南京大写知名学者论文受关注

Glutathione-Mediated Biomimetic NO Activation with Coordination Capsules for NH ₃ and α-Amino Acid Electrosynthesis

Journal of the American Chemical Society (2026) - 4 Commentsdoi: 10.1021/jacs.6c01915  pubmed: 42200426  issn: 0002-7863  issn: 1520-5126 

---

Yao Wang , Meng Lv , Xu Jing , Jinfeng Wang , Chunying Duan 

#1Liburnascincus coensis comment accepted June 2026

Duan et al. reported “Glutathione-Mediated Biomimetic NO Activation with Coordination Capsules for NH3 or a-Amino Acid Electrosynthesis” (JACS 2026, https://doi.org/10.1021/jacs.6c01915).1.The chronoamperometric (i-t) curves for 0.08 mM H1 (Fig. S43) and 0.01 mM H1 (Fig. S59) are identical at the same applied potential, with the same current densities curve. Does this indicate that varying the concentration of H1 has no effect on the reaction rate?2.The mass spectrum in Figures S2 and S4 shows an abrupt, unnatural drop in baseline intensity around m/z = 2300 and 2350, creating a "cliff-like" discontinuity in the signal. This artifact is highly unusual in standard high-resolution mass spectrometry.3.The authors report the NORR electrode effective area as 0.25 cm2 and a reaction time of 2 h. Taking the optimal catalytic data of 182.16 umol·h-1·cm-2 as an example, the total NH3 produced is 91.08 umol (182.16×0.25×2). The amount of catalyst H1 is 2.1 umol (30 mL× 0.07 mmol/L). Based on these values, the calculated TON is 43.4 (91.08/2.1) and the TOF is 21.7 h-1 (43.4/2). However, the authors report a TOF of ~90 h-1 in Figure 2h and a TON of ~70 in Figure S26, which are inconsistent with these basic calculations. Similar calculation errors appear throughout the work, raising serious concerns about the credibility of the reported catalytic data.4.The control experiment in Figure S27, intended to demonstrate the low reactivity of the intermediate NH2OH, suffers from a lack of controlled variables. While the main NORR reaction is performed in a DMSO/H2O (v/v=3:2) mixture, the NH2OH reduction study uses a different solvent system: CH3CN/H2O (v/v=3:1). Changing both the organic solvent and the solvent-to-water ratio prevents a valid comparison, as any observed difference in reactivity cannot be attributed solely to the substrate, but may instead arise from altered solvation, mass transport, or interfacial properties in the new medium.5.Calculations via the Stokes-Einstein equation using standard solvent viscosities yield hydrodynamic diameters from the stated DOSY diffusion coefficients for the identical crystalline sample: H1 (Fe) DMSO-d6: D=2.41×10-10 m2/s hydrodynamic diameters d=0.91nm H2 (Co) CD3CN: D=3.55×10-10 m2/s hydrodynamic diameters d=3.51nm The values differ by approximately 3.9-fold. For the mixed solvent: H2 (Co) CD3CN/DMSO-d6 (v:v=4:1): D=3.55×10-10 m2/s hydrodynamic diameters d=1.81nm Why does the same crystal show three vastly different hydrodynamic diameters? And the DOSY results also conflict with dimensions derived from single-crystal X-ray diffraction.6.Fig. S12 Within the chemical shift range of 2-4 ppm, identical diffusion coefficients are observed across different solvents. This contradicts the fundamental rule that varying solvent viscosities should lead to different diffusion coefficients, which is clearly evidenced in other DOSY data presented by the authors. This concern is validated by Fig. S16, where 1a-NO and H2 exhibit different diffusion coefficients.7.Fig. S69 In the DOSY spectrum, the signals assigned to free 1b are incomplete. The aromatic proton Hb and the carboxylic acid proton of 1b are absent. For free 1b, all protons should exhibit the same diffusion coefficient.8.Fig.60 We note several critical inconsistencies in the reported ATP inhibition experiments. Dissolving 30 mmol of ATP (16.53 g) in the DMSO/H2O (v/v = 3:2) reaction system appears practically unfeasible, which far exceeds the solubility limit in this medium. Even in kinetic inhibition experiments based on 0.07 mM H1 catalyst, adding ATP at the claimed concentrations (>400-fold excess over H1) would require ATP quantities vastly exceeding typical solubility limits in this mixed solvent. It is evident that ATP exerts an extremely weak inhibitory effect on H1, how can the host-guest binding constant between H1 and ATP reported by the authors be on the order of 10^5? And the ¹H NMR spectra also exhibit highly unusual behavior. No chemical shift changes are observed for the internal standard or the NH₄⁺ product even at high ATP concentrations, and the signals of the H1 catalyst disappear entirely under these conditions. This deviates significantly from the authors’ other reported kinetic NMR data (Fig.S45, S47 and S49, supposing these data are accurate). 9.Figs. S97 and S98 The post-reaction ¹H NMR spectrum shows identical splitting patterns and impurity peaks to the pure product standard, even in the presence of high concentrations of substrate, supporting electrolyte, and solvent. The absence of any baseline distortion, peak broadening, or residual signals from reaction components (e.g., starting materials, salts, or solvent-related species) in the reaction mixture spectrum is highly unusual for a 6-hour reaction in this medium.

