Electroless Nickel Plating of Magnesium Particles for Hydrogen Storage

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This work used a electroless method to synthesize Mg decorated with Ni for hydrogen storage applications. It is an interesting work and well structured work. However, several issues should be addressed before publication.

1. numerous typing errors are present and hence the whole manuscript should be double-checked before resubmission.
2. HRTEM are recommended to verify the phases of the as-synthesized catalysts.
3. The major disadvantage of Mg-based hydrides is the large thermodynamic and kinetic barrier, which results in large energy consumption for driving their hydrogen storage reaction. Hence, recent solar-driven hydrogen storage technique is developed (Nat. Commun., 2024, 15, 2815；Adv. Mater., 2023, 35, 2206946；Advanced Materials, 2020, 32, 2002647；Advanced Science, 2024, 11, 2400274; Journal of the American Chemical Society, 2025, 147, 3, 2786-2796), which should be noted in the introduction section and discussed in specific section.
4. The hydrogen desorption performance of MgH2@NiP is inferior to MgH2. Why is that?

Author Response

Please see the attachment. Thanks a lot.

 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The paper of Sindy Bello et al. presents an interesting experimental demonstration of hydrogen absorption in electroless Ni plated Mg commercial particles of about 26 micron-size . The authors report a significant improvement of the relevant sorption kinetics in regard to the untreated Mg particles, but this improvement is accompanied by a severe reduction of the hydrogen absorption capacity from 7.6 weight percent in pure Mg to 3.5 weight percent in the Ni-decorated Mg particles.

The authors do not report comprehensively some of their experimental conditions and do not represent the complete analysis of their results:

-The authors state in rows 199-201: “Hydrogen absorption tests were carried out under a hydrogen pressure of 2.1 MPa, while desorption tests were performed under a vacuum pressure of 4.5 × 10⁻⁷ MPa.” To what extent a constant pressure of 2.1 MPa has been maintained in the Sieverts system during the hydrogen absorption? To what extent a stable vacuum of 4.5 × 10⁻⁷ MPa has been maintained in the Sieverts system during the hydrogen desorption? If these values have been indeed kept absolutely constant, how the absorbed/desorbed quantities of hydrogen have been measured or calculated – by a flowmeter for example?

- The authors state in rows 201-202: “Hydrogen absorption/desorption values were calculated using the NIST (REFPROP)® database.” Which values exactly have been calculated using this database? Some more details may be helpful. In addition, a reference should be given for this data base as requested by its developers in https://www.nist.gov/srd/refprop.

- The statement in rows 202-204 “Activation energy (Ea) was estimated from kinetic curves at 325, 350, and 375 °C using the Johnson-Mehl-Avrami-Kolmogorov (JMAK) equation [31]” implies that ref. 31 gives some details and insights into the JMAK equation. Going over ref. 31 reveals that the JMAK equation is not even mentioned there. Such an attitude of citing references not in the stated context is unacceptable.   

- In rows 344-345 it is stated “Finally, Figure 11a presents the JMAK and Arrhenius plots for the MgH₂@NiB sample.”  The authors do not provide sufficient details of their analysis. Saving themselves those details may even lead to their own confusion -  see below. The authors do not provide the information of the JMAK equation used (α=1-exp(-Ktⁿ), where α is the reacted (desorbed) fraction of MgH₂, K depends on the particularities of the nucleation and growth mechanism, t is the reaction time and the exponent n is associated with the dimensionality of the reaction (desorption) process and the nucleation rate. K is probably not identical with k, which appears in Fig. 11a.). More specifically, what is the exponent n used in their fitting procedure? May this information contribute some insight into the desorption mechanism?

The units of k in Fig. 11a are not indicated. The y axis of the inset in Fig. 11a is again denoted lnk as in the main figure, which is probably not correct. Each line in the inset is assigned a specific temperature, T, (325+273, 350+273, 375+273 K) and yet the x axis is 1000/T. This of course does not make sense. (This is the confusion, mentioned above.)

- The authors present DSC measurement for MgH2@NiB in Fig. 7b for three heating rates. The authors can probably derive from these measurements the apparent activation energy for desorption utilizing the Kissinger equation, and then compare its consistency with the corresponding E_a value obtained from the kinetic analysis.

