# How did Mendeleev know elements from compounds or mixtures?

I read that Mendeleev initially thought didymium was an element, but it was actually a mixture.

How did he know the rest of the chemicals in the periodic table were elements and not compounds, especially since back then most elements would have been incredibly difficult to isolate and get a mono-atomic version? It seems like an impossible task in the 19th century.

• Element was what you can not split experimentally. For example, you can split rust to iron and oxygen if you really want to, but you can not split iron to other compounds (elements) by any chemical means. Yes, it was hard to decide, what is an element at the time. Playing with a lot of compounds, elements, checking also the mass ratio, made slowly clear the "atomic theory" (which was not very widely accepted even until the early XX. century!). Jul 28 '20 at 11:17
• It is impossible to give a simple answer in this forum. It was a difficult problem and each element is different. So a long discussion would be needed to deal with each element, from the lightest to the heaviest. It would need at least one page per element. And there are more than 100 elements ... Jul 28 '20 at 11:42

Please do not underestimate the scientists of 19th century. They were as creative, intelligent and perhaps more genuinely dedicated to science than the scientists of the 21st century. Spectroscopy was the tool of the trade to identify and verify that a given substance is not a mixture.

The original reference which established that Didymium was a mixture is in German [1] (English title: "On the decomposition of didymium in its elements"). I don't know enough German to read it conveniently, but it is clear that the authors did fractional crystallization of a lot of salts, made measurements absorption spectra, did emission spectroscopy and finally led to a beautiful result. The original paper has plenty of spectra! From [1, p. 490]:

Da sonach die exacte Zerlegung des Didyms in mehrere Elemente realisirt ist, so schlage ich vor, die Bezeichnung Didym nunmehr ganz zu streichen und beantrage, für das erste Element, entsprechendder Grünfärbung seiner Salze und seiner Abstammung die Benennung:

Praseodym mit dem Zeichen $$\ce{Pr}$$

und für das zweite, als das „neue Didym“, die Benennung:

Neodym mit dem Zeichen $$\ce{Nd}.$$

Aus den Atomgewichtsbestimmungen, die in allen Einzelnheiten nach Bunsen ausgeführt wurden, ergaben sich als vorläufige Werthe für Praseodym = 143·6, für Neodym = 140·8, wobei dem entsprechenden Oxyd die allgemeine Formel $$\ce{M2O3}$$ zukommt.

English translation with assistance of DeepL:

Since, therefore, the exact division of the didymium into several elements has been carried out, I propose that the name 'didymium' be deleted altogether and that the first element be given a name corresponding to the green colour of its salts and its origin:

Praseodymium with the sign $$\ce{Pr}$$

and for the second, as the "new didymium", the naming:

Neodymium with the character $$\ce{Nd}.$$

From the atomic weight determinations, which were carried out in all individual units according to Bunsen, it resulted as provisional values for praseodymium = 143·6, for neodymium = 140·8, whereby the corresponding oxide is assigned the general formula $$\ce{M2O3}.$$

Anyway, historically, it is not Mendeleev who had to decide whether the elements were actually elements. You may ask the right question, how did the element discoverer decided that it was an element in Mendeleev's time? The answer is that each element has a story to tell and that is why Isaac Asimov wrote a beautiful book called "The Search for the Elements". Go ahead and get hold of it.

### Reference

1. v. Welsbach, C. A. Die Zerlegung des Didyms in seine Elemente. Monatshefte für Chemie 1885, 6 (1), 477–491. DOI: 10.1007/BF01554643.
• "Please do not underestimate the scientists of 19th century. They were as creative, intelligent and perhaps more genuinely dedicated to science than the scientists of the 21st century." Ain't that the truth. Jul 29 '20 at 0:25
• To anyone interested, I strongly recommend this article as a window into the exhaustive work involved in the isolation of an element, even in the early 1900s. As the author states: The most important point proved by this work is that the element, giving the characteristic absorption bands of thulium, cannot be separated into simpler substances. After about 15,000 operations the absorption spectrum underwent no change. Jul 29 '20 at 0:25
• @NicolauSakerNeto, I was teaching a lab, and a student was collecting a Raman spectrum, which took exactly one min for acquiring the data. She was quite annoyed that it was taking too long! Jul 29 '20 at 0:46
• @M.Farooq This is why it's so important to know the history of our field! It blows my mind to think of the conceptual leaps and bounds that were made with flame and flask, yet the first time I heard of the 1800s obsessions with balloons, I myself thought it funny. The very fact that many of us initially take their work for granted/as obvious is testament to how well they did it. Jul 29 '20 at 15:32
• @jezzo, I completely agree! That is why I wanted to learn functional German. It comes to use in these historical cases! If you really like history, search "Ask the Historian" William B. Jensen. It is book of historical chemistry question and answers whose answers are not available in textbooks. For example, why R is used for universal gas constant, and so on. Jul 29 '20 at 16:04