#2Passiflora crenata comment accepted June 2026

“Glutathione-Mediated Biomimetic NO Activation with Coordination Capsules for NH3 or a-Amino Acid Electrosynthesis” (Journal of the American Chemical Society, 2026, doi.org/10.1021/jacs.6c01915). 1.The authors report the geometric surface area of the NORR electrode as 0.25 cm2. However, the voltammetric responses for identical concentrations of H1 and equivalents of 1a-NO in Figure 2d and Figure S25 exhibit drastically different currents (0.04 mA vs 0.01 mA). As evidenced by the inconsistent current densities observed under analogous conditions: Figure S24 shows a current density close to -0.06 mA/cm2 at 0.4 mM 1a-NO, while Figure S25 only shows -0.04 mA/cm2.2.The LSV curves in Figure 2f show current values inconsistent with the corresponding plot in its inset, suggesting potential unit confusion between mA/cm2 and mA. More critically, while Figure 2f displays increasing current with higher 1a-NO concentrations, Figure S41 exhibits the opposite trend under analogous conditions. Even at the same concentration of 0.12 M 1a-NO, the reported current densities differ significantly between the two datasets. And the authors report the NORR electrode geometric area as 0.25 cm². In Figure 2f, the LSV curve for 0.05 M 1a-NO at -0.1 V vs. RHE shows a current density below 6 mA/cm2, corresponding to a total current of less than 1.5 mA. However, Figure S35 reports a current greater than 3 mA at the same potential under standard reaction conditions. These values are mutually inconsistent, and similar contradictions are also present in other i-t datasets.3.In Figure 2d, the reduction potential of uncomplexed 1a-NO is shown to be at -0.6 VNHE (~0 VRHE). In the NORR reaction, with 0.05 M 1a-NO and 0.07 mM H1, even if all the catalyst were to bind the substrate, only 0.07 mM of 1a-NO would be complexed. This means the vast majority of 1a-NO remains free in solution. Applying a potential of -0.3 VRHE should reduce this abundant free 1a-NO. However, the control reaction without H1 in Figure 2f shows no NH3 product, which is inconsistent with the electrochemical behavior observed.4.The authors report electrocatalytic data that contain errors and mutual contradictions. For instance, in Figure 2g at −0.1 VRHE, the NH3 yield is ~100 μmol·h⁻¹·cm⁻² with a Faradaic efficiency of ~58%. Given the electrode area of 0.25 cm2 and reaction time of 7200 s, the total charge is calculated to be Q = (100×0.25×2×96485×10^-6) / 58% ≈ 41.6 C, corresponding to an average current I = Q/t ≈ 6 mA. However, the current shown in Figure S35 is only 3 mA, which is clearly inconsistent. If calculated using the i-t curve provided by the authors, the Faradaic efficiency would exceed 100%.5.The authors present numerous mutually contradictory kinetic datasets for the NORR reaction. For example, in Figure S55, the NH3 production rate with 0.002 M 1a-NO exceeds 10 μmol/h. However, Figure S57 reports a rate below 15 μmol/L under the same 0.002 M 1a-NO condition. Even accounting for potential unit errors, converting the latter value to a rate using the given volume (30 mL) and reaction time (2 h) yields a result that is inconsistent with Figure S55.6.Instances where figures and their captions are inconsistent, such as the mismatch observed in Figure S40, appear repeatedly throughout the paper. Furthermore, the standard electrocatalytic testing conditions are not uniform across the work. And there are different between the conditions described in the main text and those in the supporting information for identical catalytic reactions.7.The kinetic data provided by the authors lacks controll variables. For the H1 concentration- dependent kinetics in Figure S42, the standard condition specifies 0.05 M 1a-NO, yet the authors altered the 1a-NO concentration while varying H1. Conversely, for the 1a-NO concentration-dependent kinetics in Figure S50, the standard condition calls for 0.07 mM H1, but the authors changed the H1 concentration while varying 1a-NO. It is unclear why such a dual-variable test was used.8.The reported yields are inconsistent with the corresponding i-t curves, leading to Faradaic efficiencies exceeding 100%. For example, in Figures S42 and S43, at an H1 concentration of 0.08 mM, the reported NH3 yield is ~140 μmol·h-1·cm-2. With an electrode area of 0.25 cm2 and a reaction time of 2 h, this corresponds to a total NH3 production of 70 μmol. The charge required to produce this amount of NH3 is QNH₃= 70*10^-6 mol * 5 * 96485 C/mol =33.8 C. However, the corresponding i-t curve indicates a maximum total charge of only Q < 4 mA/cm2 * 0.25 cm2 * 7200s = 7.2C. Back-calculating the Faradaic efficiency from these values gives FE = 33.8/7.2 =469% > 100%. Even if the unit is written incorrectly, FE is still greater than 100%. In Figure S43, the system contains 0.02 M 1a-NO and 0.08 mM H1 with a measured current density of around 3 mA/cm2, translating to an overall current of 3 * 0.25 mA = 0.75 mA, less than 1 mA. Yet Figure S42 reports an NH₃ yield reaching approximately 150 μmol·h-1·cm-2 under matching conditions. 9.How the Raman signals assigned to NH2OH and NH3 can be detected under open-circuit potential in Figure S66. The authors attribute the sharp peak at 840 cm-1 to quartz window background vibration, yet this feature turns into a broad hump with potential-dependent intensity variation in Figure S88. In addition, two inherent background peaks located at 1200–1300 cm-1 in Figure S66 completely disappear in Figure S88. Furthermore, no noticeable peak evolution can be observed for the assigned bands at 1048 cm-1 and 1583 cm-1 in Figure S88, which conflicts sharply with the high product yield of C–N coupling reported in this work.