- The authors state in rows 390-393 “Catalyzing micrometric magnesium particles through the anhydrous Ni-B electroless coating enables a reduction in the onset temperature for hydrogen release under a pressure of 4.5 × 10⁻⁷ MPa (TPD) from 441 °C to 240.6 °C (a reduction of 200.4 °C).” First, it is not clear where the value 441°C  is taken from. It does not appear elsewhere in the text, nor a reference is given for it. Second, the onset temperature of 240.6 °C is related to a minor hydrogen desorption (see Fig. 7a) of probably some secondary phase, i.d. Mg₂NiH₄, reported to be present in the decorated particles (Figures 9 and 10). The onset temperature, associated with the major hydrogen release, is about 350^oC (Fig.7a). The results of this research should not be “decorated”.

- As already mentioned above, the cycled hydrogen capacity of the Mg@NiB decorated particles is reported to be about 3.5 wt.%. It is worthwhile to cite other relatively simple procedures of obtaining Mg-Ni alloys and their cycled hydrogen capacity. For example, Mg-16.3Ni alloy, prepared by induction melting, followed by a simple mechanical treatment, has demonstrated a cycled hydrogen capacity of about 4.5 wt.%  (J.O. Fadonougbo et al., Kinetics and thermodynamics of near eutectic Mg-Mg2Ni composites produced by casting process, Int. J. Hydrogen Energy, 45 (2020) 29009-2922).

 

The points, raised above, should be seriously considered and referred in the manuscript.

Some technical remarks follow:

- rows 33,34: ± sign is missing in 94.56±4.43 kJ/mol.

- rows 44,45: To put a full stop after “market” and start the next sentence with “Although”. To write “primary” instead of “main”: Hydrogen may become a main source of energy, but it will still be a secondary, and not a primary source of energy. To omit “is a carrier that”.

- rows 57, 58: To write Mg instead of MgH₂. It is Mg, not MgH₂, that is low-cost and naturally abundant.

- row 208: “100 to 400 °C” is not a “heating rate”. To correct this.; To write “to support” instead of “yo support”.

- row 265: To write “hydrogen absorption/desorption isothermal kinetic curves”, as stated in Fig. 5, instead of “hydrogen absorption/desorption isotherms”. Isotherms alone usually mean something else, i.e. pressure-composition isotherms.

- rows 265-269: The decorated Mg particles should be labelled (Mg@NiP), (Mg@NiB),instead of (MgH₂@NiP), (MgH₂@NiB). This applies also to other parts in the text. Also, it should be written “In 90 minutes Mg absorbs 3.2 wt.% of hydrogen” and not “In 90 minutes MgH₂ absorbs 3.2 wt.% of hydrogen”. It is the Mg that absorbs hydrogen, and not MgH₂.

- row 273: To write “absorption” instead of “abortion”.

- row 281: To write “isothermal absorption/desorption kinetic curves”, as stated in Fig. 6, instead of “absorption/desorption isotherms”.

- row 298: To correct “204.6 ^oC” to “240.6 ^oC” – see row 393. Also, the main hydrogen release occurs at onset temperature of about 350 ^oC.

- rows 350, 351, 353: The sentences “This represents a 106% improvement in desorption kinetics compared to uncoated micrometric commercial magnesium particles. Regarding the reversible hydrogen absorption capacity, the specimen maintains a reversible capacity of 3.5 wt.% ± 0.09 wt.%, showing a 9.37% improvement in hydrogen absorption kinetics.” are not clear and should be clarified.

- row 354: To write “an improvement” instead of “and improvement”.

 

Author Response

Please see the attachment. Thanks a lot.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All my comments have been well revised and this work could be published in the present form.

Author Response

We appreciate your comments and suggestions. Thank you.

Reviewer 2 Report

Comments and Suggestions for Authors

 The authors have added additional required information to their manuscript, but this brings into light some fundamental major problems in the revised manuscript of Sindy Bello et al. that require clarifications and significant changes:

