appear (two rotational domains, see text for definition). The reciprocal unit cells of Ir, Bo, and Li-related superstructures are marked in purple, red, and blue, respectively. (h) LEED-IV data obtained from the Bo’s diffraction spots marked by red circles in (e–g). Li deposition on Bo/Ir was performed at room temperature (RT) and was monitored in situ using LEEM and LEED. Upon deposition of Li up to a dose corresponding to a nominal coverage (θLi) of ∼2.8 monolayer (ML), no new morphological features were detected by LEEM. However, changes in the electron reflectivity of the Bo and Ir regions were found. An example is shown in Figure b for the θLi = 1.6 ML sample. The observed LEEM contrast inversion, along with the full spectroscopic modifications of the Bo and Ir regions, is discerned in LEEM intensity-voltage (LEEM-IV) data shown in Figure d. These results indicate that Li adsorption substantially modifies the local electronic and crystal structure of both regions, which directly govern electron reflectivity in LEEM. No new superstructures were observed in LEED during Li deposition on Bo regions, and at the same time, the background intensity has increased (see Figure f), which indicates the absence of a long-range periodic Li adlayer and the formation of a disordered structure on the surface. Disordered Li arrangement on Bo is also suggested by the LEED-IV data displayed in Figure h, where only a moderate modification of the Bo diffraction spot spectrum was detected after Li deposition. In parallel, the intensities of Bo diffraction spots in Figure f are reduced after Li deposition (see also Supporting Information Figure S1), causing the disappearance of some of the spots, most likely due to the Li-induced deterioration of the Bo lattice and attenuation by the Li adlayer. To assess the impact of Li deposition on the chemical and electronic properties of the Bo region, micro-spot X-ray photoelectron spectroscopy (μ-XPS) measurements were carried out on a freshly prepared Bo/Ir at variable θLi. As visible in the color plot in Figure a, B 1s photoemission intensity continuously shifts to higher binding energy as Li is deposited on the Bo region of the sample. For Li coverages beyond θLi ≈ 1 ML, no further peak shift is observed, while its intensity decreases. Besides the B 1s shift, the Li-induced charge redistribution is evident in the reduction of the work function (ΔΦ) of the Bo region. As displayed in Figure d, ΔΦ reached approximately −3 eV at θLi ≈ 1 ML and maintained the same value at higher Li coverages, which demonstrates a correlation between B 1s peak shift and ΔΦ. Considering that the work function of Bo/Ir was recently estimated to be 4.7 eV, saturating the sample surface with Li atoms (at θLi ≥ 1 ML) reduces its work function to approximately 1.7 eV, which is significantly lower than the work function of metallic Li (2.9 eV). A similar coverage-dependent decrease in the work function was previously predicted for freestanding borophene α-sheets decorated with Li and Cs, consistent with the strong electron-donating character of alkali atoms. , (a) Color plot of the μ-XPS data in the B 1s energy range, acquired during continuous Li deposition on Bo/Ir. (b) Comparison of B 1s peaks before Li (θLi = 0 ML) and after Li deposition (θLi = 0.5 and 1.5 ML) and in the non-irradiated region on the sample (θLi = 2.8 ML). (c) Comparison of Li 1s peaks for the same samples as shown in panel b. (d) Work function reduction in the Bo region of the sample as a function of the Li coverage. Cartoon insets illustrate the reduction of the surface dipole (from P 0 to P 1) due to the charge donation from Li atoms (blue spheres) to Bo/Ir (pink/gray). For the quantification of the B 1s energy shift, we examine in more detail a few selected Li coverages. To account for the instrumental resolution and avoid misinterpretation of the data, the main B 1s peak is fitted by a single component B1,2 rather than by two components, B1 and B2, identified previously. , At θLi = 0 ML, B1,2 is found at a binding energy (E B) of 188.62 eV (shaded red in Figure b bottom). At θLi = 0.5 ML, B1,2 is located at 188.83 eV of binding energy and shifts further to 188.89 eV at θLi = 1.5 ML. At the same coverage, two additional minor peaks denoted as BC are detected (at E B = 190.43 and 191.73 eV, shaded pink), which we ascribe to contaminated Bo. − Similar contamination was also found previously for