#3Gaudichaudia galeottiana comment accepted June 2026

“Glutathione-Mediated Biomimetic NO Activation with Coordination Capsules for NH3 or a-Amino Acid Electrosynthesis” (JACS 2026, doi:10.1021/jacs.6c01915).

1. Under standard reaction conditions in Figure S35 (0.05 M 1a-NO, 0.07 mM H1, -0.3 V vs. RHE), the total measured current exceeds 7 mA on the 0.25 cm2 working electrode, translating to a current density above 28 mA·cm-2. In comparison, Figure S43 records a current density lower than 3.5 mA·cm-2 at the same potential, with 0.02 M 1a-NO and 0.08 mM H1. This comparison indicates that the concentration of free 1a-NO primarily governs the reduction current. As shown in Figure 2d, uncomplexed 1a-NO has a reduction potential of -0.6 V vs. NHE (approximately 0 V vs. RHE). Given the large concentration gap between 0.05 M 1a-NO and 0.07 mM H1, the vast majority of 1a-NO remains unbound in the solution. Consequently, the applied potential of -0.3 V vs. RHE is sufficient to drive the reduction of abundant free 1a-NO, meaning the catalytic effect of H1 is negligible in this system.

1. Figure 3 b), if the reaction current remained stable during both the 4-7 h and 15-18 h periods, how can the NH₃ yield produced in the same 3-hour interval be drastically higher during 15-18 h, while the Faradaic efficiency simultaneously decreases? Conversely, a drop in Faradaic efficiency would normally lead to a lower yield. The reported combination of a large increase in NH3 yield and a simultaneous decrease in Faradaic efficiency under a stable current is inconsistent with basic electrochemical principles. Does this suggest the current was not actually stable, or that the reported yield and efficiency values are inconsistent?

1. In both Figure S12 and Figure S69, all signals falling within the 2~4 ppm chemical shift range are reported to share the exact same diffusion coefficient, even though these peaks originate from different species or solvents. How can all these signals yield identical diffusion coefficients across different species and solvents?

4.In Figure 4 d), the LSV trace for H1 with both 1b and 0.05 M 1a-NO shows a lower current density than the trace in Figure 2f with 0.05 M 1a-NO alone. This indicates that the addition of 1b leads to displacement of 1a-NO from the H1 cage. The ESI-MS data in Figure 4 b) further confirms the formation of a dominant 1b⸦H1 complex. This competitive binding would be expected to reduce the amount of 1a-NO available for catalysis, thereby lowering the reaction yield. The reported high product yields in the presence of 1b appear inconsistent with this competitive binding effect.

#4Scleria baroni-clarkei comment accepted June 2026

“Glutathione-Mediated Biomimetic NO Activation with Coordination Capsules for NH3 or α-Amino Acid Electrosynthesis” (J. Am. Chem. Soc. 2026, 148, 22, 22407–22412).

1. The fluorescence titration curve in Fig. S71 is inconsistent with the reported 1:1 host-guest binding model. With H1 at 0.01 mM, the titration plateaus at only ~3 μM of 1b (0.3 equivalents), far before the 1:1 stoichiometric point. How can such early saturation be explained by a 1:1 host–guest binding model? The ITC data presented in Fig. S10 further supports that the actual binding stoichiometry is less than 1:1. And no clear isosbestic points can be observed from the UV-vis titration results in Fig. S70 either.