1. The authors state in rows 202-204: “Hydrogen absorption tests were carried out under an initial hydrogen pressure of 2.1 MPa, while desorption tests were performed under an initial vacuum pressure of 4.5 × 10⁻⁷ MPa.” The addition of “initial”, with regard to the original manuscript is useful. Still, the final pressures of the sorption experiments could be indicated. It is merely informative (not mandatory) for the absorption kinetic curves as they have not been analyzed. It is though mandatory for the desorption kinetic curves, which have been analyzed, the analysis being used for the derivation of the activation energy. A very large change of the dynamic desorption pressure may cause a significant variation of the driving hydrogen desorbing force, with regard to the equilibrium hydrogen pressure at the given temperature. Such a variation may be problematic for performing a significant and correct fit of the kinetic curves. It may be deduced from the right part of Fig. 4 that the rise of the hydrogen pressure is from 4.5×10⁻⁷ MPa to 0.2-0.3 MPa. Additional complication is due to the different driving desorption forces at the different temperatures as different hydrogen equilibrium pressures characterize the corresponding temperatures. Also, the derivation of the activation energy in this work, based on the JMAK equation, is problematic -see below.

 

2. The equation given in row 215, 𝑙𝑛(−𝑙𝑛(1−∝))=𝑛𝑙𝑛(𝑘)+𝑛𝑙𝑛(𝑡), is not correct. Starting with the familiar form of the JMAK equation, α=1-exp(-ktⁿ), one readily obtains 𝑙𝑛(−𝑙𝑛(1−∝))=𝑙𝑛(𝑘)+𝑛𝑙𝑛(𝑡), i.e. without n before lnk. Then the activation energy, Ea, estimated from kinetic curves at 325, 350, and 375 °C in Figure 11a and using the calculating scheme in the manuscript, will be completely different from the one given there (94.5 kJ/mole). The authors seemingly assume that the free terms of the fitted line equations in the inset of Fig. 11a are nlnk. Consequently, lnk values, appearing in the main part of Fig. 11a, have been erroneously obtained by dividing the free terms by n. (By the way, the equations for the lines at 325^oC and 375^oC in the inset of Fig. 11a have been interchanged, but this does not affect the erroneous calculations.) If lnk values are correctly taken to be the free terms in the line equations and plotted in Fig. 11a, one can roughly estimate E_a to be 140 kJ/mole, significantly different from 94.5 kJ/mole. Still, another serious problem is that this calculating scheme, adopted in the manuscript, is not correct in principle. The k value, appearing in the JMAK equation, is not in general the rate constant for the hydrogen sorption process if nucleation and growth mechanism is assumed. The rate constant, k*, should be derived by finding a function f(α) that satisfies an equation of the type f(α)=k*t. Some f(α) functions for different rate-determining mechanisms may be found in ref.58 ((J.O. Fadonougbo et al., Kinetics and thermodynamics of near eutectic Mg-Mg2Ni composites produced by casting process, Int. J. Hydrogen Energy, 45 (2929) 29009-2922) and in J.D. Hancock and J.H. Sharp, Method of Comparing Solid-State Kinetic Data and Its Applications to the Decomposition of Kaolonite, Brucite and BaCO3, J. American Ceramic Society, 55 (1972) 74-77. It may be also useful and even a must to examine the possibility of a diffusion-controlled mechanism for the dehydrogenation of the spherical decorated Mg particles, i.e. to try to fit a kinetic diffusion-controlled equation to the experimental kinetic curves.

- It may be noted in passing that the equation in row 210, 𝑙𝑛(𝑘)=−𝐸_𝑎/𝑅𝑇+𝑙𝑛(𝑘₀), is the Arrhenius equation and not the Avrami equation, as stated in rows 207, 208.

3. It seems that there is a basic misinterpretation and misunderstanding of the temperature programmed desorption (TPD), presented in Fig. 7a, and discussed in rows 314-317,409-412. This misunderstanding has not been removed despite relevant remarks in my original reviewing report. The authors state in rows 314-317: “The onset of hydrogen release occurs at 273.7°C, with two endothermic peaks suggesting the presence of multiple desorption phases, associated with the decomposition of Mg₂NiH₄  (240.6°C) and MgH₂ (273.7°C), respectively,…” There is indeed a temperature onset at 240.6°C which is quite certainly associated with the release  of hydrogen from Mg₂NiH₄, but there is no temperature onset at 273.7°C. At this temperature the TPD curve flattens off. Differentiating the TPD curve would show that 273.7°C is the end of the first endothermic peak (or more correctly endothermic dip). In other words, 273.7°C indicates most certainly the temperature of total hydrogen release from Mg₂NiH₄. The temperature onset of hydrogen release from MgH₂ is roughly at 340-350^oC, and complete hydrogen release is not achieved until 400^oC, the end temperature of the TPD curve in Fig. 7a.  Consequently, the statement in rows 409-412 “Catalyzing micrometric magnesium particles through the anhydrous Ni-B electroless coating enables a reduction in the onset temperature for hydrogen release under a pressure of 4.5×10⁻⁷ MPa (TPD) from 441 °C to 273.7 °C (a decrease of 167.3 °C) [53].” is not correct.  In addition, it is still not clear where the value 441°C  is taken from. Ref. 53 states that the onset temperature for the release of hydrogen from MgH₂ is 302^o C. The authors are strongly advised to hydrogenate their original non-decorated Mg spherical particles and to measure their temperature onset during hydrogen desorption.