Bo/Ir due to the significant reactivity of Bo. , In the present case, those contamination peaks are likely interfered by chemically active Li atoms (see also the following Li 1s description), but remain well-separated in energy from the B1,2 peak and do not disrupt its fitting and quantification. The corresponding Li 1s spectra are shown in Figure c and are fitted well by a single component. At θLi = 0.5 ML, the Li peak is located at E B = 55.90 eV, which is an increased binding (by almost 1 eV) compared to Li metal. , Increased binding energy suggests charge transfer from Li atoms to the Bo/Ir subsystem. At θLi = 1.5 ML, Li 1s binding energy is reduced to 55.79 eV, and the peak width increased by 30% in comparison to the 0.5 ML sample. Such a width increase indicates the coexistence of Li atoms in several chemically similar states that cannot be resolved in our experiments. Besides B–Li compounds, contributions from lithium borates and lithium oxides to the observed Li 1s peak width are possible. ,− By examining another Bo region of the same sample, further shift of the B1,2 to E B = 189.07 eV is found, along with a higher Li coverage of θLi = 2.8 ML, see top panels in Figure b–c. Also, Bo contamination becomes more significant since BC peaks are more pronounced with their spectral weight shifted to higher E B (190.61 and 192.05 eV). Thus, a prolonged exposure of the first examined sample area to the X-ray beam presumably led to radiation-induced chemical reactions, which most likely include desorption of Li. Such effects of irradiation damage have been noticed before in similar experiments. Borophene’s B 1s core level shift at θLi = 2.8 ML is 0.45 eV with respect to the pristine Bo/Ir and these were the highest coverage and the largest B 1s shift attained in our experiments. At the same sample area, Li 1s peak exhibited additional small but detectable reduction of binding energy to E B = 55.76 eV. Apart from the morphological and crystallographic modifications arising from the altered Li coverage (diffraction spot intensity and electron reflectivity changes, described above), no other radiation-related effects on the sample surface have been observed in LEEM or LEED. Annealing of Li-Decorated Sample To assess the thermal stability of Li on the surface and explore the possible formation of new phases, the Li-decorated sample was annealed to elevated temperatures. During annealing of the θLi = 1.6 ML sample at a rate of ∼0.6 °C/s, LEED images averaged over several Bo domains revealed the emergence of additional diffraction spots in the temperature range between 400 and 500 °C, see Figure g and Supporting Information Figure S1 (additional details of the superstructure are given below). A LEEM image of the same annealed sample is given in Figure c. Although Bo and Ir regions can still be identified, their electron reflectivities differ from those observed before annealing (Figure a–b). These reflectivity changes are best captured by the corresponding LEEM-IV curves in Figure d and are noticeably smaller than those caused by Li deposition at RT. In addition, a “transition zone” (Tr) between Ir and Bo regions has formed, appearing very bright in Figure c, along with barely discernible but detectable submicrometer structure in the Ir regions (see Supporting Information Figure S2). The origin of the observed post-annealing diffraction features was clarified by performing micro-spot LEED (μ-LEED) measurements separately on the Bo and Ir regions. This allowed probing of single-domain Bo on a micrometer scale, providing also a better signal-to-noise ratio in comparison to the area-averaged data in Figure g. The Bo region, besides a Bo-(6 × 2) pattern, exhibits signal from the two rotational domains of a high-order commensurate (12/7×12/7)R10.9◦ superstructure (Li- 12/7 hereafter), visible in Figure a (and also in Figure g). Thus, this Li superstructure is independent of the particular Bo domain orientation and keeps its epitaxial relation to the Ir substrate. Assuming one Li atom in the basis of the Li- 12/7 unit cell, θLi associated with this superstructure is 0.58 ML. Notably, Bo diffraction spots of the annealed sample show partially restored intensity and a LEED-IV signature very similar to the Li-free sample (see Figure h). (a) BF-LEEM image (U st = 17 V) of the θLi = 1.6 ML sample annealed to 500 °C, with the μ-LEED images (E = 40 eV) of the Ir and Bo regions shown