Chemistry - Why do research papers mention experimental details in such detail?

Solution 1:

Experimental details are very, very important: they are used to ensure that results are reproducible, or at least that is the aim.

Let's talk about this specific example of suppliers and sources. One might naively think that a chemical purchased from supplier X is the same as the chemical from supplier Y. In an ideal world, that would be how it is. Unfortunately, this isn't the case, due to various considerations e.g. method of preparation, packaging, and so on. Therefore, even though the majority of the compound may be the same, there can be different impurities. And unfortunately, these impurities can lead to reactions working or failing. Many experimental chemists will personally know of cases where a reaction doesn't work until a reagent from a different company is used, or a reaction that stops working after buying a different bottle of a reagent.

There are many examples of this in the literature, but I'll choose the first one that came to mind, which is a reasonably recent paper from my old group (open access):

Mekareeya, A.; Walker, P. R.; Couce-Rios, A.; Campbell, C. D.; Steven, A.; Paton, R. S.; Anderson, E. A. Mechanistic Insight into Palladium-Catalyzed Cycloisomerization: A Combined Experimental and Theoretical Study. J. Am. Chem. Soc. 2017, 139 (29), 10104–10114. DOI: 10.1021/jacs.7b05436.

During these investigations, we also uncovered a critical dependence of the stereochemical outcome of the reaction on the batch of Pd(OAc)2 employed as catalyst. This discovery was made through the chance purchase of Pd(OAc)2 from a different supplier, which led to an unexpected ratio of enamide alkene geometries ((Z):(E) = 80:20), rather than the typical ratio of ∼97:3. Screening of further samples of Pd(OAc)2 gave variable results, the most extreme being a reversal of stereoselectivity to 40:60 in favor of the (E)-isomer.

The exact supplier isn't named, but 1H NMR spectra of various batches are given and the performance in the reaction can be correlated with the exact form of Pd(II) in the catalyst, which ensures that future readers are aware of what to look out for. This is very important for good science, because what's the point of publishing your fancy new method if nobody can replicate it?

Of course, this doesn't always happen. However, it's far better to include the information, just in case it is needed.

As mentioned by Waylander in the comments, commercial samples of samarium diiodide (SmI2) also have a reputation for being inconsistent: see e.g. this report by Szostak, M.; Spain, M.; Procter, D. J. J. Org. Chem. 2012, 77 (7), 3049–3059. DOI: 10.1021/jo300135v.

Solution 2:

You have already seen a detailed response by orthocresol. I can add one more extreme example, but it is not only the chemicals but the apparatus can also play a role in certain reactions. At times, glassware can also alter the rate of reaction. I am not an organic chemist, but some time ago, I was studying the epimerization of a certain natural sugar for developing a analytical method for its determination. Quacks claim that that particular sugar cures cancer. If you boil that sugar in water, glassware from different companies showed different rates of conversion.

If you happen to analyze commercial "enantiomerically pure" compounds, you will see this if often not the case. The story of commercial polymers is also the same.

If you get a chance to read older papers you will see very long passages and very minute details were provided often with a schematic diagram of the reaction apparatus. Today this level of detail is not deemed necessary and journals often put a limit on page numbers. For example, American Chemical Society journal Analytical Chemistry often returns the paper if the length exceeds 8 journal pages.

Once you get into the habit of reading papers, there is an online section, called Supporting Information. Good authors always write a detailed Supporting Information.

Solution 3:

wouldn't it be better if these were presented in an appendix with more data (say, with their batch numbers)?

It's certainly a matter of judgment how much detail is necessary. Sometimes the descriptions are kept very general if the authors think that is sufficient for correct reproduction. The cited text reports at the level of supplier which is a very common one (it also serves in a convenience way in telling readers at least one supplier where the reagents can be obtained). The next levels would be to give also

  • brand
  • grade/purity
  • article number
  • and finally the lot

(BTW, I put lot numbers into a manuscript just a few weeks ago.)

I don't think this is nationalistic or discriminating, just unneccessary, although I'm not the one with a PhD ;)

In my experience the amount of details given differs between fields.

