|
|
| (3 intermediate revisions by the same user not shown) |
| Line 5: |
Line 5: |
| =Exploratory V-Si RF-DC co-sputtering in Cluster Lesker PC3= | | =Exploratory V-Si RF-DC co-sputtering in Cluster Lesker PC3= |
|
| |
|
| This page documents an early, non-systematic attempt to deposit vanadium silicide by simultaneous RF sputtering of a 4-inch Si target and DC sputtering of a 3-inch V target in the <i>Cluster Lesker PC3</i> chamber. Two otherwise nominally identical depositions were compared: one without RF substrate bias and one with 30 W RF substrate bias. | | This page documents an <b>early</b>, <b>non-systematic attempt</b> to deposit vanadium silicide by simultaneous RF sputtering of a 4-inch Si target and DC sputtering of a 3-inch V target in the <i>Cluster Lesker PC3</i> chamber. Two otherwise nominally identical depositions were compared: one without RF substrate bias and one with 30 W RF substrate bias. |
| | |
| | We achieved the expected results for the V<sub>3</sub>Si process in PC1 and describe them [[Specific Process Knowledge/Thin film deposition/Deposition of Vanadium Silicide/V3Si RF-DC Co-Sputtering in Cluster Lesker PC1|in dedicated page]]. |
|
| |
|
| The experiment gave a striking result. In the measured 30–80° GiXRD range, the unbiased sample showed only reflections assigned to cubic A15 V<sub>3</sub>Si, whereas the sample deposited with 30 W RF substrate bias showed only reflections assigned to hexagonal C40 VSi<sub>2</sub>. The two samples also had very different surface morphologies and sheet resistances. | | The experiment gave a striking result. In the measured 30–80° GiXRD range, the unbiased sample showed only reflections assigned to cubic A15 V<sub>3</sub>Si, whereas the sample deposited with 30 W RF substrate bias showed only reflections assigned to hexagonal C40 VSi<sub>2</sub>. The two samples also had very different surface morphologies and sheet resistances. |
| Line 177: |
Line 179: |
| ! XRR density | | ! XRR density |
| ! XRR roughness | | ! XRR roughness |
| ! Sheet resistance
| | |
| ! Calculated resistivity
| |
| |- | | |- |
| | No substrate bias | | | No substrate bias |
| Line 186: |
Line 187: |
| | 4.741 g/cm<sup>3</sup> | | | 4.741 g/cm<sup>3</sup> |
| | 7.873 nm | | | 7.873 nm |
| | 29.34 Ω/sq
| | |
| | 276 µΩ·cm
| |
| |- | | |- |
| | 30 W RF substrate bias (69 V DC bias) | | | 30 W RF substrate bias (69 V DC bias) |
| Line 195: |
Line 195: |
| | Not obtained | | | Not obtained |
| | Not obtained | | | Not obtained |
| | 6.64 Ω/sq
| |
| | Not calculated because thickness is unknown
| |
| |} | | |} |
|
| |
|
| Line 298: |
Line 296: |
| The 30 W RF-biased film was too rough for a defensible XRR model. No thickness, density, roughness or resistivity should be quoted from this scan. A cross-sectional measurement is needed. | | The 30 W RF-biased film was too rough for a defensible XRR model. No thickness, density, roughness or resistivity should be quoted from this scan. A cross-sectional measurement is needed. |
|
| |
|
| =SEM surface morphology= | | =Further characterization and data access= |
| | |
| <gallery caption="" widths="800px" heights="570px" perrow="1">
| |
| image:eves_SEM_V3Si_PC3_bias_comparison_20260916.png|Plan-view SEM comparison at 200,000× and 300,000×. Original scale bars and microscope metadata are preserved in each panel.
| |
| </gallery>
| |
| | |
| Both films are rough and polycrystalline, but the morphologies differ strongly:
| |
| | |
| * the unbiased V<sub>3</sub>Si film consists mainly of much finer, densely distributed granular/nodular surface features, with occasional brighter agglomerates;
| |
| * the 30 W-biased VSi<sub>2</sub> film contains substantially larger faceted plate-like grains, many with visible terraces or banded surface contrast;
| |
| * the coarser faceted morphology of the biased sample is consistent with its strongly damped XRR response, but plan-view SEM cannot provide an RMS roughness or film thickness.
| |
| | |
| =Electrical results=
| |
| | |
| The unbiased V<sub>3</sub>Si sample had a sheet resistance of 29.34 Ω/sq. Combining this value with the XRR thickness gives:
| |
| | |
| <math>
| |
| \rho=R_s t
| |
| =29.34\ \Omega/\square \times 94.046\ \mathrm{nm}
| |
| =276\ \mu\Omega\cdot\mathrm{cm}
| |
| </math>
| |
| | |
| The 30 W-biased VSi<sub>2</sub> sample had a much lower sheet resistance of 6.64 Ω/sq, but this does <b>not</b> by itself demonstrate a lower material resistivity or a better film. Sheet resistance depends on both resistivity and thickness. The biased layer may be considerably thicker because of reaction with the Si wafer, may contain an electrically parallel interfacial/unreacted layer, or may simply have a different intrinsic resistivity and connectivity. Until its thickness and cross-sectional structure are measured, only the sheet resistance should be reported.