on the left. The reciprocal unit cells of Ir, Bo and Li-related superstructure are marked in purple, red, and blue, respectively. (b–c) DF-LEEM images generated from Bo-(6 × 2) (a sum of signal from all three Bo rotational domains, see Supporting Information Figure S4 for details) and Li- 12/7 (one rotational domain) diffraction spots (U st = 28 and 22 V, respectively). Yellow dashed lines in the micrographs mark boundaries between Ir and Bo regions on the sample. Ir regions disclose even more pronounced signal from the same Li- 12/7 superstructure, see Figure a, and also a very diffuse but detectable intensity which we ascribe to an incommensurate (1.28 × 1.28) R30° superstructure (Li-1.28 hereafter; see also Supporting Information Figure S2). Nominally, this superstructure corresponds to θLi = 0.61 ML. It is important to note that both of these superstructures were also observed during Li deposition on clean Ir(111) at RT (see Supporting Information Figure S3) and can thus be regarded as a signature of a close interaction between Li and Ir atoms. As expected, no signs of Bo-(6 × 2) are found in the Ir region. These observations are further supported by dark-field LEEM (DF-LEEM) measurements (Figure b-c) which show that the Li- 12/7 signal is more intense in the Ir regions than in the Bo regions. Dark-field data also demonstrates that the two rotational variants of the Li- 12/7 pattern produce indistinguishable DF-LEEM images (not shown), indicating that the corresponding domains, too small to be resolved, have equal spatial distributions. With further annealing to higher temperature, no new superstructure spots emerge while the dominant Li- 12/7 spots continuously loose intensity and disappear completely, in both Bo and Ir regions, as the temperature approaches ∼700 °C (see Supporting Information Figure S1). To better correlate temperature-induced morphology and crystallography with the respective electronic structure, photoemission electron microscopy (PEEM) measurements were conducted on the same sample. Figure d and e show data extracted from PEEM images covering B 1s, Ir 4f7/2 and Li 1s peak positions, recorded at Bo, Tr and Ir regions of the sample. To begin with, B1,2 peak is found at E B = 188.73 eV in the Bo region only, associated with the intrinsic Bo peak shifted due to interaction with remaining Li (see below for Li quantification). However, the largest peak in the Bo region is B3 located at E B = 189.17 eV which exists, even more pronounced, in the transition and Ir regions of the sample. In addition, B4 and B5 peaks are present in the Bo region at lower E B, and they can also be found with altered intensities and binding energies in the transition and Ir regions. It is surprising to find a pronounced B 1s signal in the Ir region, which was essentially free of any boron after Bo synthesis and Li deposition at RT. The amount of B in Ir region is significant, ∼90% of the amount present in the Bo regions, as deduced from the XPS peak areas. The same conclusion can be drawn from the integrated (total) B 1s signal shown in the PEEM image in Figure b, where intensity differences between Bo and Ir regions (cf. Figure a) are barely visible. (a) BF-LEEM image (U st = 17 V) of the θLi = 1.6 ML sample annealed to 500 °C. (b–c) Integrated PEEM images obtained from the B 1s and Li 1s core levels on the same sample area shown in panel a, disclosing spatial distribution of the respective elements. (d–e) PEEM-extracted data of B 1s, Ir 4f7/2 and Li 1s peaks in the Bo, transition and Ir regions of the sample shown in panels a–c. Li 1s peaks are present in Bo, transition and Ir regions at essentially identical positions after annealing56.02, 55.97, and 56.00 eV of binding energy, respectivelyshowing minimal deviation from the pre-annealing data shown in Figure c. However, integrated areas of the Li 1s peaks disclose θLi = 0.4 ML in Bo and θLi = 0.5 ML in the Ir and transition regions. This somewhat higher quantity of Li atoms in the Ir region is also visible in the integrated PEEM image in Figure c. Since θLi = 1.6 ML sample was used in this annealing experiment, it is clear that significant amount of Li atoms desorbed from the sample surface upon reaching 500 °C. Annealing to even higher temperatures (∼700 °C) results in further desorption of Li atoms and eventually yields restoration of the B1,2 peak at the position matched with that of the pristine Bo/Ir (not