In-situ Quantification of Nanoparticles Oxidation: A Fixed Energy X-ray Absorption Approach

Round 1

Reviewer 1 Report

Per the suggestions of the reviewers, the author has made significant improvement in the whole manuscript, therefore, a publication is suggested.

Author Response

Thank you for the comment

Reviewer 2 Report

After read the author’s corrections this reviewer considers that this paper is not suitable for its publication as far as the major issues were not addressed. Note that the comments provided by this review are in good agreement with comments of reviewer #1.

1) Changes in the background signal

The signal intensity in the acquisition of XAS spectra is always normalized against the incoming photon flux This is a common practice, however the details of the normalization process are now given in the experimental section.

This assumption doesn’t answer my question. In the FEXRAV measurement, how can the authors be sure that there is no change in the pre and post peak than can affect (changes in the background) the absolute intensity recorded during continuous the scan upon different polarization? In addition, any change in the hybridization of Pd 4d with the O 2p can influence the change in the peak intensity without modify the oxidation state i.e. by change the material structure.

2) Dissolution

Dissolution happens for sure in this case. However, considering the actual scan rate of the experiment and the diffusion properties of the dissolved hydroxipalladate, we can assume that the the diffusion region is well within 3 mm from the electrode surface (calculated considering a 2800 s time and 1e-9 m2/s diffusion coefficient, both cautelative estimation). The attenuation length of the PdK radiation in the electrolyte and in the carbon materials exceeds 1 cm. For this reason, we can assume that the fluorescence signal samples both the material in the electrode and the dissolved one. For sure for longer time at the dissolution potential could drive some dissolved Pd to quit the volume sampled by the fluorescence signal, however this is not the case of our experiment. Additionally, considering that the size of the electrode is much larger than the Nernst diffusion layer created in our experiment, we can assume that diffusion is planar.

Do you have any experimental evidence that support this extreme? EQCM measurements can help you to address this issue.

3) Bulk oxidation

We are not sure we have caught the meaning of the question. However, working with small monodispersed nanoparticles like the ones we have in this study, the technique is sensitive to the surface. In these conditions the depth sensitivity is not far from that of the X ray Photoelectron Spectroscopy, with the clear advantage of a much simpler way to realize in-situ experiments.

I’m expert in XPS spectroscopy and this affirmation is simply false based on the photoelectron scattering in solid. The sensitivity of photoelectron spectroscopy (depending in KE) is below 1 nm meanwhile in this measurements the sensitive is tens of nanometers (according to figure 1, TEM measurements). The information provided by the XAS spectra is mostly bulk sensitivity being the surface contribution unknown in this specific case.

This is just a different approach. Other work on the application of the same technique have been published by other authors. The effectiveness of FEXRAV in investigating electrocatalysts has been investigated and an evolution toward a quantitative approach is, in the opinion of the authors, worthy of further research. Moreover, this procedure can be in principle implemented in even in Lab XAFS spectrometer based on conventional X-ray tubes, and in beamline where no availability of QuickXAS is present.

From the point of view of this reviewer this technique provides no valuable information because the changes in the intensity can be ascribed to many different issues not only to variation in the oxidation state. Therefore, it is necessary to provide the whole spectrum and compare it to relevant references changes in the intensity cannot provide any valuable information about the oxidation state due to many different process involves in the variation of the intensity.

In the study we follow the evolution of the catalyst under the application of the electrochemical stimulus. Our focus is on the oxidation of palladium; therefore, we believe that the technique is an “Operando” method.

This is an in situ method as far as not information related to the species evolving from the reaction are provided by any on line analytic technique (in this specific case, products of the ethanol oxidation). If the authors want to use this nomenclature an online product analysis should be added to the experiments to collect simultaneously the variation in the electronic structure and products of the reaction.

We don’t understand the reviewer objection. For sure the relative entity of the oxidation depends on the surface to volume ratio of the particles and depends on the particle size distribution. What we are demonstrating here is that given a catalyst of which we know the particle distribution, we can determine the relative amount of Pd (II) that is formed during the experiment.

My concerns are in good agreement with reviewer #1. The oxidation state cannot be ascribed just to variation in the peak intensity. In this specific case the surface and the bulk of the catalysts (several tens of nanometers for your nanoparticles according to the TEM) can be potentially quite different. In addition, can the authors provide any experimental evidence that the hybridization of Pd 4d with the O 2p plays not any role in the overall change (i.e. O K-edge)?

