# Catalyst claim check

**Source:** Hongzhou Yang, Lu Shang, Qinghua Zhang, Run Shi, Geoffrey I. N. Waterhouse, Lin Gu, and Tierui Zhang (2019), *A universal ligand mediated method for large scale synthesis of transition metal single atom catalysts*, *Nature Communications* 10, 4585. DOI: [10.1038/s41467-019-12510-0](https://doi.org/10.1038/s41467-019-12510-0).

**Scope:** Source check of the supplied main article and Supplementary Information only; not an independent experimental replication.

| Claim / question | Evidence and exact location | Conditions | Assessment / limits |
|---|---|---|---|
| Ni single-atom catalyst achieves 98.9% CO Faradaic efficiency | Main PDF p. 1 (abstract) states “98.9% Faradaic efficiency at −1.2 V.” Main PDF p. 6, text beside **Fig. 6b**, identifies the metric as **Faradaic efficiency to CO (FECO)** and reports the maximum of 98.9% at −1.2 V. **Fig. 6b** plots FECO versus potential. Main PDF p. 7 and **Fig. 6e** specify that the 98.9% value belongs to **Ni-SAC-2.5** (2.5 wt.% Ni); the summary repeats this. | −1.2 V vs RHE; CO₂-saturated aqueous 0.1 M KHCO₃; sealed three-electrode H-type cell. Working electrode: catalyst ink on glassy carbon; Ag/AgCl reference and Pt-wire counter electrode. Potential held 30 min for FE measurement; gaseous products measured by GC. | **Supported as a reported result.** The precise claim is 98.9% **FECO** for Ni-SAC-2.5 under these conditions. It is not a general efficiency value across all potentials or Ni loadings. |
| Selectivity range | Main PDF p. 6, text and **Fig. 6b**, report FECO above 90% from −0.7 to −1.5 V vs RHE. | Same electrolyte/cell context as above. | Supports broad potential-window selectivity, but the exact 98.9% peak occurs only at −1.2 V. |
| Competing gaseous product | Main PDF p. 6 cites **Supplementary Fig. 51** for low H₂ Faradaic efficiency. Supplement PDF p. 47 caption: “Faradaic efficiency of H₂ for Ni-SAC-2.5 at different potentials.” | Ni-SAC-2.5; potentials are plotted in the supplied figure. | Caption is readable and the figure is supplied, but numerical point values were not recoverable from the retrieved text. It supports low H₂ qualitatively; it does not independently restate the 98.9% FECO value. |
| Liquid products | Main PDF p. 6 cites **Supplementary Fig. 52**. Supplement PDF p. 48 caption states that no liquid products were detected by NMR after CO₂RR. | Post-CO₂RR liquid-phase NMR; further acquisition parameters are not stated in the retrieved caption. | Supports absence of detected liquid products within the reported NMR method/detection capability; it does not prove absolute absence below detection limits. |
| Peak selectivity versus stability potential | Main PDF p. 6, **Fig. 6b**, gives peak FECO at **−1.2 V vs RHE**. The 20 h current-density stability experiment is **Fig. 6c at −0.8 V vs RHE**, explicitly stated in the text and caption. Main PDF pp. 6–7 and **Fig. 6d** report nearly identical pre/post Ni K-edge XANES spectra. | Peak-selectivity FE test: −1.2 V; durability: −0.8 V for 20 h. | **The stability experiment was not run at the peak-selectivity potential.** Therefore it directly supports 20 h current retention and structural similarity at −0.8 V, not 20 h durability of the 98.9% result at −1.2 V. “Very slight decrease” is qualitative in the text; no formal degradation rate or replicated uncertainty is reported in the cited passage. |
| Ni-loading dependence | Main PDF p. 7 and **Fig. 6e** report FECO at −1.2 V of 98.9%, 98.5%, 98.0%, and 95.3% for Ni loadings of 2.5, 3.4, 4.5, and 5.3 wt.%, respectively. | −1.2 V vs RHE. | Confirms that 98.9% is the optimum reported loading-specific value, not the value for the nominal 5.3 wt.% material. |
| Large-scale catalyst comparison | Main PDF p. 7 cites **Supplementary Fig. 53**. Supplement PDF p. 49 says the CO₂RR performance of material synthesized at 1.6 kg scale was very similar to that made at 70 mg scale. | 70 mg versus 1.6 kg synthesis scale; exact plotted potentials/values are not recoverable from the retrieved text. | Figure and caption are supplied and readable, but numerical trace values are not text-readable here. This supports comparability/scalability qualitatively, not the specific 98.9% claim unless the plotted data are manually digitized. |
| Faradaic efficiency versus energy efficiency and conversion | The paper defines/reports product **Faradaic efficiency** through charge allocation to CO and H₂, with GC product detection. No energy-efficiency metric or CO₂ conversion fraction is reported in the cited main-text section, Fig. 6, or Supplementary Figs. 51–53. | FE test conducted for 30 min at fixed potential; no feed/outlet carbon balance or energy-efficiency calculation is provided in the retrieved evidence. | **Do not relabel 98.9% as energy efficiency or CO₂ conversion.** FECO is the fraction of measured charge assigned to forming CO; it does not state the fraction of inlet CO₂ converted, nor electrical-to-chemical energy efficiency. |

## Bottom line

The attached sources support the narrowly framed statement that **Ni-SAC-2.5 reportedly reached 98.9% Faradaic efficiency to CO at −1.2 V vs RHE** in CO₂-saturated 0.1 M KHCO₃ in a sealed H-type cell. The claim appears in the abstract (main PDF p. 1), the results and Fig. 6b (p. 6), and the loading comparison/summary and Fig. 6e (p. 7). It must not be interpreted as 98.9% energy efficiency or CO₂ conversion. The durability evidence used a less negative potential, −0.8 V vs RHE for 20 h, so it does not establish 20 h stability at the −1.2 V peak-selectivity condition.

All cited supplementary figures (51–53) are present in the supplied supplement and have readable captions. Their plotted numerical values are not recoverable from the retrieved text; where relevant, this is marked above. No cited supplementary item was missing.