Lithium: a rare earth metal that is lighter than air?

---

Keith S. Taber

---

It was purely by accident that I had the radio on when I heard it claimed that

• lithium metal is lighter than air

The actual phrase that caught my attentions was

"the lightest metal on earth, lighter in fact than air, lithium".

This seemed bizarre, to say the least.

I went to the BBC website page for the programme concerned (an episode of a series called 'The Scramble for Rare Earths'), where it was acknowledged that a mistake had been made:

"Correction: this episode incorrectly states that lithium is lighter than air. Lithium has a lower atomic mass than oxygen or nitrogen."

So, someone at the BBC had spotted the mistake, or it had already been pointed out to them.

But I thought the correction was interesting in itself, as it made perfect sense to someone who would have the subject knowledge to have noticed the error and appreciated how it came about. I was less sure it would explain much to someone who did not already appreciate the ambiguity of the labels we give to both substances and also to the nanoscopic particles from which the are composed. More on that below, but first there is another clarification needed.

The bearable lightness of beings

Lightness strictly contrasts with heaviness, and so is about weight. Thus the 'old chestnut'¹:

which is heavier: a tonne of lead or a tonne of feathers?

Of course, by definition, the mass of a tonne of anything is fixed – so a tonne of feathers has the same mass (and so, in the same gravitational field, the same weight) as a tonne of lead, even if our instinct is that feathers are lighter.

So, really this is about density. When someone (sorry, Misha Glenny) makes the reference:

the lightest metal on earth, lighter in fact than air, lithium

We might respond in various ways:

1. this is a meaningless comparison as without knowing how much of each we are comparing it is not possible to draw any conclusion (a great deal of air would be heavier than a tiny amount of lithium);
2. to make sense of this we must assume we are considering the same amount of each; or
3. yes, that seems reasonable: the lithium metal on earth weighs less than the air.

Option 1 is clearly a pedantic response – 'we will not assume something you do not tell us: it is not for us to add necessary information you have omitted'. Now this would be my response, but then I know I have a mind which tends to such pedantry. I would like to think this is due to my scientific training but actually think it predates that. Perhaps I am just a pedantic person.

Option 3 follows a line of argument that as we have not been told how much is being considered, then the only admissible assumption from the claim is this refers to the earth ("metal on earth"). Lithium is a reactive element (which is why it is stored under oil to avoid oxygen reaching the surface – see the image below), not found native, so any lithium metal on earth is the result of deliberate processing – and, surely, there is much less weight of lithium metal here than the weight of the atmosphere. (However, invoking the earth as a whole complicates the mass:weight relationship, so we would need to specify where we are doing the weighing – say at the earth's surface beneath the atmosphere.) I suspect most people would immediately dismiss this as not being the intended meaning.

The sensible option?

Option 2 would surely be how most people would interpret "lighter than"given the context of the claim. As long as we agree on what we mean by 'the same amount', we have no problem. We have already considered one option which clearly does not work: same amount = same mass.

In chemistry, amount of substance is measured in moles. One mole of lithium weighs about 7 g. But we only apply moles to collections of what we take as identical entities. Lithium metal is comprised of lithium atoms², and although these atoms are not all identical (natural lithium has a small proportion of ⁶Li with only three nuclear neutrons mixed with the majority ⁷Li) they can be assumed near enough so for most purposes.

Air, however, is a mixture of different substances: mainly nitrogen, oxygen and argon, but also including many others in smaller amounts. And its composition is not fixed. In particular, the water vapour levels change a good deal; but also CO₂ levels (say, over a forest as opposed to a dessert, even ignoring anthropogenic inputs due to human activity); SO₂ levels being higher near active volcanoes; higher nitrogen oxides and ozone levels from car emissions in built up areas; carbon disulphide levels from algae being higher over some sea areas, etc, etc.

But dry air is mostly nitrogen and oxygen, and a mole of nitrogen weighs about 28g and a mole of oxygen about 32g, so a gas mixture which is nearly all nitrogen and oxygen and which contains 'a cumulative mole'* of these gases (say about 0.78 moles of nitrogen and 0.21 moles of oxygen, and a small amount of other atmospheric material) will weigh more than our 7g of lithium.

* But we had to do some violence there to the proper use of the mole concept.