  • Chemistry is among the picky ones in terms of very detailed reporting. Other answers already give examples where such details did matter. Personally I suspect that chemistry may be a field where actually a lot of reproduction happens compared to, say, sociological studies (at least that's my analytical chemist's view on synthesis): this reproduction of the paper is a side effect of doing the synthesis in order to obtain the product.

  • As PhD student, I attended a workshop with participants from all kinds of STEM fields and we had an excercise where e.g. chemists were reading physics papers and vice versa and then discussing them. One really prominent point in that discussion was that chemists kept complaining about the physics papers that they were of no use because the description of the experimental setups was not sufficiently detailed and clear to go to the lab and build such an apparatus. The physicists complained that the chemistry papers were full of details (implying that it is dangerous to give that level of detail since any competitor could come and pretty much immediately get to your lab's level)

Solution 4:

I would note that the precise list of reagents and descriptions of reaction conditions can be critical for select reactions to obtain a success and/or avoiding unfavorable side reactions. An extreme example of the latter can include explosions and/or the destruction of the created product of interest.

An illustrated example can be found in the field of advanced oxidation processes (AOPs) as commonly employed in environmental remediation and also disinfecting.

Apparently, for example, Oxone (active ingredient includes Potassium peroxymonosulfate) is widely employed as an oxidizing agent including for swimming pools. The action of KHSO5 in the presence of transition metals (like iron or some ions in even trace amounts, like cobalt) can produce powerful radicals. In fact, a trace of cobalt ions in the presence of HSO5- can produce the powerful sulfate radical anion (see "COBALT/PEROXYMONOSULFATE AND RELATED OXIDIZING REAGENTS FOR WATER TREATMENT" a thesis by Georgios P. Anipsitakis, available here). To quote from the dissertation:

It was found that when Co(II), Ru(III) and Fe(II) interact with KHSO5, freely diffusible sulfate radicals are the primary species formed.


$\ce{Co(ll)/Fe(ll) + HSO5- → Co(lll)/Fe(lll) + .SO4- + OH-}$ (Eq. 2.1 for Cobalt)

Note, if HSO5- is in excess:

$\ce{Co(lll)/Fe(ll) + HSO5- → Co(ll)/Fe(ll) + .SO5- + H+ }$ (Eq.1.11 & 3.8 for Fe]

And, particularly on the catalytic ability of cobalt ions:

It is also important to state that cobalt is needed only in small-catalytic amounts, thus its potential toxicity may be overcome. As low as 72.3 µg/L of dissolved cobalt were sufficient for the complete transformation of 50 mg/L of 2,4-DCP and approximately 30% organic carbon removal (Figure 2.9a). This value is between cobalt concentrations found in drinking water and below those tested in a recent study that reported the role of dissolved cobalt as a catalyst in an ozonation process (110).

My point here being, that transition metals, in even trace amounts, can particularly be significant, especially when engaging in cyclic reaction systems, producing pronounced impacts.

Bottom line, impurities should not always be assumed to be insignificant.

Solution 5:

One point that has not been explicitly covered in previous answers is the thought of the reader of the future.

In 10, 20, 50 years time the reagent/starting material you take for granted now may well not be available commercially. The procedure you usually follow to make X may be unfamiliar to the first year PhD student reading your paper. So stating exactly where your starting materials/reagents came from gives the future reader a starting point.

There are often assumptions made by researchers that the materials they consider common are well known to everyone. orthocresol mentions Pd catalysts, I never worked in a lab where anyone bought these as finished products, all were made up from PdCl2 given the relative costs of doing the synthesis vs buying the finished catalyst, but the relevant syntheses are not cited as often as they should be! I've even written a paper myself on common half sandwich compounds it was non-trivial to find a good, published, large scale prep on Aust. J. Chem. 70 (1), 113-119 despite the fact that a reasonable number of labs use them heavily.

As an example of the other extreme I've come across an older paper which stated "[non trivial, not currently commercially available] compound X was made by the usual method". No citation though! (I have not been able to dig the reference up for this, been out of chemistry a few years now and a search of my Papers library didn't throw anything up.)