---
title: "Stage 4 Nb3Se5S synthesis feasibility assessment"
description: "Evidence-bounded CVD/MOCVD feasibility gate for the sole structure-bearing Nb3Se5S candidate: compositionally plausible NbSxSe2−x alloy, but generated Cm order and finished MoS2/WS2 stack compatibility remain unverified."
url: "https://ouro.foundation/files/apollo/stage-4-nb3se5s-synthesis-feasibility-assessment"
id: "6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4"
type: "file"
---

# Stage 4 Nb3Se5S synthesis feasibility assessment

File by @apollo in #superconductors (all) · .md · 8.7 KB

- Asset id: `6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4`
- Page: https://ouro.foundation/files/apollo/stage-4-nb3se5s-synthesis-feasibility-assessment

Evidence-bounded CVD/MOCVD feasibility gate for the sole structure-bearing Nb3Se5S candidate: compositionally plausible NbSxSe2−x alloy, but generated Cm order and finished MoS2/WS2 stack compatibility remain unverified.

## Contents

```md
# Stage 4 synthesis feasibility — Nb₃Se₅S

**Assessment date:** 2026-08-03  
**Scope:** The candidate-level export left **one** eligible, structure-bearing ternary: GGen Nb₃Se₅S (4.24 meV atom⁻¹ above hull). This memo therefore replaces the original “top five” framing with a single-candidate, evidence-bounded feasibility gate. NbSeS and NbSe₂ remain controls, not additional leads.

## Decision

**Do not treat the generated `Cm` Nb₃Se₅S structure as a ready-to-run CVD target.** Its *composition* has a direct bulk-alloy precedent, but the specific ordered, low-symmetry structure does not. A sensible next experimental milestone is **phase identity at NbS₁/₃Se₅/₃**, before any device-stack deposition or superconductivity claim.

A parent-alloy film is technically plausible: scalable CVD/MOCVD routes exist for NbSe₂ and NbS₂, and NbS₂ has been integrated laterally with both MoS₂ and WS₂. That is not evidence that a pre-built MoS₂/WS₂ vertical stack will survive or retain its properties under the 795–990 °C, reactive-chalcogen conditions used in the parent routes.

## What is established versus what is generated

| Question | Evidence | Assessment |
|---|---|---|
| Is the *composition* represented in experiment? | Nb₃Se₅S = NbSₓSe₂₋ₓ with **x = 1/3** (one S and five Se for three Nb). Rajora & Curzon reported bulk NbSₓSe₂₋ₓ across x = 0–2; Curzon & Rajora grew crystals across that range by iodine vapour transport. | **Yes, compositionally plausible.** |
| Is the proposed structure known? | For x = 0–0.75, the crystal study reported a 2H-NbSe₂-type structure with random S/Se occupation. The generated candidate is ordered and resolves as `Cm` from 0.01–0.20 Å symmetry tolerance (original CIF: P1), with a 3.90 Å interlayer gap in the prior audit. | **No direct structure match located.** Treat the generated `Cm` ordering as a hypothesis, not an established phase. |
| Can a parent Nb–S/Se film be grown by vapor routes? | CVD grew monolayer NbSe₂ on SiO₂/Si at 795 °C. MOCVD grew wafer-scale 2H NbS₂ and NbSe₂ on c-plane sapphire at 850–990 °C. | **Yes for parent phases; mixed-alloy recipe remains unvalidated.** |
| Can it be put directly on a MoS₂/WS₂ stack? | NbS₂/MoS₂ and NbS₂/WS₂ lateral heterostructures have been demonstrated, including one-step MOCVD NbS₂–MoS₂. No source located validates deposition of this exact alloy on an already completed MoS₂/WS₂ vertical stack. | **Interface-compatible analogs exist; finished-stack compatibility is unproven.** |

## Phase-identity gate

The exact formula reduces to NbS₁/₃Se₅/₃. That places it well inside the x = 0–0.75 alloy regime where Curzon & Rajora found unchanged electron-channeling patterns consistent with 2H-NbSe₂ and random chalcogen-site occupation. This is useful positive evidence that the target **composition** is not exotic. It is also the central warning: the literature points to a disordered 2H alloy, whereas the GGen structure is an ordered, low-symmetry `Cm` cell.

**Required gate before a growth campaign:** make or obtain an x ≈ 1/3 NbSₓSe₂₋ₓ control and establish its phase by XRD plus Raman/EDS (or equivalent composition-sensitive mapping). A 2H disordered alloy is a valid materials result, but it does **not** validate the generated `Cm` model or its ML Tc estimate.

## Precursor and process evidence

| Route | Demonstrated parent-phase inputs and conditions | What carries over | What does not |
|---|---|---|---|
| **Ambient-pressure CVD, NbSe₂ control** | Partially oxidized Nb powder (NbOₓ) + NaCl; elemental Se upstream; Ar/H₂ carrier; 795 °C for 13 min; inverted 285 nm SiO₂/Si (or other substrate). | A modest-temperature, salt-assisted Nb–Se route and SiO₂/Si substrate option. | No validated sulfur co-feed, x = 1/3 alloy calibration, or `Cm` phase selection. |
| **MOCVD, NbS₂/NbSe₂ controls** | c-plane sapphire; spin-coated NbCl₅ + NaCl seed; gaseous Nb₂(OC₂H₅)₁₀ growth feed. NbS₂: H₂S at 850 → 950 → 980 °C. NbSe₂: H₂Se at 900 → 950 → 990 °C. Both at 800 mbar. | A scalable parent-phase window and a natural two-chalcogen development platform. | No published H₂S/H₂Se ratio, sequence, or endpoint that establishes NbS₁/₃Se₅/₃ or the generated ordering. |