shown), which proves the reversibility of the Li-induced chemical doping of Bo. Indeed, desorption of Li was also detected in mirror electron microscopy (MEM) measurements during annealing (see Supporting Information Figure S5) and coincides with the loss of the Li- 12/7 superstructure signal in LEED addressed above. Atomistic Insights DFT calculations have been performed to provide a complete picture of Li interaction with Bo monolayer on Ir(111). The χ6 polymorph, reported previously in the literature, ,,, was structurally optimized to obtain the initial atomic configuration (see Figure a) used further for studying its interplay with Li. That particular Bo structure exhibits five (slightly distorted) hexagonal holes and a corrugation (calculated as root-mean-square roughness) of 0.25 Å. The side view in Figure a also showcases small displacements of the Ir surface atoms due to the interaction with Bo. (a–c) Top and side (horizontal) views of the DFT-optimized structures of Bo/Ir(111) before and after adsorption of Li atoms, at two different Li concentrations (Irgray, Bmagenta, Liyellow). Only B–B bonds are shown for clarity. White parallelogram marks the Bo unit cell. (d–e) Charge difference plots of the Li-decorated systems (redcharge gain, bluecharge loss), displaying electronic charge rearrangement along planes indicated by dotted lines in panels b and c. Positions of Li and B atoms in the proximity of those planes are marked by colored circles, horizontal dashed lines represent (averaged) positions of the Ir surface atoms. Considering that hollow hexagons are energetically preferable sites of alkali metal adsorption on Bo sheets, ,, we used the same sites for the initial placement of Li atoms on the Bo/Ir(111) system. Two Li concentrations (coverages) have been examined in the calculations: one Li atom per Bo unit cell (Li/Bo/Ir, θLi = 0.08 ML) and five atoms per Bo unit cell (5Li/Bo/Ir, θLi = 0.42 ML). In the Li/Bo/Ir case, Bo undergoes significant structural deformation in the vicinity of the adsorbed Li atom. Most notably, the hexagonal hole turns into a heptagonal hole, and the corrugation of the Bo sheet increases to 0.38 Å, as illustrated in Figure b. Li atoms reside 1.35 Å above the (averaged) Bo plane, with a formation energy of −2.64 eV per Li atom. In the case of 5Li/Bo/Ir system, depicted in Figure c, the distortion of the Bo lattice increases even further, leading to a corrugation of 0.52 Å. Additionally, the five hexagonal holes within Bo, where Li atoms were initially placed, distort almost beyond recognition after structural optimization. The average distance of Li adatoms above the Bo plane is 1.39 Å, and the formation energy is −1.35 eV per Li atom, indicating weaker binding of Li to Bo as θLi increases. Notably, from the definition of the formation energy (see Methods), its negative values at both calculated Li coverages indicate stability against the formation of Li clusters. The absence of clustering was also theoretically predicted earlier for Li adsorption onto freestanding Bo polymorphs. , Bader charge analysis of the Li-decorated systems shows that the charge is transferred from Li adatoms to the Bo/Ir subsystem. In the Li/Bo/Ir case, the Li adatom is left with +0.86 e and averages to +0.84 e per Li adatom in the 5Li/Bo/Ir case (see also Supporting Information Figure S6). Such significant charge donation from Li atoms can be linked to surface dipole attenuation and a reduction of the sample’s work function (cf. Figure d). Notably, similar charge transfers have been calculated for the adsorption of Li on freestanding Bo polymorphs. , Details of charge redistribution are better resolved in charge difference (Δρ) plots shown in Figure d–e. In a situation of dilute adsorption (Li/Bo/Ir), charge transfer from Li is limited to B atoms in its vicinity, which can be correlated with the localized structural impairment of Bo described above. At higher coverage (5Li/Bo/Ir), electronic charge from Li adatoms is distributed to all B atoms, which causes much larger deformation of the Bo layer. Li adsorption also affects the charge associated with the Ir surface atoms, although to a smaller extent (see also Supporting Information Figure S7). The absence of charge accumulation between Li and B atoms signifies their ionic-like bonding rather than covalent. Charge donation from Li atoms is also reflected in the calculated binding energies of the B 1s core