Specific Process Knowledge/Thin film deposition/Deposition of Vanadium Nitride/VN Reactive Sputtering in Cluster Lesker PC1

This page was written by Evgeniy Shkondin, DTU Nanolab, unless otherwise stated.
All images and photographs on this page belong to DTU Nanolab.
The fabrication and characterization described below were carried out in 2026 by Evgeniy Shkondin, DTU Nanolab.

Vanadium Nitride p-DC Reactive Sputtering

This document presents the development and characterization of vanadium nitride (VN_x) thin films deposited by reactive pulsed-DC (p-DC) sputtering in the Metal-Oxide sputter system (PC1), commonly referred to as the “Cluster Lesker.” A 3-inch elemental V target was mounted in Source #5, which can be operated in DC or p-DC mode.

Vanadium nitride is a hard, electrically conductive refractory ceramic with good resistance to wear, corrosion and elevated temperatures. Depending on composition and defect concentration, VN can also become superconducting at low temperature. These properties make VN_x interesting for protective coatings, diffusion barriers, electrodes and superconducting thin-film devices. Its electrical and structural behaviour is particularly sensitive to the N/V ratio, vacancies, secondary phases, oxygen contamination and deposition temperature.

Two related process series were prepared to separate the effects of reactive-gas fraction and substrate temperature. Wafers 1–5 were deposited at 400 °C, whereas wafers 6–10 were deposited at 600 °C. Within each series, the N₂ fraction in a constant total Ar/N₂ flow of 40 sccm was increased from 5 to 25%. The pairs 1/6, 2/7, 3/8, 4/9 and 5/10 therefore have the same gas composition but different deposition temperatures.

All fabrication and characterization were carried out in 2026 by Evgeniy Shkondin, DTU Nanolab. Film thickness, density and roughness were investigated by X-ray reflectivity (XRR); crystalline phases were examined by grazing-incidence X-ray diffraction (GiXRD); surface-region composition was measured by X-ray photoelectron spectroscopy (XPS); and the near-elastic loss structure was studied by reflection electron energy-loss spectroscopy (REELS). Sheet resistance was also measured to follow changes in electrical performance.

The substrates were 150 mm Si wafers coated with 300 nm of thermally grown SiO₂. The oxide provides electrical isolation and should be considered when interpreting substrate, background and interface-related features.

The baseline recipe used in the Metal-Oxide sputter system (PC1) was as follows:

• Recipe name: MD PC1 Src5 p-DC Downstream with reactive N2 gas

• Deposition type: R p-DC
• Power: 400 W
• Pressure: 3 mTorr
• Gas: Ar and N₂ (Variable)
• Frequency: 100 kHz
• Reverse time: 1 µs
• Deposition time: 1800 s
• Temperatures: 400°C and 600°C

All depositions were performed for 1800 s at 400 W and 3 mTorr. The total process-gas flow was kept at 40 sccm; only the Ar/N₂ balance and substrate temperature were changed. Pulsed-DC operation was selected to improve discharge stability during reactive sputtering and to reduce the probability of arcing as the V target surface became progressively nitrided.

Deposition conditions

The table below lists the recorded process parameters for all ten wafers. Wafers 1–5 form the 400 °C series and wafers 6–10 the 600 °C series. In both series, the N₂ flow was raised in 2 sccm steps while the Ar flow was reduced by the same amount, maintaining a total flow of 40 sccm. Apart from substrate temperature and gas composition, the nominal settings were kept constant. The reported DC-bias values represent the measured discharge-voltage magnitude during operation.

VNx Reactive p-DC sputtering deposition conditions

Deposition conditions using Reactive p-DC mode Wafer number Power (W) DC bias (V) Pressure (mTorr) Frequency (kHz) Reverse time (µs) Temperature (°C) Ar flow (sccm) N2 flow (sccm) Time (s) wafer 1 400 417 3 100 1 400 38 2 1800 wafer 2 400 432 3 100 1 400 36 4 1800 wafer 3 400 440 3 100 1 400 34 6 1800 wafer 4 400 449 3 100 1 400 32 8 1800 wafer 5 400 454 3 100 1 400 30 10 1800 wafer 6 400 417 3 100 1 600 38 2 1800 wafer 7 400 426 3 100 1 600 36 4 1800 wafer 8 400 436 3 100 1 600 34 6 1800 wafer 9 400 446 3 100 1 600 32 8 1800 wafer 10 400 451 3 100 1 600 30 10 1800

The discharge voltage increases as the N₂ fraction is raised: from 417 to 454 V at 400 °C and from 417 to 451 V at 600 °C. Such a systematic change is consistent with a changing target-surface state during reactive sputtering. It is therefore important not to interpret the N₂ percentage in the gas as the stoichiometric value x in VN_x; the relationship between gas flow and incorporated nitrogen is generally nonlinear.

