Talk:Lennard-Jones potential

Does Buckingham potential belong?

Should the equation for Buckingham potential be presented? Should it even be mentioned except for a link in the see also? I believe that this should be cut because it doesn't really add anything about the LJ pontential and I don't see how something presented earlier(LJ) could approximate something presented later(Buckingham). If anything the Buckingham refined the LJ but since it is not very common that is questionable as well.

Phancy Physicist (talk) 18:50, 29 December 2010 (UTC)

It was added Dec.1 and incorrectly attributed to LJ himself, which would be a reason to include it if true. On Dec.25 I changed the attribution to Buckingham, but did not really think about removing it. The reference cited (Atkins) does say that "there is plenty of evidence to show that 1/r¹² is a very poor representation of the repulsive potential, and that an exponential form is greatly superior". However admittedly Atkins is only a physical chemistry text and those of you (not me) who use such potentials are probably more up-to-date with the evidence.

I suggest that both the Buckingham and Morse potentials be placed in a brief last section called, perhaps called Refinements of the L-J potential. We could say (with dates) that LJ was the first attempt at a potential function between two neutral atoms, and mention Buckingham and Morse as the first two attempts at refinement, and perhaps a few more recent versions too. This would give a little more context then just a See also section. Dirac66 (talk) 19:33, 29 December 2010 (UTC)

I like the idea of a refinements section. We just have to be careful of what potentials we call refinements of LJ in order to not take due credit from others. I have to check Morse's paper but he proposed his potential in 1929 so it could have been independently developed by him as easily as it could have been inspired by Lennard-Jones.

Phancy Physicist (talk) 19:54, 29 December 2010 (UTC)

Example values for Lennard_Jones Potential

After working on this forever, I could never find the proper values to match a theoretical atom that would properly work within this formula.

Argon Values: sigma, σ 3.405 epsilon, ε 120 m 11

Source: http://inside.mines.edu/~dwu/classes/chen610/projects/Sp06/Barkley%20statmech2.pdf

I would like to suggest sharing this information on the wiki page. — Preceding unsigned comment added by 74.167.120.101 (talk) 21:35, 20 February 2012 (UTC)

I am a little confused: the 3.405, 120 values are very similar to those of Rowley, Nicholson and Parsonage (see Argon on SklogWiki) for the Lennard-Jones model. However, the m=11 corresponds to equation 6 in you project for the m-6-8 potential function and the parameters are very similar to those of Klein and Hanley (see Caligaris and Rodriguez) --Spud Gun (talk) 11:54, 21 February 2012 (UTC)

I would not be opposed to adding in example values but I would like to be very careful about how it is presented. Though there are numbers which people swear to for σ and ε, these number are highly situation dependent and even sometimes come down to what exactly you are trying to model. What I mostly worry about is whatever values we decide to show would become "The Values" for anyone who casually reads up on the LJ potential. Perhaps giving the impression of a kind of "universality" for the numbers shown.

Phancy Physicist (talk) 21:37, 21 February 2012 (UTC)

Rayleigh-Taylor instability

One of the figures in this article is a Rayleigh-Taylor instability. It is in the computational physics box, but it seems irrelevant to me for this article. Any suggestions here? — Preceding unsigned comment added by 145.94.187.200 (talk) 07:29, 21 August 2012 (UTC)

Move to "Lennard–Jones potential" (containing en-dash instead of hypen-minus)

The following discussion is an archived discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review. No further edits should be made to this section.

The result of the move request was: request withdrawn. Favonian (talk) 21:22, 25 August 2012 (UTC)

---

Lennard-Jones potential → Lennard–Jones potential – "Lennard–Jones" contains an en-dash (see for example the article Navier–Stokes equations as a comparison), and not a hyphen-minus as is currently in the article title. I tried moving the article manually, but there's already a redirection page to this article with that name. —Kri (talk) 03:19, 24 August 2012 (UTC)

• Oppose, this is correct here. The endash is only used if we combine the names of two people, e.g. Claude-Louis Navier and George Gabriel Stokes, but this is only named after one guy. --The Evil IP address (talk) 15:36, 24 August 2012 (UTC)

Aha, I thought Lennard and Jones were two different guys :P I guess this settles it then. Should the move request be removed now or doesn't it matter if we leave it here? —Kri (talk) 21:17, 25 August 2012 (UTC)

The above discussion is preserved as an archive of a requested move. Please do not modify it. Subsequent comments should be made in a new section on this talk page or in a move review. No further edits should be made to this section.

