Interpretation of the K-edge XANES spectrum of 3d metal is not easy. The forbidden s-d transition contributes only a small part of the XANES spectra (normally the "pre-edge" peak). A rather comprehensive explanation of several features of the Fe K-edge XANES spectra can be found in a paper ( *ACS Nano* 2015, 9, 12, 12496–12505, Figure 2 and discussions). Ni should be similar. The "white line intensity" of 3d metals is affected by many factors including oxidation state but actually not that strongly. O coordination number and especially source (O, OH, H2O) also matter. Regarding the in-temperature XAS, you may refer to our recent publication (*J. Am. Chem. Soc.* 2020, 142, 3, 1417–1423) that gives a good example showing how FeOx transforms to metallic Fe or Fe-Nx with increasing temperature, providing temperature-dependent both XANES and FT-EXAFS spectra. Jia On Wed, Dec 15, 2021 at 8:59 AM 王茂林 wrote:

Dear Robert:

Thanks for your such a detailed and quick reply!

Here I get some points: Because s→d transition is forbidden, we cannot assure that the higher "white line" intensity, the higher the metals' oxidation states are. But we can also get metal's oxidation information via comparation the edge positions and features in spectrum with standard sample (in this case is Ni foil and NiO, the "white line" intensity of NiO is higher, so when the "white line" of spectrum in samples is getting higher, the formal oxidation number is higher). (First we assume that ligands are all oxygen atoms.)

I'm a beginner of EXAFS, is that right? Thanks again for your patience!

Marco Wang

-----原始郵件----- *發件人:*"Robert Gordon" *發送時間:*2021-12-12 12:04:35 (星期日) *收件人:* ifeffit@millenia.cars.aps.anl.gov *抄送:* *主題:* Re: [Ifeffit] Any coorelation about 'white line" in K-edge XAS with oxidation states?

Hi Marco,

Yes, a s - d transition is dipole forbidden, but the quadrupole transition can occur, as can a transition s - mixed p/d. The edge position in XAFS for metals like Ni and Cu is taken at the inflection point leading to the first main absorption feature, but is considered to have s - d character. For NiO, features near this edge indicator in the figure would be considered pre-edge. For NiO, the edge position would be considered shifted from that of Ni foil.

The white line for a K-edge generally corresponds to a noticeable number of unoccupied states with p character on the absorbing atom in the presence of the core hole (i.e. intermediate states) with "small" energy dispersion. Look at the energy difference between the edge indicator and "white line" indicator. That is about 18 eV. For most materials, the work function is 4 - 6 eV between the highest occupied states and the continuum in the ground state. Edge position is sensitive to charge transfer between target atom and ligands, not specifically a formal oxidation state. What you see is not a measurement of the ground state Density of States, but an influence of the core hole on intermediate states, and of the emitted photoelectron during the lifetime of the corehole.

Much of this terminology also arises from consideration of bulk materials, not nanoscale. Features in the XANES can arise from long range interactions, which are lessened as the scale gets considerably reduced.

The figure shows a clear transition from something with a pronounced white line feature to something with a less-pronounced feature and increased intensity near the foil edge energy, which I would interpret as a loss of coordinating oxygen yielding something more metallic, or at least more covalent, in character on the substrate. I am curious as to what Ni2Mo3N Ni K-edge would reveal in comparison to their results. It is not unreasonable for the authors to conclude a reduction of the Ni is occurring by loss of coordinating oxygen - "fully reduced to Ni(0)"...well, that is a bit of an overstatement - "extent of electron transfer from the Ni reduced to that of a more metallic/covalent environment" would be more exact, but I understand what is meant and would not argue with another author's writing style if it is not an egregious overstatement.

So, to answer your question: "how should we correctly interpret the “white line” and edge energy from XAS spectrum of metals’ K-edge and L-edge?"

We do this by a careful consideration of known standards and knowledge of the influence of bonding environment and extent of electron transfer between absorber and ligands

The concept of formal oxidation state (oxidation/reduction) is a convenient means of discussing electron transfer in more ionic systems but less useful as the degree of covalency increases.

-R.

On 2021-12-11 6:11 a.m., Marco Wang wrote:

Dear everyone:

Ihave some questions about the interpretation of "white line", from anarticle (https://doi.org/10.1038/s41467-021-27116-8) and the figure was attached. In the article, it is said that *“the sample was fully reduced to Ni (0) state at ~480 °C judging from the edge position and ‘white-line’ profiles”*.

First, we assumed that the metal was coordinated with the same anion.

We all know that *K*-edge transition in 3d and 4d metalis' spin forbidden, so it may be inappropriate to conclude that the intensity of the "white line" is correlated with metals' oxidation states. But in other words, the probability for s→d transition may be proportional to thenumber of empty d orbitals, so it seems like if the the intensity of "white line" is higher, the oxidation state is higher.

So, how should we correctly interpret the “white line” and edge energy from XAS spectrum of metals’ *K*-edge and *L*-edge?

Thank you in advance for your time.

