Combinatorial Deposition & Ex-situ Selenization

• Combinatorial deposition and ex-situ selenization is a method for synthesizing tunable TMD thin films by sequential oxide layering and CVD selenization.
• The process enables precise control over substitutional dopant levels, facilitating systematic composition–property mapping across wafer-scale substrates.
• Comprehensive characterization via RBS, XPS, Raman, and FET measurements confirms the conversion efficiency and reveals key optoelectronic properties.

Combinatorial deposition combined with ex-situ selenization is a workflow for the synthesis of compositionally tunable transition metal dichalcogenide (TMD) thin films, enabling systematic control of substitutional dopant levels and their impact on optoelectronic properties. As demonstrated for the synthesis of vanadium-doped tungsten diselenide, W$_{1-x}$V$_x$Se$_2$, this strategy uses sequential deposition of V$_2$O$_5$ and WO$_3$ bilayers under high vacuum on SiO$_2$/Si substrates, followed by atmospheric-pressure chemical vapor deposition (CVD) in selenium vapor to yield alloyed TMD films with precise stoichiometry (Bajgain et al., 28 May 2025). This approach allows composition–property mapping across a wafer and supports scalable doping protocols for device-oriented studies.

1. Deposition Protocols and Combinatorial Bilayer Design

Highly doped silicon wafers with a 300 nm thermal SiO$_2$ top layer serve as substrates, supporting field-effect device fabrication via global back-gate configuration. Bilayer oxide films are prepared by sequential thermal evaporation of WO$_3$ and V$_2$O$_5$ in a high vacuum environment ($\sim$10^{-6}$$Torr) at room temperature. No sputtering or radio-frequency power is used. The bilayer thicknesses are systematically set to target specific vanadium fractions ($$x$$) in the final alloy via a combinatorial rule-of-mixtures estimate. For discrete compositional control, two architectures are reported:</p> <div class='overflow-x-auto max-w-full my-4'><table class='table border-collapse w-full' style='table-layout: fixed'><thead><tr> <th>Sample</th> <th>$$t_V$(nm)</th> <th>$t_W$(nm)</th> <th>$x_\mathrm{RBS}$$</th> </tr> </thead><tbody><tr> <td>A</td> <td>1.9</td> <td>11</td> <td>0.17</td> </tr> <tr> <td>B</td> <td>3.8</td> <td>9.1</td> <td>0.32</td> </tr> </tbody></table></div> <p>The alloy fraction for a given bilayer set is estimated as</p> <p>$$x \approx \frac{t_V}{t_V+t_W}$</p> <p>where$t_V$is the thickness of V$_2$O$_5$, and$t_W$is that of WO$_3$$. Empirical calibration by Rutherford Backscattering Spectroscopy (RBS) yields an approximate scaling law:</p> <p>$$x(t_V) \simeq 0.084\,t_V\,,\quad t_V\text{ in nm}$$</p> <p>For combinatorial mapping, wedge deposition of$$t_V$across the wafer would produce a continuous$x$$gradient suitable for high-throughput composition–property mapping.</p> <h2 class='paper-heading' id='ex-situ-selenization-conditions-and-reaction-chemistry'>2. Ex-situ Selenization Conditions and Reaction Chemistry</h2> <p>Conversion of the oxide bilayers to chalcogenide alloys employs an ex-situ, atmospheric-pressure CVD furnace. Elemental Se placed upstream is sublimed to generate a Se vapor flux, while Ar in excess and trace H$$_2$$flow facilitate delivery and oxide reduction; typical flows are 200–500 sccm (Ar) and 10–50 sccm (H$$_2$). The temperature is ramped from ambient to$700\textrm{–}800\,^\circ$$C and held for 10–30 minutes before natural cooling. The overall reaction for arbitrary$$x$is:</p> <p>$(1-x)\,\mathrm{WO}_3 + \tfrac{x}{2}\,\mathrm{V}_2\mathrm{O}_5 + 2\,\mathrm{Se} \to \mathrm{W}_{1-x}\mathrm{V}_x\mathrm{Se}_2 + \Bigl(3-\frac{x}{2}\Bigr)\,\mathrm{O}_2$$</p> <p>This chemistry converts the oxide bilayer to a uniform, ∼15–16 nm thick W$$_{1-x}$V$_x$Se$_2$$film, fully preserving compositional gradients established by the initial bilayer design.</p> <h2 class='paper-heading' id='structural-and-compositional-characterization'>3. Structural and Compositional Characterization</h2> <p>Comprehensive assessment of film composition and structure is achieved via RBS, X-ray photoelectron spectroscopy (XPS), and Raman spectroscopy. RBS confirms full conversion and quantifies$$x$$at 0.17 and 0.32 for the representative bilayers. XPS corroborates the presence of W$$^{4+}$, V$^{3+}$/V$^{4+}$, and Se$^{2-}$$, consistent with complete selenization and alloying. Raman spectra display systematic evolution:</p> <ul> <li>Pristine WSe$$_2$:$A_{1g}$at 257 cm$^{-1}$,$E_{2g}^1$at 253 cm$^{-1}$.