| |
| | |
| =Interpretation of the bias-induced phase change=
| |
| | |
| The most likely working interpretation is that the oxide-stripped Si substrate was an active reactant and that RF substrate bias changed the early interface reaction sufficiently to select a different silicide phase.
| |
| | |
| # <b>The targets supplied a V-rich flux.</b> The estimated target-derived atomic ratio was V:Si = 4.22, which is more V-rich than V<sub>3</sub>Si. Formation of V<sub>3</sub>Si in the unbiased sample therefore requires some Si uptake from the wafer, but only a modest amount: approximately 11.7 nm of substrate-equivalent Si in the simple full-retention mass balance.
| |
| # <b>Bare crystalline Si provides a very large Si reservoir.</b> Earlier V/Si interface studies report that VSi<sub>2</sub> is the first silicide phase to nucleate and grow at a single-crystal Si/V interface. More recent V<sub>3</sub>Si work likewise shows that limiting the available Si favors the V-rich A15 phase, whereas an unrestricted Si supply favors Si-rich silicides.
| |
| # <b>The unbiased run may have formed an early V<sub>3</sub>Si-rich reaction layer.</b> With the V-rich incoming flux, such a layer could kinetically limit further Si transport from the wafer and preserve the A15 phase during the remainder of the 1200 s deposition.
| |
| # <b>The 69 V substrate DC bias introduced energetic ion bombardment.</b> Ion bombardment can increase defect generation, surface mobility, atomic mixing and re-sputtering. In this experiment it may have disrupted or delayed formation of the V<sub>3</sub>Si-rich barrier, enhanced access to substrate Si, and allowed the thermodynamically/kinetically favored Si-rich VSi<sub>2</sub> reaction to continue. This is a plausible mechanism, not a direct measurement.
| |
| # <b>The bias also changed the plasma state.</b> The Si display changed from 100 to 110 V and the V discharge voltage from 322 to 309 V at fixed applied target powers. The actual V and Si arrival rates may therefore have shifted relative to the separate elemental calibrations. Voltage alone is insufficient to calculate the change in composition.
| |
| | |
| The data demonstrate that 30 W RF substrate bias is not a small microstructural adjustment for this process. On HF-stripped Si at 600 °C it changes the detected crystalline product from V<sub>3</sub>Si to VSi<sub>2</sub> and produces a much coarser surface morphology.
| |
| | |
| =Limitations=
| |
| | |
| * Only two depositions were compared; there is no bias-power series and no repeatability test.
| |
| * The elemental deposition rates were measured separately, not simultaneously during co-sputtering at 600 °C.
| |
| * No direct V:Si composition measurement is available.
| |
| * GiXRD establishes crystalline phase identity within its detection limits but does not quantify amorphous material or reliable phase fractions.
| |
| * The biased sample has no measured thickness; therefore its resistivity and true deposition/reaction rate are unknown.
| |
| * Plan-view SEM does not reveal the film/substrate interface or how much Si was consumed.
| |
| * The influence of the HF-to-load delay and any regrown interfacial oxide was not isolated.
| |
| * Phase-pure A15 diffraction does not by itself prove superconductivity; a temperature-dependent resistance measurement is required.
| |
| | |
| =References= | |
| | |
| * W. Zhang et al., “Thin-film synthesis of superconductor-on-insulator A15 vanadium silicide,” <i>Scientific Reports</i> 11, 2358 (2021), [https://doi.org/10.1038/s41598-021-82046-1 DOI: 10.1038/s41598-021-82046-1]. This work discusses control of Si supply during V<sub>3</sub>Si formation and provides the V/Si molar-volume calculation used as a cross-check here.
| |
| * R. J. Schutz et al., “The formation of vanadium silicides at thin-film interfaces,” <i>Journal of Applied Physics</i> 50, 5773–5781 (1979), [https://doi.org/10.1063/1.326718 DOI: 10.1063/1.326718]. This study reports VSi<sub>2</sub> as the first silicide phase to nucleate and grow at single-crystal Si/V interfaces.
| |
| * M. Bose et al., “In-situ Investigation of the Phase Formation and Superconductivity in V<sub>3</sub>Si Thin Films at High Temperatures,” [https://arxiv.org/abs/2412.04159 arXiv:2412.04159]. This work shows that interface/barrier consumption and excessive Si access can drive formation of undesirable Si-rich silicides.
| |
| * M. Manzo-Perez et al., “Substrate effects on phase formation and interfacial stability in superconducting vanadium silicide thin films,” <i>Applied Physics Letters</i> 127, 122601 (2025), [https://doi.org/10.1063/5.0291576 DOI: 10.1063/5.0291576].
| |
|
| |
|
| [[Category:Thin film deposition]]
| | SEM images, electrical properties, sheet resistances, and the complete integrated interpretation are available on the corresponding <b>LabIntra page</b> or upon request. Please contact <b>Evgeniy Shkondin, DTU Nanolab</b>, for further details and results. |
| [[Category:Sputtering]]
| |
| [[Category:Vanadium silicide]]
| |
| [[Category:Cluster Lesker PC3]]
| |