Deposition performance

The performance table combines the measured thickness, calculated average deposition rate, principal GiXRD phase assignment and sheet resistance. The average rate was calculated simply as film thickness divided by the 1800 s deposition time. No reliable thickness was available for wafers 1 and 6, so their deposition rates are left blank.

VNx Reactive p-DC sputtering deposition performance

Deposition performance summary for reactive p-DC sputtering Wafer number N2 percentage (%) Deposition Temperature (°C) Measured thickness (nm) Deposition rate (nm/s) Main phase (XRD) Sheet resistance (Ohm/sq) wafer 1 5 400 - - β-V2Si 4.24 wafer 2 10 400 250 0.1388 mixture of β-V2Si and cubic VN 7.15 wafer 3 15 400 200.2 0.1112 cubic VN 6.63 wafer 4 20 400 167.2 0.0930 cubic VN 5.50 wafer 5 25 400 149.6 0.0831 cubic VN 4.83 wafer 6 5 600 - - β-V2Si 4.33 wafer 7 10 600 218 0.1211 mixture of β-V2Si and cubic VN 5.54 wafer 8 15 600 187.5 0.1042 cubic VN 6.39 wafer 9 20 600 161 0.089 cubic VN 6.40 wafer 10 25 600 144.5 0.080 cubic VN 6.31

The deposition rate decreases markedly as the reactive-gas fraction increases. At 400 °C it falls from 0.1388 nm/s for wafer 2 to 0.0831 nm/s for wafer 5, while at 600 °C it falls from 0.1211 nm/s for wafer 7 to approximately 0.080 nm/s for wafer 10. This behaviour is typical of a reactive-sputtering process in which formation of a nitride-rich target surface reduces the effective sputter yield. A contribution from composition-dependent resputtering at the growing film cannot be excluded.

The GiXRD assignments show the same general phase sequence at both temperatures: the 5% N₂ samples are assigned mainly to β-V₂Si, the 10% samples contain both β-V₂Si and cubic VN, and the 15–25% samples are dominated by cubic VN. These labels are based on diffraction-pattern matching and should not be interpreted as quantitative phase fractions.

Sheet resistance should be considered together with thickness. For example, a thinner film may have a similar sheet resistance while possessing a different resistivity. Using $\rho =R_{s}t$, the approximate resistivity decreases from about 179 to 72 µΩ·cm across wafers 2–5 and from about 121 to 91 µΩ·cm across wafers 7–10. These values are useful for comparison within this sample set, although contact geometry, thickness uncertainty and any parallel conduction paths should be considered.

Characterization

The following sections describe the structural, compositional and near-surface characterization. For clarity, each method is presented in the same order: measurement conditions, figures or fitted values, and a short interpretation of the observed trends.

X-Ray Reflectivity (XRR) analysis

XRR measurements were performed using a Rigaku SmartLab X-ray diffractometer. The Cu X-ray tube was operated at 40 kV and 30 mA. The incident optics included an IPS (incident parallel-slit) adapter with a 5° Soller slit. The slit configuration was IS = 0.05 mm, RS1 = 0.05 mm and RS2 = 0.075 mm. Scans were acquired with a step size of 0.01° and a counting time of 2 s per point.

XRR provides information from the critical-angle region and interference fringes. Their positions and damping were modelled to obtain film thickness, scattering-length-density-related mass density and interface/surface roughness. Fitting was carried out in SmartLab Studio. The model contained an effective Si-based substrate, a possible interfacial Si–metal-containing layer where required by the data, the deposited VN_x film and a thin, low-density surface overlayer attributed to oxidation and/or adsorbed moisture. The top layer is a phenomenological fit component and should not be interpreted as a chemically pure water layer.

• 
  Measured and fitted XRR curves for the VN_x series deposited at 600 °C. The fringe period changes systematically with film thickness.
• 
  Measured and fitted XRR curves for the VN_x series deposited at 400 °C. The increasingly wide fringe spacing is consistent with decreasing film thickness at higher N₂ fraction.

Extracted values

The values below are fit results and should be interpreted as model-dependent effective parameters. Thickness is generally the most robust XRR result. Density and roughness may be correlated with each other and with the assumed surface and interface layers, especially for comparatively thick or rough films.