The sharp-eyed will also have noticed that the molar masses of the substances oxygen and nitrogen were based on molecular masses – for the good reason that in the air these substances exist as diatomic molecules. That is, while it is true that "lithium has a lower atomic mass than oxygen or nitrogen" that cannot be a full explanation as atomic mass is not the most relevant factor for these gases. ³

So, 'same amount' cannot be weight, and it is not moles, but rather can sensibly be volume. Volume for volume, a denser material is heavier than a rarer (less dense) one. Sensibly, then, the claim being made in the radio programme/podcast can be understood to be that lithium is less dense than air – thus explaining all those lithium balloons we see floating around.⁴

---

---

The chemist's triplet

Now I have rather made a lot of a single broadcast error, that it seems had already been acknowledged (though I doubt more than a very small proportion of radio listeners systematically check out webpages to find out if the programmes they listen to contain any acknowledged errors). But I think what happened here is symptomatic of a core issue in chemistry, and the teaching of chemistry.

This refers to something first widely highlighted by the Glasgow based scholar Prof. Alex Johnstone in 1982, and known as his triangle. Chemistry deals with substances which are handled at bulk scale in the laboratory and explains the proprieties of these substances in terms of models concerning theoretical entities at the nanometre scale: electrons, bonds, ions, molecules, atoms and so forth. Johnstone referred to the 'microscopic' level as opposed to the 'macroscopic level', and so this term is widely used, but a better term might be submicroscopic (or nanoscopic, cf. the nanometre) as single ions and discrete molecules are not even close to being visible under any optical microscope. ⁵

Further, chemistry has its specialised symbolic language, such that, for example, a chemical reaction would commonly be represented in terms of a reaction equation using chemical formulae. The expert chemist or science teacher moves effortlessly around the macroscopic/microscopic/symbolic apices of this chemist's triangle, but Johnstone strongly argued that classroom presentations that readily moved back and forth between bench phenomena, equations, and talk of molecules, would overwhelm the working memory of the novice learner. (If you are not familiar with the critical notion of working memory, perhaps check out 'how fat is your memory?')

---

---

The idea of this 'chemist's triplet' (as it is sometimes labelled, borrowing a term from spectroscopy) has been considered the most important factor specific to chemistry pedagogy, and indeed there is at least one whole book on the topic (Reid, 2021).

Johnstone was right to alert science teachers to this issue, although an authentic chemistry education does have to work to move learners to a position where they can interpret and move across the triplet. Chemistry students need to learn to associate phenomena (e.g., burning) with both technical descriptions (e.g., combustion, oxidation) and with molecular level models of what is going on in terms of bond breaking and so forth (Taber, 2013). But learners will need much support and practice, and to be given sufficient time, to develop competence in this ability.

---

---

Perhaps a trivial example is moving from the everyday description that, say, mercury is 'heavy' (with only an implied notion of how much mercury we are talking about) to the more technical reference to density, and an explanation based on the micro-structure of mercury in terms of ions (and conduction electrons) in the condensed state.

---

---

Useful and misleading ambiguity

Part of the challenge for teachers (and so learners) is how key terms can mean different things according to context. So, the labels 'mercury' and 'Hg' can refer to an element in abstract, but they also refer to some of the substance mercury (a macrosopic sample of the element) and to an individual atom of mercury. A chemical equation (such as 2H₂ + O₂ ⇾ 2H₂O) can refer to bulk quantities of substances reacting, but also to individual molecules within the theoretical models used to explain what is going during the reaction.

Often during a chemical explanation the same label ('mercury') or equation (2H₂ + O₂ ⇾ 2H₂O) will be referred to several times, as the focus shifts from macroscopic substance to submicroscopic entities, and back again. This adds to the learning demand, but can also be confusing unless a teacher is very explicit at each point about which is being signified. An ambiguity which can be useful for the fluent, can also be confusing and frustrating for the novice (Taber, 2009).

So, in moving from considering atoms (where lithium has less mass than nitrogen) to substances (where under standard conditions lithium is much more dense than nitrogen) the meanings of 'lithium' and 'nitrogen' change – and perhaps that is what confused journalist Misha Glenny or whoever prepared his script. The experienced science teacher may find this a surprising basic error in a programme from an elite broadcaster like the BBC, but perhaps this is a useful reminder just how easily learners in introductory classes can be confused by the ambiguity reflected in the 'chemist's triplet'.

Read more about this macro-micro confusion

Rare earths – a double misnomer

The radio programme claiming lighter-than-air lithium was part of a short series on 'The Scramble for Rare Earths'. Now 'rare earths' refers to the elements also known as the lanthanides, part of the 'f-block' in the periodic table.

The term 'earth' is also found in the common name of the group of metals, including calcium and magnesium, known as the 'alkaline earths'. 'Earths' is an anachronistic reference to materials such as oxides which were not actually elements. So, the term alkaline earths is misleading as these elements are not 'earths' but the label has become established. (Who would be a chemistry student?)

