Fred Hoyle An observer of the world and a ponderer on its problems ...

Fred Hoyle and Nucleosynthesis in Stars by Donald D. Clayton

Shortly after WWII Fred Hoyle created the theory that the chemical elements were assembled by nuclear reactions in the interiors of stars and subsequently ejected into the interstellar gas. Previous attempts at nucleosynthesis had focused on nuclear residues of an expanded dense state in the early universe. Hoyle advanced this new idea with two seminal papers [“The synthesis of the elements from hydrogen” Mon.Not.Roy. Astr. Soc.106, 343-383 (1946); “The synthesis of the elements from carbon to nickel” Astrophys. J. Suppl. 1, 121- (1954)]. This theory was proved correct by astronomical observation. This magic moment in mankind’s history solved an ancient philosophical puzzle: have the atoms always been here, and if not, when and how were they created?

At the conclusion of WWII, Hoyle (1946) demonstrated that stars in hydrostatic equilibrium must contract further when their nuclear energy supply runs out. By exhausting a sequence of fuels, Hoyle wrote, a star continues contraction unless halted by an angular momentum barrier. This continues until central temperatures reach 6x109K. At such high temperature internal matter is locally in nuclear statistical equilibrium. Hoyle suggested that portions of that NSE matter are ejected from the star when core contraction eventually produces a rotational instability. Hoyle argued that falling temperature in the expanding matter being ejected causes that matter to “freeze” at a lower NSE temperature. Specifically Hoyle showed that among this ejecta was matter frozen into a prominent abundance peak at iron, whose abundances he calculated. This was the first successful scenario for interpreting a puzzling feature of the observed abundances of chemical elements; namely that iron is 100 times more abundant than nearby elements combined. When BBFH (Burbidge, Burbidge, Fowler & Hoyle Revs. Mod. Phys. 29, 547-650 (1957)) reviewed this problem a decade later they named this the e process, which they recalculated with considerably better nuclear data and plotted in their Fig. IV.1 (p. 579). This triumph brought increased fame to Hoyle (1946) as the founder of nucleosynthesis in stars. In the following years Hoyle realized a problem, however. Although nuclear equilibrium would be achieved near the collapsing core, a large mass of overlying material would not achieve equilibrium. So in 1953 while a guest of the Caltech astronomy department, he worked out a masterly scenario for what would occur (Hoyle 1954) in a paper entitled “The synthesis of the elements from carbon to nickel”.

However, Hoyle’s 1954 paper has been forgotten by the scientific world, an oversight that needs correcting by practicing astrophysicists. While I attempt to do this with the plan described below, I call attention to the fact that the modern computerized world follows precisely the program described by Hoyle (1954) to understand the growth of the abundances of the chemical elements during galactic history. But it has done this without acknowledging the founding paper that it follows. The curious puzzle involves how the science community came to instead cite the later BBFH (1957) as the source for nucleosynthesis in stars. What Hoyle (1954) had done was to realize that the sequence of advanced core burning states in massive stars leaves in its wake a series of concentric shells containing the “ashes” of earlier burning phases. Ashes are the compositions remaining after a burning phase has completed. For example, when the carbon-burning core contracts to ignite neon burning in a smaller central core it leaves carbon-burning ashes outside that neon-burning core. Hoyle (1954) identified these sequences of shells and the major nuclei within them in his paper.

My Image
How could the source for this sweeping picture for nucleosynthesis have become forgotten? The answer, I maintain, is one of presentation. Without today’s graphics software in 1953, Hoyle (1954) could not include a figure such as the one shown here. Nor did Hoyle (1954) write the key equation for metallicity growth that he so clearly described in words, (Clayton’s “Hoyle’s Equation”, Science 318, 1876 (2007)); namely, in its simplest form, dmk/dt=B(t)Δmk. Here mk is the galaxy’s mass of primary isotope k between carbon and nickel and the left-hand side is the rate of growth of that mass. The right-hand side of Hoyle’s unwritten equation is the product of the galactic birthrate of stars more massive than 10Msun and Δmk is the mass of isotope k ejected from each. Hoyle stated this clearly, and had he included that equation and that figure, his paper would have been celebrated as original and unforgettable. One great idea that was not forgotten was the need for a 0+ excited state of 12C near 7.65 MeV excitation in order that carbon could acquire its large observed galactic abundance. Hoyle’s requirement of this nuclear state for the growth of a carbon core within presupernovae was one of the earliest and most dramatic examples of what is now called the anthropic principle; namely, if that excited state did not exist, our form of life would probably not exist. Although this idea is celebrated, its source is usually hearsay rather than citation of Hoyle’s 1954 paper.

The scientific world of nucleosynthesis within the stars began incorrect citations of the source of nucleosynthesis theory after BBFH (1957) was published. I discovered this shortcoming in the evolution of nucleosynthesis theory only gradually, and somewhat painfully inasmuch as I shared that guilt. It was, however, personally awkward for me to call attention to this while BBFH authors remained alive. So I waited until after Fowler and Hoyle died. Correction of historical credit is not easy. Effectively doing so requires high visibility. My plan became to deliver a prominent talk at a relevant conference and then to publish papers exposing the incorrect citation history.