XRR results for VNx layers deposited at 400 °C

Top layer (moisture) VNx Main layer Si wafer with 300 nm thermal oxide Deposition name VNx/Moisture thickness (nm) VNx/Moisture density (g/cm3) VNx/Moisture roughness (nm) VNx thickness (nm) VNx density (g/cm3) VNx roughness (nm) Si thickness (nm) Si density (g/cm3) Si roughness (nm) wafer 2 0.148 1.013 2.454 250.928 5.448 4.171 $\mathrm{\infty }$ 2.3360 0.141 wafer 3 0.525 0.854 2.922 199.401 5.555 4.992 $\mathrm{\infty }$ 2.3360 0.151 wafer 4 0.729 0.609 2.979 167.250 5.541 5.524 $\mathrm{\infty }$ 2.3360 0.225 wafer 5 1.002 0.850 2.633 149.104 5.613 5.008 $\mathrm{\infty }$ 2.3360 0.372

XRR results for VNx layers deposited at 600 °C

Top layer (moisture) VNx Main layer Si substrate Deposition name VNx/Moisture thickness (nm) VNx/Moisture density (g/cm3) VNx/Moisture roughness (nm) VNx thickness (nm) VNx density (g/cm3) VNx roughness (nm) Si thickness (nm) Si density (g/cm3) Si roughness (nm) wafer 7 1.106 1.184 2.988 218.010 5.705 4.106 $\mathrm{\infty }$ 2.3360 0.104 wafer 8 1.196 0.600 1.499 187.549 5.703 4.274 $\mathrm{\infty }$ 2.3360 0.107 wafer 9 0.327 0.639 1.477 161.297 5.938 3.900 $\mathrm{\infty }$ 2.3360 0.236 wafer 10 0.644 0.610 1.499 144.469 5.891 3.660 $\mathrm{\infty }$ 2.3360 0.526

The fitted VN_x thickness decreases monotonically with increasing N₂ fraction at both temperatures, confirming the deposition-rate trend in the performance table. For wafers 2–5, the fitted thickness falls from 250.928 to 149.104 nm; for wafers 7–10, it falls from 218.010 to 144.469 nm.

The fitted main-layer density is approximately 5.45–5.61 g/cm³ for the 400 °C series and 5.70–5.94 g/cm³ for the 600 °C series. Within the limitations of the model, the higher values at 600 °C suggest denser film growth. The fitted roughness of the main layer is approximately 4.17–5.52 nm at 400 °C and 3.66–4.27 nm at 600 °C, suggesting somewhat smoother layers at the higher temperature. These conclusions are comparative; they depend on the selected layer stack and should not be treated as independent chemical identification.

Grazing-incidence X-ray diffraction (GiXRD)

GiXRD measurements were performed on the VN_x-coated Si/SiO₂ wafers using a Rigaku SmartLab diffractometer with a Cu sealed X-ray tube operated at 40 kV and 30 mA. The scans covered a 2θ range of 30–80°, using a 0.5 mm incident slit, a step size of 0.01° and a scan speed of 2°/min. The incident angle was fixed at $\omega =2^{\circ }$ relative to the sample surface throughout the measurement. This shallow fixed-incidence geometry increases the relative contribution from the deposited film, although contributions from the substrate and interface may still be present. Peak positions were compared with reference patterns from the Crystallography Open Database (COD).

The measurements reveal the same general phase evolution at both deposition temperatures. At 5% N₂, the patterns are dominated by reflections tentatively assigned to β-V₂Si. At 10% N₂, reflections from both β-V₂Si and cubic VN are observed. When the N₂ fraction is increased to 15–25%, cubic VN becomes the dominant crystalline phase. Within the investigated parameter range, the reactive-gas fraction therefore has the clearest effect on the phase development, while increasing the deposition temperature from 400 to 600 °C does not change the overall phase sequence.

The phase assignments are qualitative. Film texture, overlapping reflections and possible substrate or interface contributions can influence the observed relative intensities; consequently, the GiXRD patterns should not be used to estimate quantitative phase fractions without further analysis.

Additional characterization and complete results

A complete GiXRD description, including the original diffraction patterns and detailed peak assignments, together with the full XPS composition analysis, REELS results and supporting figures, is available on the corresponding internal LabIntra page. For further information, access to the complete data set or a discussion of the results, please contact Evgeniy Shkondin, DTU Nanolab.