 

Technical remarks are not given here in view of the major problems raised above. These problems should be resolved before a further consideration to publish this manuscript.

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

 First, the authors have indeed used a modified form, α=1−exp(−kt)n, of the original JMAK equation, α=1−exp(−k’tn). This is achieved by reforming the original JMAK equation, and not by regrouping it. In that case kn=k’, and the advantage is that f(α)=kt (see my previous report). In the case of nucleation and growth f(α)= [ln(1/(1−α))]1/n.  
That said, fundamental major problems still persist in the second revised manuscript of Sindy Bello et al.

1. It is now stated in the manuscript that during the kinetic desorption experiments the hydrogen pressure rises from 4.5×10⁻⁷ MPa to about 0.17 MPa. The H2 pressure thus changes many orders of magnitude during the dehydrogenation of MgH2 with a concomitant significant approach to the relevant hydrogen equilibrium pressures, p. The latter may be estimated by utilizing ∆H=−74.06 kJ/(mol H2) and ∆S=−133.4 J/(K×mol H2) (Paskevicius, M., Sheppard, D.A. and Buckley, C.E. ‘Thermodynamic changes in mechanochemically synthesized magnesium hydride nanoparticles, J. American Chemical Society, 132 (2010) 5077–5083, doi:10.1021/ja908398u) and the equation RTln(p/po)=∆H−T∆S. The results for the equilibrium pressures at 325, 350 and 375 oC are approximately 0.32, 0.57 and 1 MPa, respectively. The kinetic desorption curves, obtained in this work, may give some practical information. Their fits, provided the mentioned very significant change of the dynamic pressure, may be though problematic or fortuitous for deriving relevant rate constants and subsequently activation energy.

2. The rate constants, k(T), derived in this work, correspond to slightly different nucleation and growth mechanisms (different n values) for the different temperatures, i.e. n=1.208, 0.929 and 0.82 at 325, 350 and 375oC, respectively  (Fig. 11a). In my opinion, a significant comparison of rate constants at different temperatures for the same type of sample should be done for the same desorption mechanism. In other words, one cannot compare rate constants if the desorption mechanism changes (even if others have done so). Going to extremes, it is clear that rate constants cannot be compared for completely different kinetic mechanisms. It is sometimes possible to fit completely different kinetic mechanisms to the same kinetic curves. Without going into details, this point could have been taken care of.

3. Even if one follows the calculating scheme of this work, there is a basic error in the estimation of the activation energy: It is stated that kinetic desorption curves have been performed at 325, 350 and 375 oC ( row 208 and Fig. 11a). These temperatures in Kelvin are 598, 623 and 648K, respectively. The corresponding 1000/T values are 1.67, 1.61 and 1.54 K−1,i.e. a 1000/T span of 0.13 K−1. The 1000/T span, plotted in Fig.11a, is approximately 0.22 K−1 - between 1.6 and 1.82 K−1. This wrong 1000/T span yields activation energy of ~94.6 kJ/mol, as given in the manuscript. Provided the linear fit still holds for the correct 1000/T span (I have not checked that), the result for the activation energy is ~160 kJ/mol. This is completely different from the reported 94.6 kJ/mol, and this relatively low activation energy is one of the central claims of the work.   
 
Despite the significant experimental effort invested in this work, it is recommended to reject publishing the manuscript in view of the above points.
 

Author Response

Dear Reviewer

We really appreciate the reviewer's comments and suggestions, and we have carefully revised the manuscript accordingly. The required changes have been incorporated and highlighted in red font in the corrected version of the manuscript.

R./ The text has been modified according to the reviewer's comments.

 

Author Response File: Author Response.pdf