A TEM image has been added.

Thanks!

In XAS no such signal onset is defined, because the “onset” just refers to the density of states of the electrons near to the Fermi level. In cyclic voltammetry, it may depend on the scan rate, however this typically happens for scan rates much higher than the one here adopted.

Can the authors provide information related to the ethanol oxidation over-potential?

The actual potential window is selected in such a way that a little superposition with the oxygen evolution region happens. The cell configuration allows an easy removal of possible bubble as the electrode surface is orthogonal to the floor plan. A short discussion on this has been added to the paper.

Thanks for add this information to the text.

Author Response

1) Changes in the background signal

The signal intensity in the acquisition of XAS spectra is always normalized against the incoming photon flux This is a common practice, however the details of the normalization process are now given in the experimental section.

This assumption doesn’t answer my question. In the FEXRAV measurement, how can the authors be sure that there is no change in the pre and post peak than can affect (changes in the background) the absolute intensity recorded during continuous the scan upon different polarization? In addition, any change in the hybridization of Pd 4d with the O 2p can influence the change in the peak intensity without modify the oxidation state i.e. by change the material structure.

There are no physical reasons (but the changes in the chemical speciation of Pd) to which a change in the “background” can be addressed. Any kind of drift of the system (due to changes in the I₀ intensity and focus, and to eventual energy drifts) were tested and excluded as source of uncertainty in the measurements. Calibration procedures were repeated after all the executed FEXRAV measurements. The data presented in the study are normalized, i.e. acquisition of pre-edge and post edge reference intensities (these latter far from the XANES region) were also acquired.

Concerning the chemical source of changes in the intensity, we want here to stress some specific points that we consider valuable in our text:

1)   Any kind of change in the “real” electronic structure and configuration of the Pd-O system is in principle the aim of our investigation, with an unnecessary link to the “formal” oxidation state. We clearly affirm that we are describing the system using a mixture of elemental Pd and PdO, in order to account the observed changes in the FEXRAV pattern. We don’t want to affirm that all the Pd(II) in the system is PdO. The large uncertainty in the determination of the fraction of Pd(II), declared by our study, accounts also for the eventual changes in the spectral intensity arising from different Pd(II) oxide/hydroxide/solvated/dissolved species.

2)   The measurements carried out at the half edge energy value are not affected by other factors but actual redox state of Pd: these measurements fully confirm the other measurements carried out at 24370 eV.

3)   The fact that at certain polarization values part of the Pd (the surface fraction) is in its divalent state is not a matter of discussion, being already fully established in numerous studies in the literature (see answer to the following point).

This is supported by the wide lliteriture In Pd XAS and and Pd electrooxidation. It is well known that under mild anodic conditions Pd oxidizes to Pd(II) and that Pd(II) oxides and hydroxides can go in the solution but still having the same XAS feature of solid oxides and hydroxides . All the following references are already present in the draft:

FOR SIMILARITY OF XAS FEATURES

Kim, Y.; Kim, J.; Kim, D.H. Investigation on the enhanced catalytic activity of a Ni-promoted Pd/C catalyst for formic acid dehydrogenation: Effects of preparation methods and Ni/Pd ratios. RSC Adv. 2018, 8, 2441–2448,     

Waser, J.; Levy, H.A.; Peterson, S.W. The structure of PdO. Acta Crystallogr. 1953, 6, 661–663,     

Moore, W.J.; Pauling, L. The Crystal Structures of the Tetragonal Monoxides of Lead, Tin, Palladium, and Platinum. J. Am. Chem. Soc. 1941, 63, 1392–1394,    

GREENWOOD, N.N.; EARNSHAW, A. Chemistry of the Elements; Elsevier, 1997; ISBN 9780750633659,      

Torapava, N.; Elding, L.I.; Mändar, H.; Roosalu, K.; Persson, I. Structures of polynuclear complexes of palladium(ii) and platinum(ii) formed by slow hydrolysis in acidic aqueous solution. Dalt. Trans. 2013, 42, 7755–7760).,

FOR Pd CHEMISTRY/DISSOLUTION UNDER ELECTROCHEMICAL STIMULUS       

Grdeń, M.; Łukaszewski, M.; Jerkiewicz, G.; Czerwiński, A. Electrochemical behaviour of palladium electrode: Oxidation, electrodissolution and ionic adsorption. Electrochim. Acta 2008, 53, 7583–7598,         

Wang, L.; Lavacchi, A.; Bellini, M.; D’Acapito, F.; Benedetto, F. Di; Innocenti, M.; Miller, H.A.; Montegrossi, G.; Zafferoni, C.; Vizza, F. Deactivation of Palladium Electrocatalysts for Alcohols Oxidation in Basic Electrolytes. Electrochim. Acta 2015.                  

