Synthesis of piperidines and pyridine from furfural over a surface single-atom alloy Ru1CoNP catalyst

The sustainable production of value-added N-heterocycles from available biomass allows to reduce the reliance on fossil resources and creates possibilities for economically and ecologically improved synthesis of fine and bulk chemicals. Herein, we present a unique Ru1CoNP/HAP surface single-atom alloy (SSAA) catalyst, which enables a new type of transformation from the bio-based platform chemical furfural to give N-heterocyclic piperidine. In the presence of NH3 and H2, the desired product is formed under mild conditions with a yield up to 93%. Kinetic studies show that the formation of piperidine proceeds via a series of reaction steps. Initially, in this cascade process, furfural amination to furfurylamine takes place, followed by hydrogenation to tetrahydrofurfurylamine (THFAM) and then ring rearrangement to piperidine. DFT calculations suggest that the Ru1CoNP SSAA structure facilitates the direct ring opening of THFAM resulting in 5-amino-1-pentanol which is quickly converted to piperidine. The value of the presented catalytic strategy is highlighted by the synthesis of an actual drug, alkylated piperidines, and pyridine.

Editorial Note: This manuscript has been previously reviewed at another journal that is not operating a transparent peer review scheme.This document only contains reviewer comments and rebuttal letters for versions considered at Nature Communications.

REVIEWER COMMENTS
Reviewer #1 (Remarks to the Author): The current form of manuscript is substantially revised and updated by addressing the mentioned suggestions.Though, I am not fully convinced, in fact, surprised that why authors are avoiding even the basic characterization (even the PXRD of Ru1Co80/HAP and Ru1Co180/HAP is not provided) of the mentioned catalysts, which is essentially important to affix the primary form of the material (here catalyst).As author emphasized, I would also like to understand why it is a negative to have sufficiently isolated Ru atoms, particularly when Ru1Co80/HAP (4x) and Ru1Co180/HAP (9x) displayed several folds higher activity than Ru1Co20/HAP ?After convincingly addressing the above queries this manuscript may be considered for publication.
Reviewer #2 (Remarks to the Author): The authors have adequately responded to all of my previous concerns.I recommend the paper for publication.
Reviewer #3 (Remarks to the Author): The authors have addressed most of my comments.Before it can be accepted, I still have one question/suggestion: The characterization of Ru1Co20/HAP shows that few Ru-Ru exists in it.Based on the different loadings (Ru1Co40, Ru1Co80, Ru1Co160), the author concluded that the surface single-atom alloy Ru1CoNP Catalyst plays a key role in this tandem reaction.However, no characterization results can prove that Ru1Co40, Ru1Co80, and Ru1Co160 are single-atom alloy catalysts.Is there any way (such as CO-FTIR) to prove this though it is difficult as mentioned by the authors?If not, it would be better to rethink the conclusion as no strong evidence.