levels, which exhibit large variations across the computational unit cell and do not show an unambiguous energy shift (see Supporting Information Figure S8). Such energy variations call for additional calculations and a dedicated investigation of a larger configurational space of this susceptible system. It is insightful to compare calculations of Li adsorption on Bo/Ir with Li adsorption on bare Ir. Considering θLi = 0.08 ML on Ir, it is found that the Li formation energy is −1.56 eV (at favorable Ir hcp sites), thus being less bound than Li on Bo/Ir. For the same Li/Ir system, the charge of the Li atom is +0.88 e. As the Li coverage increases, the Li adsorption energy increases (i.e., Li binding to Ir is reduced), and Li becomes less charged: at θLi = 0.58 ML, the adsorption energy equals to −1.04 eV per Li atom, and the calculated Li charge is +0.8 e (see also Supporting Information Figure S9). Importantly, such Li coverages of approximately 0.5–0.6 ML, corresponding to the Li- 12/7 superstructure observed in the experiments, become preferred for a specific range of Li chemical potentials (see Supporting Information Figure S10). Lastly, to assess the likelihood of Li intercalation at the interface of Bo and Ir, the optimized Li/Bo/Ir system shown in Figure b was taken as the initial configuration for the movement of Li atom downward (along the z-direction) in a sequence of 0.2 Å steps. Snapshots of five selected configurations are shown in Figure a, and the corresponding energy of the system, referenced to the initial configuration, is shown in Figure b. Clearly, the total energy increases as the Li atom is pushed through the heptagonal hole of the Bo layer, indicating that the residence of the Li atom at Bo-Ir interface is not preferable. (a) A sequence of selected Li/Bo/Ir configurations calculated by DFT, where a Li atom is placed at different distances above and below the heptagonal hole of the Bo layer. The system is shown at reaction coordinates c = 0, 3, 5, 7, and 9, simulating the movement of a Li atom (yellow sphere) through the Bo layer (magenta spheres) toward the Ir substrate (gray spheres). (b) Total energy of the system (ΔE tot) at ten examined reaction coordinates, relative to the initial configuration at c = 0. Discussion Our calculations suggest a preference for Li adsorption on Bo/Ir over its intercalation, where intercalation in a conventional sense might even be an inappropriate term since the Bo lattice significantly distorts during the Li adsorption process, especially at higher θLi. The distortion of the Bo crystal structure is also suggested by the diminishing LEED spots of Bo during Li deposition. Furthermore, the absence of intercalation is supported by only minor modifications of the Li 1s peak in a wide θLi range, whereas distinctive peaks could be expected due to significant changes in the chemical environment of Li atoms upon intercalation. Li intercalation at defects and domain boundaries of Bo is unlikely since our data does not show signatures of defective as-synthesized Bo (e.g., defect-related peaks in B 1s XPS or spot broadening in LEED) or the evolution of any distinct morphological features during Li deposition, such as the progression of the intercalation front away from the domain boundaries in LEEM. Thus, we infer that Li atoms reside on top of the Bo layer after deposition at RT, contrary to the case of Au intercalation at the Bo/Ir interface. In such an adsorption geometry, Li adatoms do not form ordered structures, as deduced from LEEM and LEED, nor do they form clusters, according to the DFT calculations. Thus, both the Coulomb repulsion between positively charged Li atoms (in the low θLi regime) and the Li–Li attraction which would be relevant in the metallic Li phase (in the high θLi regime) are overshadowed by a strong interaction between Li and B atoms, which limits Li surface diffusion and results in a disordered Li structure. Annealing induces structural and chemical changes in the Li-decorated sample. Heating to 500 °C causes partial Li desorption from the Bo regions, but only down to θLi ≈ 0.4 ML when a metastable configuration of the system is achieved. At the same time, the structural integrity of the Bo layer seems to be compromised: Bo regions consist of chemically distinct patches detectable in μ-XPS and μ-LEED, but apparently too small to be clearly visible in LEEM. B1,2 peak can be associated with