Montegrossi, G.; Giaccherini, A.; Berretti, E.; Di Benedetto, F.; Innocenti, M.; D’Acapito, F.; Lavacchi, A. Computational speciation models: A tool for the interpretation of spectroelectrochemistry for catalytic layers under operative conditions. J. Electrochem. Soc. 2017, 164, 3690–3695.

2) Dissolution

Dissolution happens for sure in this case. However, considering the actual scan rate of the experiment and the diffusion properties of the dissolved hydroxipalladate, we can assume that the the diffusion region is well within 3 mm from the electrode surface (calculated considering a 2800 s time and 1e-9 m2/s diffusion coefficient, both cautelative estimation). The attenuation length of the PdK radiation in the electrolyte and in the carbon materials exceeds 1 cm. For this reason, we can assume that the fluorescence signal samples both the material in the electrode and the dissolved one. For sure for longer time at the dissolution potential could drive some dissolved Pd to quit the volume sampled by the fluorescence signal, however this is not the case of our experiment. Additionally, considering that the size of the electrode is much larger than the Nernst diffusion layer created in our experiment, we can assume that diffusion is planar.

Do you have any experimental evidence that support this extreme? EQCM measurements can help you to address this issue.

About the occurrence of Pd electrodissolution there are several authors who already addressed this issue. One example, among others, of the Pd electrodissolution under alkaline and oxidating conditions is reported in this paper Journal of Electroanalytical Chemistry 380 (1995) 127-138. The extent of the electrodissolution process is well reckoned by Grden et al. by means of EQCM data reported in this paper (already cited in our manuscript): Grdeń, M.; Łukaszewski, M.; Jerkiewicz, G.; Czerwiński, A. Electrochemical behaviour of palladium electrode: Oxidation, electrodissolution and ionic adsorption. Electrochim. Acta 2008, 53, 7583–7598. Some of us recently showed that the dissolution is expected from thermodynamic calculation in the exact same conditions of our experiment (Journal of The Electrochemical Society 164(11):E3690-E3695 2017 that is also cited in the paper). Palladium dissolution under alkaline conditions is even reported in classical textbook on palladium chemistry (Pd Palladium: Palladium Compounds - Gmelin Handbook of Inorganic and Organometallic Chemistry - 8th edition 8th ed. 1989 by William Griffith). This data are so convincing that we did not expected any deviation for our Pd from those reported in literature. On this ground, we presented this manuscript whose scope is beyond what already done on Pd electrodissolution by the authors we cited. Eventually, about the diffusion of Pd species in the electrolyte, we do not see how to exploit an EQCM setup to clarify this point. Indeed, since hydroxipalladiates are generated by electrodissolution at electrode surface we do not see why we shouldn’t expect a diffusion following well know physical laws.

3) We are not sure we have caught the meaning of the question. However, working with small monodispersed nanoparticles like the ones we have in this study, the technique is sensitive to the surface. In these conditions the depth sensitivity is not far from that of the X ray Photoelectron Spectroscopy, with the clear advantage of a much simpler way to realize in-situ experiments.

I’m expert in XPS spectroscopy and this affirmation is simply false based on the photoelectron scattering in solid. The sensitivity of photoelectron spectroscopy (depending in KE) is below 1 nm meanwhile in this measurements the sensitive is tens of nanometers (according to figure 1, TEM measurements). The information provided by the XAS spectra is mostly bulk sensitivity being the surface contribution unknown in this specific case.