Marco Wang

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

_______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.gov http://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit

Dear Robert and Qingying: Thanks for your further interpretation! I'm far more clear about "white line" now! Marco Wang -----原始郵件----- 發件人:"Robert Gordon" 發送時間:2021-12-16 00:07:24 (星期四) 收件人: ifeffit@millenia.cars.aps.anl.gov 抄送: 主題: Re: [Ifeffit] Any coorelation about 'white line" in K-edge XAS with oxidation states? Hi Marco, As a beginner, you should take a look at some tutorials to get a better understanding of what XAFS is: https://xafs.xrayabsorption.org/tutorials.html as well as the articles Jia mentioned in his reply. Many synchrotron facilities (or universities in partnership with them) offer introductory courses, with remote participation these days. A K-edge involves transitions from the core 1s level to what are primarily states with p character s --> p transition...the s --> d is a pre-edge feature. The white line is a s --> p feature for K-edge. In general, the more electron-withdrawing the ligands, the more the edge of the absorbing atom will shift to higher energy. For an element like Arsenic, correlation between white line intensity, edge position and formal oxidation state (for a given ligand type and arrangement) can be seen, but it can be more complicated. Type of near-neighbour, number of neighbours, arrangement of neighbours can all affect the near edge as well as the extended region. That is fundamentally why XAFS is a useful technique. Comparison to known compounds is a good first step in identifying what is happening with an unknown sample. -R. On 2021-12-15 5:58 a.m., 王茂林 wrote: Dear Robert: Thanks for your such a detailed and quick reply! Here I get some points: Because s→d transition is forbidden, we cannot assure that the higher "white line" intensity, the higher the metals' oxidation states are. But we can also get metal's oxidation information via comparation the edge positions and features in spectrum with standard sample (in this case is Ni foil and NiO, the "white line" intensity of NiO is higher, so when the "white line" of spectrum in samples is getting higher, the formal oxidation number is higher). (First we assume that ligands are all oxygen atoms.) I'm a beginner of EXAFS, is that right? Thanks again for your patience! Marco Wang -----原始郵件----- 發件人:"Robert Gordon" 發送時間:2021-12-12 12:04:35 (星期日) 收件人:ifeffit@millenia.cars.aps.anl.gov 抄送: 主題: Re: [Ifeffit] Any coorelation about 'white line" in K-edge XAS with oxidation states? Hi Marco, Yes, a s - d transition is dipole forbidden, but the quadrupole transition can occur, as can a transition s - mixed p/d. The edge position in XAFS for metals like Ni and Cu is taken at the inflection point leading to the first main absorption feature, but is considered to have s - d character. For NiO, features near this edge indicator in the figure would be considered pre-edge. For NiO, the edge position would be considered shifted from that of Ni foil. The white line for a K-edge generally corresponds to a noticeable number of unoccupied states with p character on the absorbing atom in the presence of the core hole (i.e. intermediate states) with "small" energy dispersion. Look at the energy difference between the edge indicator and "white line" indicator. That is about 18 eV. For most materials, the work function is 4 - 6 eV between the highest occupied states and the continuum in the ground state. Edge position is sensitive to charge transfer between target atom and ligands, not specifically a formal oxidation state. What you see is not a measurement of the ground state Density of States, but an influence of the core hole on intermediate states, and of the emitted photoelectron during the lifetime of the corehole. Much of this terminology also arises from consideration of bulk materials, not nanoscale. Features in the XANES can arise from long range interactions, which are lessened as the scale gets considerably reduced. The figure shows a clear transition from something with a pronounced white line feature to something with a less-pronounced feature and increased intensity near the foil edge energy, which I would interpret as a loss of coordinating oxygen yielding something more metallic, or at least more covalent, in character on the substrate. I am curious as to what Ni2Mo3N Ni K-edge would reveal in comparison to their results. It is not unreasonable for the authors to conclude a reduction of the Ni is occurring by loss of coordinating oxygen - "fully reduced to Ni(0)"...well, that is a bit of an overstatement - "extent of electron transfer from the Ni reduced to that of a more metallic/covalent environment" would be more exact, but I understand what is meant and would not argue with another author's writing style if it is not an egregious overstatement. So, to answer your question: "how should we correctly interpret the “white line” and edge energy from XAS spectrum of metals’ K-edge and L-edge?" We do this by a careful consideration of known standards and knowledge of the influence of bonding environment and extent of electron transfer between absorber and ligands The concept of formal oxidation state (oxidation/reduction) is a convenient means of discussing electron transfer in more ionic systems but less useful as the degree of covalency increases. -R. On 2021-12-11 6:11 a.m., Marco Wang wrote: Dear everyone: Ihave some questions about the interpretation of "white line", from anarticle (https://doi.org/10.1038/s41467-021-27116-8) and the figure was attached. In the article, it is said that “the sample was fully reduced to Ni (0) state at ~480 °C judging from the edge position and ‘white-line’ profiles”. First, we assumed that the metal was coordinated with the same anion. We all know that K-edge transition in 3d and 4d metalis' spin forbidden, so it may be inappropriate to conclude that the intensity of the "white line" is correlated with metals' oxidation states. But in other words, the probability for s→d transition may be proportional to thenumber of empty d orbitals, so it seems like if the the intensity of "white line" is higher, the oxidation state is higher. So, how should we correctly interpret the “white line” and edge energy from XAS spectrum of metals’ K-edge and L-edge? Thank you in advance for your time. Marco Wang _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.govhttp://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit _______________________________________________ Ifeffit mailing list Ifeffit@millenia.cars.aps.anl.govhttp://millenia.cars.aps.anl.gov/mailman/listinfo/ifeffit Unsubscribe: http://millenia.cars.aps.anl.gov/mailman/options/ifeffit