The birth, life and death of stars

An old sepia-toned photograph of some twenty women in dresses, in three rows, in front of the door of a brick building. — The women who worked at the Harvard observatory, photographed together in 1925; Cecilia Payne is in the back row, with Margaret Harwood, Arville Walker and Edith Gill. Harvard University Archives, CC BY-SA 4.0, source

This page explains what stars are made of, the historical discovery of their composition and the exact origin of the atoms that make up our world and our bodies. (How a star is born and dies will be covered in detail later in this chapter).

In 1925, a young student discovers what stars are made of. Then she writes, in her own thesis, that her result is probably not real.¹

She was right, and she knew it.

What was believed before

Astronomers thought that stars had roughly the same composition as the Earth: plenty of iron, rock, the same elements in comparable proportions. Simply much hotter.

The hypothesis seemed reasonable: the lines of elements known on Earth were observed in the light of the Sun, and they seemed to point the same way.

What Cecilia Payne understood

She works on the spectra of stars — light split into colours, in which dark lines appear that reveal the elements present.²

To them she applies a brand-new line of reasoning, borrowed from the Indian physicist Meghnad Saha, who had just established how heat strips atoms of their electrons. The theory was hardly four years old and was far from universally accepted.³

From it she deduces the following: the dark lines visible in a star’s light do not measure the number of atoms present, but only the number of atoms that are in the state required to absorb that precise colour. It is the heat that sets this proportion. Once this effect is corrected for, the weakness of the hydrogen lines no longer means “there is little of it”.

Her conclusion: hydrogen dominates the atmospheres of the stars.⁴

She was right. The values accepted today for the Sun, by mass, are as follows: about seventy-four per cent hydrogen, twenty-five per cent helium, and barely more than one per cent for absolutely everything else — including almost everything that makes up the human body. Seven elements alone make up 99.8% of the Sun’s mass. The “almost” refers to the body’s hydrogen: the only part of ourselves older than the stars, as we shall see below.⁵

What happened

Her thesis is reviewed externally by Henry Norris Russell, the influential Princeton astronomer — the former teacher of Harlow Shapley, who was then director of the Harvard observatory.⁶

Her measurement of temperatures is not at issue. But her reviewers judge that she goes beyond what her data prove, and she has to moderate her most disturbing conclusion: that hydrogen dominates the atmospheres of the stars.⁷

She gives way. In the published version, she writes that these quantities of hydrogen and helium are improbably high and almost certainly not real.

But she leaves all her calculations in place.

The historian David DeVorkin, who has studied this case closely, analyses it as follows: she gave way in a manner designed to record for posterity that she was the first to make this observation — right or wrong.⁸

This is not the act of a researcher reduced to silence, but that of someone who yields on form but keeps the evidence.

1929: Russell reaches the same result

In 1929, Russell publishes a long study, by another method, and reaches the same conclusion: hydrogen dominates.⁹

Four years have passed. The heart of the story fits in one sentence: the man who could validate this result did not believe it, and he had to arrive at it himself for it to be accepted.

What can be established

This story is widely told, and its most common version goes beyond what can be established.

One often reads that her discovery was “stolen” from her. We will not write that: we have found nothing that would establish it, and the historian who has worked most on this case describes something else — a demand for evidence, applied by reviewers who found the conclusion too strong for the data.¹⁰

A precise figure also circulates: Russell’s paper is said to have received about three times as many citations as her thesis. This figure has not been found in any historian’s work, any bibliometric study or any biography. A number that looks precise but cannot be traced to any source is worse than no number at all: it borrows the authority of a measurement without having made one.

So what should we remember?

This, which is established and is enough: a correct discovery was delayed by four years because those who could validate it did not believe it. The researcher who had made it had to write, in her own hand, that her result was probably not real — writing it in such a way that posterity would know who had seen it first.

This story is more interesting than a theft. It also has the merit of being verifiable.

Where the atoms we are made of come from

The phrase is famous: “we are made of stardust”.

It is true. It is also incomplete, and dated.

Three origins must be distinguished, and it is on the third that knowledge has changed.

One. Hydrogen and helium, with a little lithium, formed in the first minutes of the Universe. They never needed a star.

Two. From carbon up to iron, the elements are made by fusion, inside stars. It is for these that the phrase holds true.

Three. Beyond iron, fusion no longer produces energy — it consumes it. Another mechanism is needed.

This third case is not a marginal exception: half of all the elements heavier than iron were made by the fastest version of this mechanism.¹¹

The mechanism and the origin of heavy elements

To go beyond iron, nuclei must capture neutrons. This capture can happen slowly, in the envelope of certain ageing stars. Or very rapidly — this is the “r-process”, which requires an enormous neutron flux.¹²

The theory of this rapid process was worked out six decades ago. But where does it happen? The question remained under debate all that time.¹³

Models and circumstantial evidence pointed to one candidate: the merger of two neutron stars — the meeting of two dead stars.¹⁴

In 2017, such a merger was detected through the gravitational waves it sent out into space. In the days that followed, a glow appeared at the same spot: a “kilonova”, the first for which detailed spectra were recorded. Re-examining them in 2019, a team identified strontium.¹⁵

It was the missing piece: a freshly made element, identified at the very site where it was made.

What this means for us

Some of our atoms do not come from stars in the ordinary sense. They come from the meeting of two dead stars.¹⁶

The same detection taught another lesson along the way: neutron stars are indeed made of neutron-rich matter, as their name implied without proving it.¹⁷

The phrase “stardust” dates from the mid-twentieth century, when fusion in stars was understood, and holds for the second origin. It has not kept up with the discovery of the third. That is no reason to abandon it: it is a reason to complete it rather than repeat it.

Note 1

this led her to realize that the light elements hydrogen and helium dominated in the atmospheres of the Sun and stars; and how she was counseled to reject this conclusion in her thesis by her external advisor, Henry Norris Russell (1877–1957) of Princeton, in January 1925.

DeVorkin, David H., “Extraordinary Claims Require Extraordinary Evidence: C. H. Payne, H. N. Russell and Standards of Evidence in Early Quantitative Stellar Spectroscopy”, Journal of Astronomical History and Heritage, vol. 13, no. 2, 2010, p. 139-144