The 'rare earths' are not earths either, but metallic elements. They are also not all especially rare. They were not initially readily recognised and characterised as different rare earth elements often have similar chemical properties and they occur in the same ores, and so historically they proved difficult to separate and identify. This label of 'rare earth' is then also something of a historical hang-over: although these elements are widely dispersed in the earth's crust so although not actually rare, they are seldom found in highly concentrated sources from which they can be readily extracted.

Rare earths do not seem to be well understood by the lay-person: quite a few websites claim that "quinine contains rare earth compounds" (see 'Would you like some rare earths with that?') As quinine (an antimalarial compound often taken as a 'tonic') is a single chemical substance, it clearly does not contain other compounds: but, in any case, its formula is C₂₀H₂₄N₂O₂ so its elemental composition is just of hydrogen, carbon, nitrogen and oxygen. Thus the repeated claim about rare earths in quinine seems curious.

The paradox of 'The Scramble for Rare Earths'

But then the BBC's blurb for the radio programme I came across, an episode of 'The Scramble for Rare Earths' called 'The Hidden Paradox', tells us:

"Misha Glenny explores the world of rare earth metals. Reducing CO₂ emissions requires critical raw materials like lithium, cobalt and nickel but mining and processing them can pose a serious threat to the environment. Can we solve the paradox?"

Presumably the paradox being that Misha Glenny explores the world of rare earth metals with reference to the alkali metal lithium, and transition elements cobalt and nickel: none of which are rare earths. Perhaps he found the rare earths too rare to include? Perhaps I need to listen to the rest of the programme.

---

Work cited:

• Johnstone A.H. (1982). Macro and microchemistry. School Science Review, 64, 377-379.
• Reid, N. (2021) The Johnstone Triangle. The Key to Understanding Chemistry, Royal Society of Chemistry.
• Taber, K. S. (2009). Learning at the symbolic level. Chapter 4, in J. K. Gilbert & D. F. Treagust (Eds.), Multiple Representations in Chemical Education, Dordrecht: Springer, pp.75-108. [Download chapter]
• Taber, K. S. (2013). Revisiting the chemistry triplet: drawing upon the nature of chemical knowledge and the psychology of learning to inform chemistry education. Chemistry Education Research and Practice, 14(2), 156-168. doi:10.1039/C3RP00012E. [Download paper]

---

Notes

---

¹ This is an idiom – a phrase which has currency in the language, and has come by convention to have a particular meaning; but where that sense is not clear from the literal meanings of the individual words. An 'old chestnut' is (when it is not a chestnut that is old) something which has been repeated so often it is familiar and loses any impact.

I suspect most readers will have met this question before (or a variation on it) and will not be caught out to suggest the feathers weigh less than the lead.

Read about idioms in science discourse

---

² Strictly, solid lithium metal contains an array of Li⁺ ions in a field of delocalised electrons. No particular electron is associated with any specific lithium ion – they are in molecular orbitals and – being delocalised – do not stay in the same place. So a mole of lithium is a mole of Li⁺ ions with a mole of (collectively, but not individually) associated electrons. This is just one complication that must make chemistry difficult for learners.

---

³ We say a mole of lithium is 7g, although this counts the individual 'atoms' (see note ²) even in the solid state where a single metallic crystal could be seen to be more akin to a single molecule. Of course, a mole of lithium crystals would have a mass MUCH more than 7g. Arguably, it is somewhat arbitrary how we define a mole of metallic lithium (and a mole of an ionic solid even more so) compared with a mole of simple molecular substances such as methane or carbon dioxide – but the convention is well established. This is a formalism that could be different – but not if you want to score the marks in a chemistry examination.

---

⁴ If lithium was heated to give off vapour, then that vapour would be less dense than air. However lithium is highly reactive and the vapour can only be kept stable in a vacuum or an inert atmosphere (and would only remain a vapour in an atmosphere at a high enough temperature). In the earth's atmosphere any slight leakage from a balloon would likely quickly lead to an explosive failure. Perhaps there is a planet somewhere with a hot enough argon atmosphere where lithium filled balloons could in principle be safely used if a suitable inert material could be found to make the balloon itself – but I doubt this would ever be a preferred option.

---

⁵ Arguably, a pure single crystal diamond could be considered a molecule, but that is not what we normally mean by a molecule. Again the choice of how to label different entities is somewhat arbitrary (in the sense that different decisions could rationally have been reached), and learners have to acquire the canonical, historically contingent, labels.

---