Figure 1 clearly shows that the particles of the catalyst are well below 5 nm in diameter. The “tens of nm” features, are the carbon structure that act as a conductive high surface area support for the catalyst. It is the high surface to volume ratio of the Pd particles that make the technique sensitive to the oxidation of the first layer from the surface. The advantage of this method over XPS it is the capability to probe the oxidation in-situ, something that cannot be obtained with XPS. We have added the indication of the Pd particles in the TEM images, along with a clear description of this in the text.

4) This is just a different approach. Other work on the application of the same technique have been published by other authors. The effectiveness of FEXRAV in investigating electrocatalysts has been investigated and an evolution toward a quantitative approach is, in the opinion of the authors, worthy of further research. Moreover, this procedure can be in principle implemented in even in Lab XAFS spectrometer based on conventional X-ray tubes, and in beamline where no availability of QuickXAS is present.

From the point of view of this reviewer this technique provides no valuable information because the changes in the intensity can be ascribed to many different issues not only to variation in the oxidation state. Therefore, it is necessary to provide the whole spectrum and compare it to relevant references changes in the intensity cannot provide any valuable information about the oxidation state due to many different process involves in the variation of the intensity.

See the answer to point 1.

5)  In the study we follow the evolution of the catalyst under the application of the electrochemical stimulus. Our focus is on the oxidation of palladium; therefore, we believe that the technique is an “Operando” method.

This is an in situ method as far as not information related to the species evolving from the reaction are provided by any on line analytic technique (in this specific case, products of the ethanol oxidation). If the authors want to use this nomenclature an online product analysis should be added to the experiments to collect simultaneously the variation in the electronic structure and products of the reaction.

We have removed any reference to the operando methods.

6)  We don’t understand the reviewer objection. For sure the relative entity of the oxidation depends on the surface to volume ratio of the particles and depends on the particle size distribution. What we are demonstrating here is that given a catalyst of which we know the particle distribution, we can determine the relative amount of Pd (II) that is formed during the experiment.

My concerns are in good agreement with reviewer #1. The oxidation state cannot be ascribed just to variation in the peak intensity. In this specific case the surface and the bulk of the catalysts (several tens of nanometers for your nanoparticles according to the TEM) can be potentially quite different. In addition, can the authors provide any experimental evidence that the hybridization of Pd 4d with the O 2p plays not any role in the overall change (i.e. O K-edge)?

Concerning the TEM, see the answer to point 3. Concerning the chemistry of Pd the indications reported in the answer to point 1 still hold. The key point here is that all the Pd species that we can obtain share the same XAS features as shown by the characteristics of the first coordination shells in Pd oxides, hydroxides and acquocomplexes. (see references in point 1)

Reviewer 3 Report

Berretti et al. reported a new approach to quantify Pd(II)/Pd ratio while applying oxidative potential. This approach is very similar to other XAS methods used in electrocatalysis and catalysis to determine the oxidation state and coordination number, but unlike those methods a fixed energy is used here while swiping the potential. The method is very well described, all details are carefully explained. I can recommend this manuscript for publication after the authors consider the following minor comments:

In authors' response to the reviewers, not all the responses are satisfactory. While many of the previous reviewers' comments and suggestions are constructive and considering them will improve the quality of the paper, for some of them, the authors have just argued without any change in the discussion. I would urge the authors to re-consider the previous comments again and add to the discussion. 

Although in the introduction the importance of the oxide/metal ratio in determining the stability and activity of Pd catalyst for DEFC is discussed, there is no direct correlation between these two parameters. The authors have referred to previous studies that hypothesized this oxidation ratio is the main reason for degradation of the catalyst and consequently its activity. However, this hypothesis needs to be proved and confirmed. The authors need to provide such a correlation.

In catalysis and electrocatalysis the surface of the particle is the determining factor for activity and selectivity, and the bulk of the particle is of less importance. The reviewer is wondering why the bulk oxidation (which in addition to the surface oxidation is quantified here) should matter in determining the catalyst activity or selectivity? Pd and its oxides seem very stable, and perhaps ex-situ XPS measurements should give us enough information about the oxidation state of the surface. The reviewer understand that Pd in DEFC is just a case study in this manuscript, but the case study should chosen in a way that could convince the community that the proposed method is important, it gives additional information and has application. 

Author Response

1)   In authors' response to the reviewers, not all the responses are satisfactory. While many of the previous reviewers' comments and suggestions are constructive and considering them will improve the quality of the paper, for some of them, the authors have just argued without any change in the discussion. I would urge the authors to re-consider the previous comments again and add to the discussion. 