the Bo layer, chemically shifted to higher binding energy due to the presence of remaining Li. The same Bo patches exhibit Bo-(6 × 2) pattern in μ-LEED data, confirming the persistence of the χ6 polymorph. However, the most abundant form of boron in the annealed Bo region, giving rise to the largest B3 component, presumably exhibits the Li- 12/7 superstructure. This is inferred from a comparison with the μ-XPS and μ-LEED data of Ir regions, where the same boron peak and superstructure dominate. Since the Li- 12/7 superstructure was detected after Li deposition on clean Ir (cf. Supporting Information Figure S3), we assume that the structure of this material is governed by the Li–Ir interaction and that B atoms play a secondary role in it, without forming a well-defined ternary interfacial compound. The two remaining B species in the Bo region, yielding minor peaks B4 and B5, might be related to other disordered forms of boron or the Li-1.28 superstructure, which can also be better recognized in the Ir region of the sample. Temperature-triggered Li intercalation is unlikely due to the significant modifications in sample crystallography and the diversity of post-annealing B 1s peaks, in addition to the already discussed unfavorable intercalation inferred from DFT. Therefore, the deposition of Li on epitaxial Bo differs significantly from the deposition on conventional van der Waals materials in the monolayer limit, such as graphene or hexagonal boron nitride on Ir(111), where Li intercalation was easily achieved already at RT. ,− From μ-XPS and PEEM, we found that boron atoms are distributed across all regions on the sample after Li deposition and annealing at 500 °C, with Ir regions containing as much as 90% of B compared to Bo regions. Thus, Ir regions get populated by B atoms in the annealing process. We assume that boron atoms might come to the Ir regions either (i) from the Ir subsurface or (ii) by detaching from the surrounding Bo layer at elevated temperature. On one hand, segregation of additional B from Ir already at 500 °C might be facilitated by the presence of Li which could lower the segregation energy barrier, as was demonstrated for subsurface boron dissolved in iridium and iron in the presence of NH3 or N2. − On the other hand, boron detachment from Bo layer and its diffusion toward the Ir regions, with possible temporary accumulation in the transition regions, could be mediated by the high chemical reactivity of Li atoms, especially at elevated temperatures. Either way, the available boron atoms do not self-assemble into a χ6 polymorph with a (6 × 2) diffraction pattern but rather bind to Li–Ir surface compounds without disrupting their intrinsic Li- 12/7 and Li-1.28 superstructures. Such material could be the origin of the stripe-like contrast detected in LEEM (see Supporting Information Figure S2). Even though other metal substrates host different Bo polymorphs, we believe our findings could be relevant for them since all polymorphs consist of a triangular boron mesh perforated by an array of hexagonal holes, which could result in similar Li-Bo interaction. However, even then, and especially during annealing, differences in Bo-substrate binding and Li-substrate interaction might lead to the formation of novel surface structures with altered chemical signatures. Therefore, dedicated studies of other alkali-decorated epitaxial Bo polymorphs are suggested to further elucidate polymorph- and substrate-specific effects. Conclusions Deposition of Li onto borophene on Ir(111) results in an intricate interplay between Li and B atoms, which is promoted by the significant chemical reactivity and binding diversity of these elements. Our results show that the deposition of Li onto Bo/Ir(111) at room temperature leads to a surface adsorption rather than intercalation while inducing structural changes in the borophene lattice. The observed structural modifications originate from electronic charge donation by Li atoms and consequent chemical doping of borophene, which interrupt the charge balance existing within the pristine borophene polymorph. Annealing of Li-decorated samples results in partial desorption of Li and triggers the formation of an inhomogeneous material on the surface, consisting of patches of borophene and B–Li–Ir complexes whose crystallography is governed by the interaction between Li and Ir atoms. The present study points to significant structural