One of the ways in which chemical elements have long been defined is as substances that cannot be decomposed. But what happens in cases where it is difficult to tell whether a substance has been decomposed or not? In this post, Sarah Hijmans illustrates some of the challenges in identifying chemical elements using the debate surrounding the nature of chlorine.
From the late eighteenth and well into the nineteenth century, chemical elements were defined as simple or indecomposable substances. The widely accepted definition was that promoted by the French chemist Antoine-Laurent de Lavoisier (1743-1794), who proposed to do away with all metaphysical speculation regarding the elements and instead establish a hierarchy of composition purely on the basis of operations that could be carried out in a laboratory.
It may seem simple enough to identify a (new) element according to Lavoisier’s definition: as long as a substance cannot be decomposed any further, it should be considered elementary. However, what happens in cases where it is difficult to tell whether a substance has been decomposed or not? The debate on the elementary nature of chlorine shows that the identification of simple substances can be more complicated than it might seem at first glance.
Chlorine was first produced and identified as ‘dephlogisticated muriatic acid’ (muriatic acid being the name for hydrochloric acid at the time) in the 1770’s. Later it was renamed following the nomenclature developed by Lavoisier and his colleagues, according to which acidity was thought to be linked to the presence of oxygen. Chemists working within this system referred to chlorine as ‘oxygenated muriatic acid’ or ‘oxymuriatic acid’.
In 1809, the French chemists Joseph-Louis Gay-Lussac (1778-1850) and Louis-Jacques Thenard (1777-1857) found that it was impossible to extract the oxygen from oxymuriatic acid, even when they used white hot charcoal, their strongest reduction method. The following year, the English chemist Humphry Davy (1778-1829) confirmed that oxymuriatic acid could not be decomposed, not even when using the voltaic pile (one of the first electric batteries). When he searched whether the presence of oxygen in the compound could be detected using different chemical methods, he found that this too was impossible. He therefore argued that oxymuriatic acid had been wrongly identified as a compound of oxygen and that it was in fact a simple substance. Because of the colour of its gas, he called the new simple substance “chlorine” after the Greek word for ‘green’ (‘χλωρός’).
However, despite the multiple failed attempts at decomposing oxymuriatic acid, the existence of chlorine was subject to debate for around six years. Jöns Jacob Berzelius (1779-1848), one of most famous chemists of the time, even opposed the new simple body until 1820. One of the main issues in this debate was that the two sides disagreed on which reactions to view as decompositions and which as combinations. Chlorine/oxymuriatic acid was known to react with hydrogen and this reaction could either be interpreted as a combination producing hydrochloric acid or a decomposition producing muriatic acid and water (more on this below).
As Hasok Chang has remarked, some of the difficulties in these kinds of cases stem from the fact that Lavoisier’s definition is rather circular: an element, or simple substance, is defined as a body that can’t be decomposed any further, and decompositions are defined as operations that break compounds down into simpler substances. Depending on which substance one defines as simple to begin with, chemical reactions might therefore be interpreted in various ways. Besides the debate surrounding the decomposition of oxymuriatic acid, this has led to a number of other disagreements on the identification of substances as either simple or composed (such as the case of water, as Chang has discussed elsewhere).
In order to break the circularity, Lavoisier and his followers relied on the principle of weight conservation: the idea that no amount of matter is created or lost during a chemical reaction. According to this principle, if substances gain weight, it gains in quantity of matter and the reaction must be a combination.
In the case of chlorine however, weighing alone could not solve the debate because the opponents of chlorine proposed an alternative hypothesis that could explain the experimental results. They thought it was impossible to extract oxygen from oxymuriatic acid because muriatic acid could only be formed in the presence of “intimately combined water”, water that was somehow contained in the acid and could not be extracted from it. In their view, oxygen could be separated from oxymuriatic acid only to combine with hydrogen, producing intimately combined water (thus explaining the weight gain that took place during the reaction). They explained the impossibility to decompose oxymuriatic acid in absence of hydrogen by the idea that ‘dry muriatic acid’, containing no combined water, could not exist. In other words, instead of viewing the reaction between chlorine and hydrogen as a combination (as we would today), these chemists saw it as a decomposition.