Inconsistency between the first plot and the LJ equation

According to the first equation and the x axis of the plot being in units of sigma, the minimum of potential in the figure should be r_m/sigma (not r_m) — Preceding unsigned comment added by 129.237.99.102 (talk) 02:05, 26 March 2014 (UTC)

Indexes should be upright

To write "LJ" as an index for a potential V ist to name the potential (and LJ is a mere name not a parameter), so one should have upright indexes. May someone please change this. Jjh1993 (talk) 14:43, 4 February 2015 (UTC)

I just changed all non-variable subscripts to upright, that includes abbreviations for truncated, cutoff, etc. --Jkokorian (talk) 09:58, 2 September 2016 (UTC)

Variables should be declared with units

When variables are first introduced, their units should be explicitly mentioned. To me, as an electrical engineer 'potential' implies that the unit is Volt, but I've even seen Lennard-Jones potential curves with a 'potential' in 1/cm. Could someone please add the correct units? I would do it myself, but I don't know what they are... --Jkokorian (talk) 10:04, 2 September 2016 (UTC)

Notation inconsistencies

σ and r_m are used in the LJ potential equations but later on in "Explanation", r_0 is used for the Buckingham potential without an explanation as to what it is. Is it σ, r_m or something else? r_0 is also used in "Alternative expressions", subsection "AB form" where is states that at r_0, V = − ε. This would mean, at least for this part, that r_0 = r_m and two different notations are used. — Preceding unsigned comment added by Jvillatoro leal (talk • contribs) 09:58, 19 January 2018 (UTC)

Pauli repulsion

I do not understand how 1/r^12 may model a Pauli repulsion. I have searched INTERNET without results, then I have tried the formulae for H2 of Heitler-London for a such divergent term. Then I realized that the Pauli repulsion may only promote the electrons to the empty orbitals, resulting in a FINITE energy increase as r->0. I feel that the exponent 12 is chosen to override the 1/r^6 divergence in r=0 and reduce rapidly the potential to the physically sounded 1/r^6 as r increases. 151.29.78.113 (talk) 20:16, 27 April 2024 (UTC)

I removed the incorrect material claiming the 12 was related to Pauli repulsion. Johnjbarton (talk) 17:56, 29 April 2024 (UTC)

A question: the two links named "Pauli repulsion" point to different articles. Is normal? 151.29.78.113 (talk) 20:30, 27 April 2024 (UTC)

I change the link to point to Exchange interaction. Johnjbarton (talk) 02:07, 28 April 2024 (UTC)

Needs work.

• The intro here is too long and does not summarized the article.
• LJ fluids (fictitious fluids with properties caused by LJ potentia) should be grouped in separate section.
• No secondary refs for basic LJ concepts.
• Too many primary refs without secondary refs.
• editorial comments such as "frequently" "extensively" are not referenced.

Johnjbarton (talk) 02:13, 28 April 2024 (UTC)

The article fails to explain that 1) the LJ potential began as a model for noble gases (atoms), 2) the repulsive term is heuristic: it was fit to data. Johnjbarton (talk) 02:55, 28 April 2024 (UTC)

@Johnjbarton:Thanks for the prompt answer and the accurate reading of the article. The point that hurts a non-expert is that he immediately notes the strong divergence and thinks that the exponent 12 is needed for this; I feel that good results require a good fit only from (about) r0 outward ... I expect that someone has read somewhere this, but finding an explicit statement seems as difficult as to find explicitly 2*375=750. In statistical mechanics the teachers often say that int V exp(-V/kT) dx3 has little importance where V is very large and positive, but this is rarely written in the books pietro151.29.78.113 (talk) 05:13, 28 April 2024 (UTC)

Maybe add/borrow from https://doi.org/10.1021/acs.jctc.4c00135, which looks like a good overview. Ldm1954 (talk) 23:10, 28 April 2024 (UTC)

@Ldm1954 thanks but I don't have access to the full article. Johnjbarton (talk) 00:47, 29 April 2024 (UTC)

I have been able to add that ref now, thanks. I'll add a short history section as well. Johnjbarton (talk) 17:57, 29 April 2024 (UTC)

Unsupported opinionated statements

I have rewrote the following section as it violated many of Wikipedia's policies. No original research and no personal opinions are tolerated. The reference provided (by Stephan et al.) does support these statements.