and chemical modifications of monolayer borophene induced by Li adsorption and can be taken as a benchmark in future alkali deposition experiments as well as in electric battery-related examinations of borophene. Methods Sample Preparation All sample preparation was conducted in ultra-high vacuum (UHV) conditions. The Ir(111) single-crystal (MaTecK GmbH) was cleaned by RT argon sputtering at 1.5 keV, oxygen glowing at p O2 = 10–6 mbar and 850 °C, and annealing at 1200 °C. Sharp Ir diffraction spots in LEED and absence of contamination peaks in survey XPS data were taken as indicators of a well-prepared Ir surface. Bo was synthesized by repeated exposure of Ir crystal to borazine (B3H6N3, Katchem spol. s r. o.) vapors at a partial pressure p B = 2 × 10–8 mbar and temperature 1100 °C, thus dissolving B atoms in Ir bulk which segregated back to the surface and condensed into a borophene layer during sample cooldown. , Li deposition was carried out by resistive heating of commercial alkali dispensers (SAES Getters S.p.A.) with currents up to 7.5 A. θLi was estimated from a comparison of Ir 4f and Li 1s peak areas, considering differences in photoionization cross sections, transmission of the electron analyzer, and the finite probing depth of the photon beam. Within this scheme, at θLi = 1 ML, the surface concentration of Li atoms is equal to the concentration of Ir atoms in the topmost substrate layer. Due to the approximation just described, as well as photon and electron beam effects described in the text, error bars of θLi are estimated to be ± 0.1 ML. Instrumentation All experimental measurements were performed in situ in UHV using the Spectroscopic Photoemission and Low Energy Electron Microscope (SPELEEM III, Elmitec GmbH), installed at the Nanospectroscopy beamline of the Elettra synchrotron in Trieste, Italy, with a base pressure below 2 × 10–10 mbar. The microscope is equipped with a hemispherical energy analyzer (Elmitec R200) having an energy resolution of 100 meV. During LEEM (resolution better than 10 nm) and LEED imaging, the sample was illuminated with electrons, while in PEEM (resolution better than 30 nm) and XPS mode, soft X-rays with linear polarization were used to generate photoelectrons. Apertures were used to limit the illumination or probing area on the sample to micrometer-sized regions, thus enabling μ-LEED and μ-XPS studies. Post-processing of PEEM data allowed the extraction of photoemission signal from even smaller and irregular regions on the sample. In PEEM and XPS, photons with energies of 130 eV for Li 1s (and Ir 4f) and 270 eV for B 1s (and Ir 4f) were used. The binding energies of the photoemission spectra are referenced to the Fermi level. XPS peaks were fitted by Voigt line shapes and plotted with the Shirley background subtracted. Work function measurements were done by observing the shift of the secondary electron cutoff in the dispersive plane of the microscope. In all modes of operation, electrons were imaged on a 2D detector comprising a phosphorous screen and a CCD camera (QImaging Retiga). The electron energy was set by applying a voltage bias (start voltage, U st) to the sample. Unless otherwise noted, all data were recorded at room temperature. Theoretical Calculations First-principles calculations were conducted using DFT as implemented in the Vienna Ab initio Simulation Package − and OpenMX code. The exchange-correlation effect among electrons was examined using the vdW-DF2-B86R functional, which incorporates van der Waals corrections. We employed a slab model for the borophene/Ir superstructure, with the in-plane lattice parameter constrained to that of the Ir(111)-(6 × 2) lattice and four Ir layers. The separation between the slabs was determined to exceed 12 Å. The core-hole pseudopotential for B 1s and Li 1s, incorporating a penalty functional and a Coulomb-cutoff method as proposed by Ozaki and Lee, was employed in the calculations of the core-level excitation. To reduce artificial interactions between core holes, the calculations employed a supercell corresponding to the Ir(111)-(6 × 6) lattice. The formation energy was defined as E form = E nLi/Bo/Ir – (E Bo/Ir + nE rhom–Li) where n is the number of Li atoms per unit cell and E rhom–Li is the energy per Li atom in rhombohedral Li used as the reference state, which was found to be approximately 1.5 meV per Li atom more stable than bcc Li. Supplementary Material