The application of this technique to the chosen system was made possible by an extensive bibliography on Pd and Pd species electrochemical behaviour in alkaline environment. This previous knowledge, together with the use of small nanoparticles to maximize surface to bulk signal, enabled us to propose this case-study. References have been added to the text, as suggested also by the previous review. Some sentences have been rephrased a/o extended  We believe that now, thanks to all the changes proposed by reviewers, text has reached consistency.

2)  Although in the introduction the importance of the oxide/metal ratio in determining the stability and activity of Pd catalyst for DEFC is discussed, there is no direct correlation between these two parameters. The authors have referred to previous studies that hypothesized this oxidation ratio is the main reason for degradation of the catalyst and consequently its activity. However, this hypothesis needs to be proved and confirmed. The authors need to provide such a correlation.

Experimental evidence on the correlation between palladium deactivation due to the formation of Pd(II) species was already present in the text (10.1016/j.electacta.2008.10.034) A new sentence, stressing this phenomenon has been added to the text. From this reference, it appears clear that palladium activity towards ethanol electrooxidation is present only when Pd is in its metallic state.

3)   In catalysis and electrocatalysis the surface of the particle is the determining factor for activity and selectivity, and the bulk of the particle is of less importance. The reviewer is wondering why the bulk oxidation (which in addition to the surface oxidation is quantified here) should matter in determining the catalyst activity or selectivity? Pd and its oxides seem very stable, and perhaps ex-situ XPS measurements should give us enough information about the oxidation state of the surface. The reviewer understand that Pd in DEFC is just a case study in this manuscript, but the case study should chosen in a way that could convince the community that the proposed method is important, it gives additional information and has application. 

For this reason, we choose particles of less than 5 nm. It is the high surface to volume ration of the Pd particles that make the technique sensitive to the oxidation of the first layer from the surface. The advantage of this method over XPS it is the capability to probe the oxidation in-situ, something that cannot be obtained with XPS. A brief note on that has been added to the text.

Reviewer 4 Report

The authors conducted in situ XANES analysis using Pd in ethanol containing alkaline media. To reveal reaction condition using in situ analysis is important and i believe in situ XANES can be a important tool to understand. However, current result only shows certain cases which cannot be generalized and compared. For that reason, I strongly recommend the author to conduct same experiment without ethanol and compare the oxidation of palladium feature is originated from the ethanol.

Author Response

1)   The authors conducted in situ XANES analysis using Pd in ethanol containing alkaline media. To reveal reaction condition using in situ analysis is important and i believe in situ XANES can be a important tool to understand. However, current result only shows certain cases which cannot be generalized and compared. For that reason, I strongly recommend the author to conduct same experiment without ethanol and compare the oxidation of palladium feature is originated from the ethanol.

In this paper, Pd oxidation in alkaline ethanol electrolyte is presented as a case-study for FEXRAV quantification. Experimental results obtained by the same FEXRAV technique to study Pd oxidation in bare alkaline media were already presented in our previous article, already cited in the text as [31]:

Montegrossi, G.; Giaccherini, A.; Berretti, E.; Di Benedetto, F.; Innocenti, M.; D’Acapito, F.; Lavacchi, A. Computational speciation models: A tool for the interpretation of spectroelectrochemistry for catalytic layers under operative conditions. J. Electrochem. Soc. 2017, 164, 3690–3695.

Round 2

Reviewer 2 Report

The resubmitted manuscript quality was improved significantly and now I can recommend this manuscript for its publication in this journal.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

In this work, Alessandro et al aimed to show that the utilization of the XAFS that is collected at certain energy can work as an indicator for the oxidation states of Pd catalysts. The motivation is clear and the conjunction work on electrochemistry and X-ray spectroscopy is also rewarding. However, significant concerns have aroused in terms of the basis for such assumption. This manuscript requires significant correction with more supportive data, since its current form is quite misleading for future readers.