The current state of research of thermophysical properties of the Lennard-Jones substance is summarized in the following. The most comprehensive summary and digital database was given by Stephan et al. Presently, no data repository covers and maintains this database (or any other model potential) – the concise data selection stated by the NIST website should be treated with caution regarding referencing and coverage (it contains a small fraction of the available data). Most of the data on NIST website provides non-peer-reviewed data generated in-house by NIST.

If this author disagrees, please provide evidence that **explicitly** supports these statements.

A.Cython (talk) 15:04, 17 May 2024 (UTC)

The statement seems factually correct looking at the NIST website that is linked. There seems to be mostly data procured in house by NIST that did not go the best-practice way of peer-review. But I agree that it is better to put this discussion here on the talk page. TimeStep89 (talk) 20:53, 20 May 2024 (UTC)

I disagree simply because, as stated on the NIST website [[1]], they reproduce peer-reviewed results with established methods and tools and make them available to the public for further evaluation and use, suggesting that the results are peer-reviewed constantly (once by the journal from which they are compared, by researchers at NIST and elsewhere, and by the public). By the way, the data in every academic group is produced in-house, meaning that it is often not always accessible during the review process or after. However, the NIST website provides the codes to reproduce the published data. As stated on the NIST web page, reproducibility is a priority.

Unfortunately, we live in times when a single round of "best-practice way of peer review" is insufficient (of failure to reproduce results from peer-reviewed papers), whether because researchers are trying to monetize upon the research or for other reasons. This is well documented in peer-reviewed journals that study the failure of the peer-review process #1, #2 just to name a few. Not to mention that the retraction of peer-reviewed publications has skyrocketed in the last decade; see [2] for details, resulting in trust deterioration between the public and academic institutions. This is why it is so essential for the reproducibility of scientific results.

So, unless you have explicit evidence to the contrary, I am reinserting the statement. This is the second time I have requested explicit evidence, so do not undo my edits without providing evidence; this is not the wiki-way. Be inclusive and constructive, the wiki-way. A.Cython (talk) 22:01, 23 May 2024 (UTC)

@A.Cython Your claim concerning "the wiki-way" does not match the policy of Wikipedia. That policy reads:

• The burden to demonstrate verifiability lies with the editor who adds or restores material

I encourage you to focus your argument in support of the content on the content and skip all of the extraneous commentary about "we live in times".

The content you restored was:

• A concise data selection along with codes to reproduce the results can be found in the National Institute of Standards and Technology (NIST) website.

This sentence has several problems independent of the reliability of NIST, but let's start there. NIST has a solid conservative reputation and it has a great deal of internal review. Absent direct evidence I think we should consider the work to be as declared on the site:

• The purpose of these pages is to provide some explicit results from molecular dynamics and Monte Carlo simulations for the Lennard-Jones fluid. It is intended to provide guides for testing codes. Reproducing these results is a test of the correctness of codes, either written by the user or obtained elsewhere.

That's a low bar.

In the sentence you added "concise"; that editorial comment is not something NIST says. The sentence is incomplete in characterizing the content of the NIST site and its purpose. The reference is incomplete.

Based on this I will make some changes. Johnjbarton (talk) 00:41, 24 May 2024 (UTC)

This is my new version:

• The US National Institute of Standards and Technology (NIST) provides its molecular dynamics and Monte Carlo simulation results as a guide for testing correctness of code.

With a ref to:

• Daniel Siderius (2017-09-01). NIST Standard Reference Simulation Website - SRD 173. doi:10.18434/T4M88Q.