Thus, two possible hypotheses could explain the facts, and principles like theoretical coherence or simplicity had to come into play in order to decide between the two. To any current-day reader, the explanation using combined water might seem rather far-fetched and requiring more suppositions than the idea that chlorine was a simple body. However, muriatic acid was not the only mineral acid to have this property: nitric and sulphuric acid only existed in solution or as hydrates, and it was experimentally impossible to produce ‘dry’ nitric and sulphuric acid. It therefore made sense to generalize this idea and assume that the production of all mineral acids required intimately combined water. Moreover, the identification of chlorine as a simple body required such a significant rearrangement of chemical classifications that the opponents of chlorine saw their alternative hypothesis as the simplest option.
The most important issue was the classification of salts. At the time, salts were seen as composed of an acid, a metal and oxygen. The so-called ‘metallic muriates’, compounds of chlorine such as sodium chloride and potassium chloride, had all the properties of salts and it would therefore make sense to classify them as such. However, Davy had presented chlorine as a simple substance in the family of oxygen, and this meant metallic muriates would have to be classified as a kind of oxide. This completely contradicted the idea that similarities in properties were linked to similarities in composition, a key principle of chemical classification at the time.
Summing up the debate, Scottish chemist Thomas Thomson (1773-1852) identified the classification of salts as “the vulnerable part” of Davy’s view:
“Indeed, [Davy’s] opinions respecting [the muriates] cannot be embraced without overturning all the received doctrines respecting the neutral salts, doctrines upon which every thing resembling theory in chemistry is founded.” Since metallic muriates were composed of chlorine and a metal, chlorine needed to contain oxygen in order to mirror the composition of the other metallic salts. This was one of multiple arguments that motivated the opponents of chlorine to prefer the hypothesis that oxymuriatic acid could only be decomposed in the presence of hydrogen. In Berzelius’s words:
“While, therefore, the new doctrine [that chlorine is a simple body], when compared with the whole of chemistry as a theoretical science, is inconclusive, and inconsistent with itself, all the phenomena, when viewed according to the old doctrine [that oxymuriatic acid contains oxygen], are simple, consistent, and more than probable.”Only when iodine was discovered and identified as a simple body similar to chlorine did it become clear that chlorine was not just a strange exception, but a member of a family of substances which would come to be called the halogens. This constituted an important new argument in favour of the existence of chlorine, because it turned chlorine from an anomaly to a substance that had its place in a classification. Gradually the different views of salts and acidity were adapted to fit this new information in.
Rather than resulting from the experimental identification of chlorine as indecomposable, the acceptance of chlorine as an element can therefore be attributed to the accumulation of evidence that it was similar to the new element iodine, as Robert Siegfried, Jan Golinski and Tamsin Gray, Rosemary Coates and Mårten Åkesson argue as well. Even Berzelius, who still refused to accept chlorine himself, admitted that the discovery of iodine had “certainly induced many chemists to abandon the old doctrine”.
This famous debate from the history of chemistry can teach us many different things. Here I have mainly wanted to highlight the difficulty of evaluating the elementary nature of substances purely on the basis of laboratory operations. Often, the interpretation of such operations already requires a choice between different theories, and factors like coherence with existing knowledge therefore play a role in the choice of hypothesis. Even though Lavoisier and his followers argued in favour of a pragmatic or empirical definition of the chemical element, this does not mean that they could do away with any and all kinds of speculation.
Based in part on the paper "How to investigate the underpinnings of sciences? The case of the element chlorine", co-authored with Jean-Pierre Llored and published in Foundations of Chemistry (2020).