Major issue:

1# It is always of great controversy to determine the oxidation states based on the peak intensity at certain energy in Pd K-edge XANES. However, the author made such assumption without any supporting reference or theories. They made such assumption based on the linear combination fitting, regardless of the obvious spectra shifts before the edges. What is worse,  it is worth noticing that the linear combination fitting is only effective when the phases of material exist between two standards. I do not believe that the electrochemically oxidized Pd(OH)x species could be close to PdO. For instance, the XANES and EXAFS of various FeOOHs differs from any FeOx (many papers on this fields). So, how could the author claim that PdO can be used a eligible and standard reference for linear combination fitting?

2# According to the author ‘Although the main Pd(II) species investigated by FEXRAV is most probably Pd(OH)42- (due to the dissolution process already mentioned) the preparation of an eligible standard solution is extremely difficult and could affect the reproducibility of the measurements. Still, PdO and Pd(OH)42- share the same coordination number and the same ligand  species [19].’

However, in fact, no experimental or XAFS evidence support from reference 19 can support author’s claim on Pd(OH)42- and PdO. The author should read reference and literature carefully before they cite these references. This is regarding with comment 1# also, if the author can not identify the structure of Pd(OH)42- and compare with PdO, the base of this paper is then ruined.

3# In Figure 6, it seemed that the Pd species would get oxidized with positive scans and then get reduced at opposite scans, with a nearly same intensities at the beginning and ending points. It indicated the electrochemical oxidation of Pd to Pd(OH)x, and reduction of Pd(OH)x to Pd…To the best of the reviewer’s knowledge, it does not make any sense in aqueous system. It seems that the so-called ‘Fixed Energy X-ray Absorption Approach’ is not corrected used and interpreted here…

Minor issues:

1# Too little discussion on XAS is covered in the introduction part. Several important works on the use of XAFS in the similar fields, electrocatalysis, should be cited accordingly: Nat. Commun. 8(1): 957.; Small 14 (15), 1704319-1704325; J. Am. Chem. Soc., 2015, 137 (3),1305–1313.

2# In Figure 2. It is good to show the standard XRD pattern of Pd metals and compare/identify these Bragg peaks from there, instead of labeling the assignment facets.

3# Too many typos in the whole manuscripts!

Reviewer 2 Report

In this manuscript E. Berretti et al investigated the variation of the Pd oxidation state in an alkali medium in situ by means of an EC-cell in FY. In order to follow how the oxidation changes depending on the applied potential the authors fixed the energy at a point of interest and collected the change in the intensity. Even it is an interesting approach from my point of view the results cannot be used as an effective technique to describe the changes in the oxidation state. It is because changes in the intensity can be related not only to the variation in the oxidation state but to changes in the background signal, dissolution, bulk oxidation etc… Furthermore, I don’t see the advantage of this technique vs. Quick XAS. Other issues:

1)      “In-operando” should be change by “operando”. Furthermore, in this manuscript the investigation was done in situ but not under operando conditions which require the analysis of the reaction products at the same time that the spectra is collected at a given potential.

2)      The way in which the oxidation state is calculated is not accurate and depends strongly in the particle size and homogeneous particle size distribution. Then this method is not useful.

3)      In the SEM figure the particle size cannot be distinguished, because of this this study requires a TEM characterization.

4)      Is the signal intensity onset depending in the potential scan rate?

5)      The use of a static cell is problematic from the point of view of bubbles formation during evolution process.

6)       The spectra in figure 4a are normalized, which is not possible during the potential tracking experiments.

7)      The quantification using a lineal dependence is not a good method. Changes observed here can be ascribed easily to changes in the bulk oxide formation and not in the surface. Therefore, the values observed can varied strongly in samples with different size and/or not homogenous particle size distributions.

8)      One indication that the oxidation is mostly related to changes in the bulk is the fact that the intensity increases linearly with the applied potential. In electrochemistry is expected that the oxidation/reduction occurs ones the redox potential is reached and that it is not change until the next redox potential is reached.

Reviewer 3 Report

From my viewpoint, conclusions are not so obvious, but applied approach is important in the corresponding fields. Therefore, I recommend publication of this work in Catalysts. Please see below necessary revisions.

1) Introduction, the first or next sentence had better have a few recent references (for example, J. Electroanal. Soc. 2018, 165, F3118-F3131, Bull. Chem. Soc. Jpn. 2017, 90, 1017-1026).

2) Please provide scale bars to images in Figure 1.

3) Figure 2 is fine, but measurement systems and mechanisms cannot be well understandable from this figure. Please add more information on analytical (measurement) methods.