Johnjbarton (talk) 00:50, 24 May 2024 (UTC)

I agree with @Johnjbarton. I find the proposed statement 'The US National Institute of Standards and Technology (NIST) provides its molecular dynamics and Monte Carlo simulation results as a guide for testing correctness of code.' better. However, I would add to the discussion that the Siderius/nist data should accordingly be considered a primary source, i.e. original research. From my understanding, a Wikipedia article should not contain such. (as @A.Cython initially him/herself ironically initiated the discussion) 2A02:810B:4B40:1CD4:8F1:98FF:BFD:C127 (talk) 11:35, 24 May 2024 (UTC)

Anonymous entries are treated with suspicion in Wiki. Not to mention that they undermine Johnjbarton arguments, I have to admit. A.Cython (talk) 17:42, 27 May 2024 (UTC)

It is actually not clear if the data on the website under discussion is a "(concise) data selection" (as edited by @A.Cython), i.e. a selection of data from different people/ codes etc. or data created by "Daniel Siderius", i.e. "provides its .. results" with a reference to D. Siderius (as in the current version edited by @Johnjbarton). In my view, the origin of data to be referred to in a Wiki article should be clear. It is a bit surprising that an institution with a reputation such as NIST leaves room for such unclarity. TimeStep89 (talk) 12:03, 24 May 2024 (UTC)

I'm puzzled by all of the attention the NIST site is given here. These simulation results are not "data" in the sense of a measured result of an experimental apparatus. They are just numbers computed from a program and the code for the program is available at the site. No scientific conclusions are proposed or discussed. I see zero reasons to question the origin of the data.

My question would be notability. But honestly the entire section that includes the NIST single sentence is too large by a factor of 3 or more in my opinion. Johnjbarton (talk) 15:26, 24 May 2024 (UTC)

The NIST site reference is a primary source and my version of the text follows WP:PRIMARY as far as I can tell. The claim that Wikipedia articles should not contain primary sources is not correct (I guess more than half of all citation in science articles are primary, may be more). Johnjbarton (talk) 15:17, 24 May 2024 (UTC)

Science articles are considered in most cases "secondary sources" not primary.

"A secondary source provides thought and reflection based on primary sources, generally at least one step removed from an event. It contains analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas taken from primary sources. Secondary sources are not necessarily independent sources. They rely on primary sources for their material, making analytic or evaluative claims about them."

From a history framework a primary source would be the diary of a notable person, while a secondary source would be a historian evaluating and placing the contents of the diary into context.

In more a science framework things are not so clear cut. However, the following can fit the definitions by Wiki: a primary source might be the original paper, in our case, the first paper by Lennard Jones, while all the subsequent papers are secondary since they reevaluate and extend the original work by Lennard Jones. A.Cython (talk) 17:13, 27 May 2024 (UTC)

I am ok-ish with the current version, which is a much improvement from the original version: "... the concise data selection stated by the NIST website should be treated with caution regarding referencing and coverage (it contains a small fraction of the available data)." Clearly, from your responses, you are unfamiliar with Wikipedia, so I strongly suggest you familiarize yourself with its policies and how they are implemented. Note that I still disagree with some of the arguments presented above, but these are small potatoes that I will let other editors fix these issues. My immediate task was to fix the obvious. Also, regarding "The burden to demonstrate verifiability..." I have demonstrated that the original version violated Wiki policies. On the other hand, you have not proven most of your arguments nor why my changes were repeatedly removed. A.Cython (talk) 17:24, 27 May 2024 (UTC)

Applications

@TimeStep89 added back a chunk of content I had deleted and put it in the Overview.

As I said in my edit summary, the material is just repetition. But I'm fine if you prefer this version over other ones in the article. The real problem is placement in the Overview section. The effect is to create an article within an article. Just for one example, the paragraph that starts:

• The Lennard-Jones potential yields a good approximation of intermolecular interactions for many applications:

clearly belongs in Applications.

The re-added material should be merged into the appropriate locations and the duplicate content removed. Johnjbarton (talk) 15:14, 20 May 2024 (UTC)

In general, yes. This could be done. Following this line of thought, the 'Lennard-Jones substance' section should in my view be seen as a application. Different applications would then be 'building block as force fields'.

I would suggest:

- Applications

- -- Lennard-Jones substance

- -- building block as force fields

Duplicate material should in my view be 'carefully' removed, i.e. not three long paragraphs in a single delete. That was the mainreason why I put it back in so it can be improved step-by-step, where appropriate. TimeStep89 (talk) 20:31, 20 May 2024 (UTC)

Sounds great, please go ahead. Johnjbarton (talk) 20:32, 20 May 2024 (UTC)

it makes much more sense seeing it now this way. But I think there can be done some for fine tuning (which essentially follows your statement that there are in parts some duplicate statements). TimeStep89 (talk) 20:59, 20 May 2024 (UTC)