</li> <li>$x=0.17$:$E_{2g}^1$redshifted by 4 cm$^{-1}$, FWHM increased by 8 cm$^{-1}$.</li> <li>$x=0.32$:$E_{2g}^1$redshifted by 6 cm$^{-1}$, FWHM increased by 11 cm$^{-1}$. A feature at 197 cm$^{-1}$, assigned to 1T-VSe$_2$, emerges at higher$x$$.</li> </ul> <p>No X-ray diffraction data are reported in the main text.</p> <h2 class='paper-heading' id='optoelectronic-property-mapping'>4. Optoelectronic Property Mapping</h2> <p>Back-gated field-effect transistors (FETs) are fabricated from the alloy films using a two-probe geometry ($$L \approx 40~\mu$m). Linear$I_d$–$V_d$$characteristics verify ohmic contacts. The field-effect mobility in the linear regime is</p> <p>$$\mu = \frac{L}{W\,C_\mathrm{ox}\,V_d}\, \frac{dI_d}{dV_g}$</p> <p>with$C_\mathrm{ox} = \epsilon_0\epsilon_r/d_\mathrm{SiO_2}$. The on/off ratio is</p> <p>$R_\mathrm{on/off} = \frac{I_d(V_g=V_{g,\max})}{I_d(V_g=V_{g,\min})}$$</p> <p>Temperature-dependent transport (200–380 K) reveals:</p> <ul> <li>Pristine WSe$$_2$:$I_d$increases with$T$$(semiconducting behavior), fit by activated transport$$I_d \propto \exp(-E_a/k_BT)$.</li> <li>$x=0.17,\,0.32$:$I_d$decreases with$T$, phenomenologically fit as$\rho(T) = \rho_0 + A\,T^n$$(metallic-like).</li> </ul> <p>Photoconductive gain at 532 nm excitation is defined as</p> <p>$$G(T) = \frac{I_d^\mathrm{light} - I_d^\mathrm{dark}}{I_d^\mathrm{dark}} \times 100\%$$</p> <p>with observed maxima:</p> <ul> <li>Pristine:$$G_\mathrm{max} \approx 30\%$at low$T$, decreasing with$T$.</li> <li>$x=0.17$:$G_\mathrm{max} \approx 8\%$.</li> <li>$x=0.32$$: Gain negligible.</li> </ul> <p>Photoresponse times are 50–60 ms rise time, with no evidence of long-lived carriers ($$\tau \lesssim$ms). The photocurrent power-law is$I_\mathrm{ph} = A\,P^b$, with$b \simeq 0.29$($x=0$),$b \simeq 0.40$($x=0.17$). A simplified gain expression is$G \approx \tau\,\mu\,V_d/L^2$$(not fit in the paper).</p> <h2 class='paper-heading' id='composition-property-mapping-and-wafer-scale-implications'>5. Composition–Property Mapping and Wafer-Scale Implications</h2> <p>Systematic variation of$$t_V$(and therefore$x$$) via combinatorial deposition enables high-throughput exploration of carrier concentration, field-effect mobility, on/off ratio, activation energy (for insulator–metal transitions), and photoconductive gain. Notably, an insulator-to-metal transition is observed between$$x \approx 0$and$x \approx 0.17$. For$x \gtrsim 0.17$, the alloy demonstrates degenerate$p$$-type transport with diminished photoresponse. The workflow:</p> <ul> <li>Bilayer oxide “wedge” deposition (spatially varying$$t_V$$).</li> <li>Ex-situ CVD selenization.</li> <li>Global assessment by FET/photoconductivity mapping. enables composition–property maps over large substrates without multiple CVD cycles.</li> </ul> <p>The observed empirical scaling,$$x \simeq 0.084\,t_V$$(nm), provides a practical handle to achieve arbitrary V-fraction alloys in the 0–0.32 regime given precise deposition control.</p> <h2 class='paper-heading' id='broader-applications-and-methodological-context'>6. Broader Applications and Methodological Context</h2> <p>This combinatorial deposition and ex-situ selenization strategy offers a pathway to tunable, wafer-scale transition metal dichalcogenide alloys and heterostructures. Its precision in setting dopant fraction by deposit geometry and scalability for property mapping render it compatible with exploration of insulator–metal transitions, dopant-induced band structure modification, and device optimization in 2D material electronics and optoelectronics. The methodology is directly extensible to other oxide/selenide or oxide/sulfide systems where substitutional doping of <a href="https://www.emergentmind.com/topics/janus-transition-metal-dichalcogenides-tmds" title="" rel="nofollow" data-turbo="false" class="assistant-link" x-data x-tooltip.raw="">TMDs</a> is required, subject to adaptation of selenization conditions.</p> <p>The approach as demonstrated for V$$_2$O$_5$/WO$_3$bilayers and W$_{1-x}$V$_x$Se$_2$ thin films illustrates the potential for systematic exploration of composition–structure–property relationships using a minimal number of growth cycles, establishing a robust paradigm for dopant tuning in layered chalcogenide materials (Bajgain et al., 28 May 2025).