Along about 1975 I had needed to read both Hoyle (1954) and BBFH (1957) very carefully because of our influential Rice University research program on explosive nucleosynthesis in massive stars (e.g. W.D. Arnett & D. D. Clayton, Nature 237, 780-784 (1970); Woosley, Arnett & Clayton, ApJSuppl 26, 231 (1973)). I was stunned to discover that we had been wrongly attributing the source for the supernova-shell theory that we had begun evaluating in those years. The scientific plan we followed had been clearly laid out by Hoyle (1954) in a fashion far superior to the somewhat muddled picture within the later BBFH paper. Research showed that nucleosynthesis in supernova shells is overwhelmingly the source of the abundant primary elements (those between carbon and nickel that can be created from initial hydrogen). This process causes the astronomically observed increase of galactic metallicity. Checking other influential papers for their attributions I found primarily abundant pro forma citations of BBFH and almost none of Hoyle (1954). Although I spoke with both Hoyle and Fowler privately about this, their close friendship with each other (and with me) made it awkward to go public with the deluge of past misattributions to BBFH. When I asked Hoyle about the muddled state of BBFH supernova nucleosynthesis, Hoyle said only, “It was my fault”, and declined to describe how this had happened with him being a coauthor of BBFH.

Fred Hoyle died in 2001, one year after
my final visit with him. The public talk component of my plan arose when Caltech announced sponsorship of a 2007 conference to celebrate the 50th anniversary of BBFH. I delivered my invited talk to attendees at Caltech on July 23, 2007. My notes for this talk can be seen here. Text displayed in black was the text shown on my PPt slides. Colored text indicates my spoken comments augmenting those written on the PPt slide. Most spoken comments are in red, but some are in other colors chosen to help my eye find them easily while I spoke. These notes were edited on August 3, 2007, immediately after the conference, to reflect what I actually had said during my presentation. My WORD file bears that electronic date and is thus demonstrably unedited subsequently.

Prior to the first slide of my talk I warned the audience at Caltech, “Instead of adulation of BBFH, I will cast the cold eye of the historian on what that paper achieved scientifically”. Nonetheless the audience was shocked that my talk did not echo the praises for BBFH that it expected. I did praise BBFH, but my praise was expressed as sociological reasons for it becoming one of the most influential papers in the history of astronomy. I did not praise BBFH as a source of nucleosynthesis theory, as very many subsequently have done. I pointed instead to the more than 100 citations of specific stars by BBFH that showed evidence of nucleosynthesis and to 111 citations of observational papers by practicing astronomers that contributed to that emerging science. These citations brilliantly heightened the sense of astronomical participation in creating the science of nucleosynthesis. Astronomers therefore embraced the theory with an intensity that would not have been generated by any theoretical paper. The indifference to Hoyle (1954) demonstrates that point.

My 2007 Caltech talk emphatically credited Fred Hoyle with creating the theory of nucleosynthesis in stars. His 1946 and 1954 papers creating the theory were more powerful than the subsequent BBFH presentation, and I said so clearly.
My notes. His 1954 paper single handedly laid down the program that the supernova theory for the increase of galactic metallicity would follow for the six decades! But research papers (including my own) nonetheless began immediately to cite BBFH as an appropriate outline of the theory, even as a fountainhead for it. To test that suggestion I selected for my talk 30 reprints near 1970 from my own personal collection, papers that I judged to be significant research contributions to the theory. Twelve of those made no citation of BBFH at all, undoubtedly because their policy was to cite only those papers that contributed the ideas used for their own studies. Of the remaining 18 papers in my survey all cited BBFH, whereas Hoyle (1954) had one citation. Authors began to cite BBFH for the origin of the idea of nucleosynthesis in stars even when they had read neither it nor Hoyle (1954)!

More recent evidence of the citation negligence of Hoyle’s papers reveals itself in the special Reviews of Modern Physics issue (Rev Mod Phys 69, 995-1084 (1997)) entitled “Synthesis of the elements in stars: forty years of progress”, in which 15 experts in nucleosynthesis purport to analyze forty years of progress since publication of BBFH (1957). I had declined an invitation to contribute because I sensed how unjust to Fred’s memory the collection would become. Those 15 experts made not a single citation of Hoyle (1954) in their ten-page list of references! The most important paper in the history of nucleosynthesis in stars was not referred to even once. Several conclusions come to mind in an historical analysis. The publication of BBFH prompted a collective repression of Hoyle’s papers. Secondly, active researchers have not even read Hoyle (1954). Many probably assumed that Hoyle’s coauthorship of BBFH would have ensured adequate visible credit to Hoyle’s ideas within BBFH, but his ideas were not discussed in BBFH. Why? Hoyle was an indifferent editor of work being polished for publication, and never blew his own horn, perhaps even to a fault. Hoyle’s restless intellect was always pushing on to new frontiers. His weakness was his calm assurance that scientists collectively would sort out citation credit and his underestimation of the length of time that would require. So the only explanation he ever would offer me was “It was my fault”. I believe this to be a correct confession, but I discount the notion that BBFH was intentionally written so as to hide Hoyle’s map for nucleosynthesis. Hoyle’s coauthors, frankly, did not fully understand and appreciate what Hoyle (1954) had achieved. Although coauthors drafted BBFH, it was surely the fault of coauthor Hoyle for not correcting those drafts to give higher visibility to the supernova-shell theory. The coauthors became enamored with defining and giving names to nuclear processes. Those process names were loved by astronomers and astrophysicists alike, and those names contributed to the great fame of BBFH. That fame not only suppressed Hoyle’s paper, but it also began the perception that nucleosynthesis was a Caltech creation even though three of its coauthors were British.

The third step of my plan was to publish a more complete scientific review of original ideas in Hoyle (1954). I did this in the following year (Clayton
New Astronomy Reviews 52, 360 (2008)) in a review entitled “Fred Hoyle, primary nucleosynthesis and radioactivity”.

Fred Hoyle can never be forgotten by those of us who knew him. As time goes by he should also be remembered by those who never knew him personally as being the originator of the theory of nucleosynthesis within the stars. Astronomers must realize that even if they personally had been introduced to nucleosynthesis by BBFH, they should nonetheless cite its true source.

Donald D. Clayton
Centennial Professor of Physics and Astronomy Emeritus
Clemson University
September 6, 2013