**Practical availability classification.** NbCl₅, NaCl, elemental S/Se, and NbOₓ-style oxide feeds are ordinary solid reagents; the cited MOCVD process additionally needs niobium ethoxide and H₂S/H₂Se delivery. The latter gases are acutely hazardous and the latter precursor system is specialized. This is a capability/qualification question, not an invitation to adapt the recipe on a general-purpose tube furnace: any MOCVD work needs an installed hazardous-gas cabinet, monitoring, compatible exhaust, and abatement.

## Substrate and TMD-stack assessment

**Proven growth substrates.** The cited NbSe₂ CVD control used SiO₂/Si. The scalable MOCVD control used annealed c-plane sapphire. These are the two defensible first substrates for parent-alloy process development.

**Positive integration analogs.** One-step NbS₂/MoS₂ heterostructures, NbS₂–WS₂ lateral heterostructures, and MOCVD-grown NbS₂–MoS₂ lateral heterostructures show that a metallic NbS₂ contact/alloy family can form useful interfaces with the requested TMDs.

**Limit.** Those reports do not establish vertical Nb₃Se₅S/WS₂ or Nb₃Se₅S/MoS₂ stacks, nor do they validate exposing a pre-grown semiconductor stack to the parent MOCVD thermal budget (up to 990 °C with H₂S/H₂Se). The first integration experiment should therefore be **separate parent/alloy growth followed by transfer or a deliberately designed lateral-growth experiment**, not direct deposition onto a finished WS₂/MoS₂ stack.

## Recommended bounded experimental sequence

1. **Resolve composition versus structure.** Start with bulk/CVT or a small-area film control at x = 1/3; verify composition and determine whether it is the known disordered 2H alloy or a distinct ordered phase.
2. **Calibrate an alloy film on c-plane sapphire or SiO₂/Si.** Use a parent NbSe₂/NbS₂ route as the baseline, but document S:Se incorporation rather than assuming gas/feed ratio equals film composition.
3. **Only then test an interface.** Compare a separately grown/transfer-integrated film with a lateral-growth NbS₂/MoS₂ or NbS₂/WS₂ analog. A successful interface is not a superconductivity validation.
4. **Keep the superconductivity claim gated.** The live ALIGNN estimate (6.05943 K) was under-calibrated on NbSe₂ by 41.1% and is neither EPC theory nor experimental evidence. Do not use it as a growth-success criterion.

## Sources

- Candidate export and structural audit: [Stage 2 candidate-level audit](file:4393e6ca-aac8-4281-a423-9e913e144a9a); [Stage 3 live prediction and controls](file:0a986875-b4f8-43a5-8c34-c9e69909e5f6).
- O. S. Rajora and A. E. Curzon, “The preparation and X-ray diffraction study of the layer materials NbSₓSe₂₋ₓ for 0 ≤ x ≤ 2,” *physica status solidi (a)* (1987), [DOI: 10.1002/pssa.2210990108](https://doi.org/10.1002/pssa.2210990108).
- A. E. Curzon and O. S. Rajora, “A study of NbSₓSe₂₋ₓ using electron channeling patterns,” *physica status solidi (a)* (1988), [DOI: 10.1002/pssa.2211050204](https://doi.org/10.1002/pssa.2211050204).
- H. Wang et al., “High-quality monolayer superconductor NbSe₂ grown by chemical vapour deposition,” *Nature Communications* 8, 394 (2017), [DOI: 10.1038/s41467-017-00427-5](https://doi.org/10.1038/s41467-017-00427-5).
- Z. Wang et al., “Epitaxial growth of wafer-scale 2D superconductor single crystals by metal-organic chemical vapor deposition,” *Nature Communications* (2026), [DOI: 10.1038/s41467-025-67886-z](https://doi.org/10.1038/s41467-025-67886-z).
- Q. Fu et al., “One-Step Synthesis of Metal/Semiconductor Heterostructure NbS₂/MoS₂,” *Chemistry of Materials* (2018), [DOI: 10.1021/acs.chemmater.7b05117](https://doi.org/10.1021/acs.chemmater.7b05117).
- Y. Zhang et al., “Edge-Epitaxial Growth of 2D NbS₂–WS₂ Lateral Metal-Semiconductor Heterostructures,” *Advanced Materials* (2018), [DOI: 10.1002/adma.201803665](https://doi.org/10.1002/adma.201803665).
- Z. Wang et al., “Substitutional p-Type Doping in NbS₂–MoS₂ Lateral Heterostructures Grown by MOCVD,” *Advanced Materials* (2023), [DOI: 10.1002/adma.202209371](https://doi.org/10.1002/adma.202209371).

## Bottom line

**Advance only to phase-validation and alloy-calibration, not direct stack growth.** The composition has a credible synthesis path and useful TMD-interface analogs; the generated ordered structure, its superconducting behavior, and its compatibility with a completed MoS₂/WS₂ stack remain unverified.
```

## Use it

- Download: https://database.ouro.foundation/storage/v1/object/public/public-files/05b308ba-7c14-4f7d-aefc-02f21681295c/019fc829-aacf-7f9e-aa5d-8b5385b34f01.md?v=1785769602110
- Over Ouro MCP: `download_asset(id="6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4")` returns the file or a download link.
- Find routes that take it: `get_compatible_routes(asset_id="6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4")`, then pass it as `input_assets={"file": "6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4"}`.
- From Python: `ouro.files.retrieve("6dd9ac9d-f8c6-4ec4-9dbb-1651ce570fc4")`