Responses to the referees
Reviewer #1 (Remarks to the Author): 1.The current form of manuscript is substantially revised and updated by addressing the mentioned suggestions.
Response: Thanks for your positive comment.
2. Though, I am not fully convinced, in fact, surprised that why authors are avoiding even the basic characterization (even the PXRD of Ru1Co80/HAP and Ru1Co180/HAP is not provided) of the mentioned catalysts, which is essentially important to affix the primary form of the material (here catalyst).
Response: Previously, in order to obtain a direct evidence of single-atom Ru species, we have already done the basic characterizations of PXRD (Supplementary Fig. 3 in SI), CO-DRIFTs (Supplementary Fig. 6 in SI), aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (AC-HAADF-STEM, Fig. R1a) and in-situ H2-reduction environmental TEM (ETEM, Fig. R1b) for Ru1Co20/HAP sample, but no obvious Ru atomic signal were identified.So, we thought it would be also difficult to obtain effective structural information even characterizing Ru1Coy/HAP (y=40, 80, and 160) materials with lower Ru content, that's the reason why we didn't further perform the basic characterizations for these low-Ru-loading catalysts.In fact, according to reviewer's suggestion for last version of manuscript, we have already performed the PXRD characterization for Ru1Co40/HAP catalyst (see in Supplementary Fig. 3 in SI), indeed neither Ru signal nor other effective information can be obtained.Fortunately, at this stage we have obtained the reliable XAS data for the Ru1Co40/HAP and Ru1Co80/HAP samples (Fig. R2 and Table R1), which proves the single-atom Ru dispersion.In order to further prove our conclusion "a unique surface single-atom alloy (SAA) structure enables the one-pot formation of piperidine from furfural", the Ru1Co40/HAP and Ru1Co80/HAP catalysts with relatively lower Ru content were chosen for X-ray absorption spectra (XAS) characterizations.This time, a long-time XAS signal acquisition was applied for trying to obtain high-quality XAS data.As shown in Fig. R2a-b, the obvious different k-space patterns of the Ru1Co40/HAP and Ru1Co80/HAP with that of Ru foil as well as only one peak at ~1.98 Å (not phase-corrected) in r-space of these two catalysts, suggest the formation of sole Ru-Co coordination, which is also confirmed by wavelet transforms (Fig. R2c).The best-fitted EXAFS results of the Ru1Co40/HAP and Ru1Co80/HAP samples reveal Ru-Co shell at 2.46 Å with coordination number (CN) of 5.3 and Ru-Co shell at 2.48 Å with CN of 5.7, respectively (Fig. R2e-f and Table R1).The absence of Ru-Ru shell undoubtably indicates the single-Ru-atom dispersion on Co nanoparticles in the Ru1Co40/HAP and Ru1Co80/HAP catalysts.
(Note: High-energy XAS experiment, e.g., for Ru XAS characterization, is usually open during July-August and November-December in Shanghai Synchrotron Radiation Facility (SSRF) every year.Therefore, we can not guarantee the XAS measurement for Ru element at any time of the year.Last year, we performed the XAS characterizations for the Ru1Co20/HAP and Ru1Co40/HAP catalysts, but only a reliable signal of Ru1Co20/HAP sample was obtained.According to reviewers' suggestions, in July this year, again we tried the XAS characterizations for the low-Ru-loading Ru1Co40/HAP and Ru1Co80/HAP samples, but a long-time XAS signal acquisition was applied.
Fortunately, moderate-quality XAS data of both Ru1Co40/HAP and Ru1Co80/HAP samples were obtained this time, seen in Fig. R2).The data range used for data fitting in k-space (∆k) and R-space (∆R) are 3.0-11.3Å −1 and 1.0-2.6Å, respectively.

3.
As author emphasized, I would also like to understand why it is a negative to have sufficiently isolated Ru atoms, particularly when Ru1Co80/HAP (4x) and Ru1Co180/HAP (9x) displayed several folds higher activity than Ru1Co20/HAP?

Response:
We are sorry for the misunderstanding of the negativity for having sufficiently isolated Ru atoms, here the Ru content in Ru1Co40/HAP (0.21 wt% Ru), Ru1Co80/HAP (0.10 wt% Ru), and Ru1Co160/HAP (0.05 wt% Ru, see Table 1 in manuscript) decreases gradually but the Co loading remains constant (5 wt% Co).
Because of the complicated two-step tandem catalysis (100 °C for 6 h: reductive amination of furfural, and 180 °C for 14 h: hydrogenative ring arrangement of furfurylamine to piperidine), so we calculated the overall piperidine production rate based on total Ru content and whole reaction time: overall piperidine production rate ( 2 ×  −1 ×ℎ −1 ) = produced piperidine (mol) Ru loading (mol) ´20 (h) (already been supplemented in the manuscript), and just wanted to show that only at very dilute concentrations of Ru piperidine formation occurred.So, we can not conclude that it is negative to have sufficiently isolated Ru atoms based on the piperidine production rate.