aa r X i v : . [ phy s i c s . h i s t - ph ] O c t Mem. S.A.It. Vol. 75, 282 c (cid:13) SAIt 2008
Memorie della
My chemistry with Francesco
D. Galli
INAF – Osservatorio Astrofisico di Arcetri, Largo E. Fermi 5, I-50125 Firenze, Italye-mail: [email protected]
Abstract.
Through all his career, Francesco maintained a keen interest in primordial starformation and the chemistry of the early Universe. It was therefore quite natural for me,his former student and his o ffi cemate for more than 12 years, to be also involved in thesestudies. In this contribution I summarize the chemistry that Francesco and I developed,pointing out the main findings and false steps of our lifelong collaboration. Key words.
Early Universe – Molecular processes – Atomic Processes
1. Introduction
In the summer of 1987, Ralph Pudritz andMichel Fich organized a NATO School on“Galactic and Extragalactic Star Formation”in Whistler, a resort town in beautifulBritish Columbia, where Francesco was in-vited to give a review lecture on “PrimordialStar Formation”. Among the participants tothe School were three PhD student of theUniversity of Florence (Riccardo Cesaroni,Paolo Lenzuni and myself), and the some-what more mature Director of the ArcetriObservatory (Franco Pacini). For the three ofus, the youngest members of this scientificexpedition, the 1987 NATO School was thefirst experience of an international astronom-ical conference, and, in my case, my first ex-citing contact with the great American conti-nent. A picture taken during the School (Fig. 1)shows a relaxing moment during one of ourexcursions, when Francesco and Franco Paciniengaged in a resistance contest in the freezing-cold waters of the Garibaldi Lake (for therecords, Francesco lost). Francesco’s lecture,published one year later (Palla 1988), stim-
Fig. 1.
Francesco Palla and Franco Pacini swim inthe Garibaldi Lake, British Columbia (July 1987). ulated my interest in the chemistry of theearly Universe, and our discussions duringlong walks in the Canadian woods marked thebeginning of a lifelong collaboration with himon this subject.It is instructive to look in retrospect at thetopics that Francesco addressed in his 1987lecture, to assess the progress (or the lack of alli: My chemistry with Francesco 283 it) and the expected (or unexpected) develop-ments that have occurred in this field in theintervening 30 years. Francesco’s lecture cov-ered four broad subjects: (1) the search forzero-metal stars, (2) the chemistry of the earlyUniverse, (3) the evolution of collapsing gasclouds, and (4) the formation of primordialstars. These topics are still in the forefront ofcurrent research, as also witnessed by theseProceedings. In his lecture, Francesco gavealso a summary of his own work done in col-laboration with Ed Salpeter and Steve Stahlerthat had been published only a few years be-fore: the famous Palla-Salpeter-Stahler (andpermutations) “trilogy” of 1983–1986 (Palla etal. 1983; Stahler et al. 1986a,b)Our collaboration on primordial chemistrydeveloped from these premises. The first pa-per of the trilogy (which is Francesco’s mostcited paper) pointed out the importance of 3-body reactions, 3H → H + H and 2H + H → H + H for the gas-phase forma-tion of molecular hydrogen. The conversionof H into H was followed with the help of avery simple model for the collapse of a zero-metal cloud. Francesco was aware of the lim-itations of the collapse model and the chem-ical network adopted, and wanted to improveboth. Even though his main interest was onthe former aspect (collapse and star forma-tion), we resolved to work first on the latter(the chemistry), and continued to do so until2013. The motivation for proceeding in thisorder was the poor state of primordial chem-istry at the time, with many reaction channelspoorly identified, reactions rates badly guessedand often largely discrepant from author to au-thor, and the relevant data scattered in pub-lications not easily accessible (it was beforethe ADS). Especially annoying was the un-certainty on the H cooling rate for collisionswith H atoms. In the 1980’s, two indepen-dent calculation of this fundamental ingredi-ent for any recipe of primordial star formationwere available: one by Lepp & Shull (1983),and one by Hollenbach & McKee (1979, re-vised in 1989), based on independent sets ofcollisional rate coe ffi cients. The problem wasthat the two cooling rates in the low-densityregime di ff ered by as much as one order of magnitude or more below T ≈
100 K, a tem-perature range relevant for collapse calcula-tions. As a result, the estimate of the minimummass needed to collapse at redshift z ≈ ∼
30 (see e.g.Fig. 7 of Palla 1999). Clearly, before attack-ing our planned collapse calculation, a system-atic and critical reanalysis of the microphysicswas in order. The collapse calculation we orig-inally planned was eventually done, but notby us. The work by Kazuyuki Omukai andcoworkers is in my opinion the ideal fulfil-ment of our program (Omukai & Nishi 1998,Omukai 2000). Later on, Francesco would “ex-ploit” K. Omukai’s capabilities to further de-velop his trilogy papers: first, to explore theformation of massive stars by the enhanced ac-cretion rate expected in primordial conditions(Omukai & Palla 2001); and, second, to extendhis 1986 calculation of the mass-radius relationfor protostars to the case of a zero-metallicitycloud (Omukai & Palla 2003).Turning back to our chemistry, our criti-cal selection and analysis of gas-phase reactionrates for a mixture of H, He, D, Li and theirproducts was eventually completed and pub-lished in 1998, together with our “new and im-proved” cooling rates for the main molecularspecies (Galli & Palla 1998, hereafter GP98).Our recommended reaction and cooling rateswere widely adopted by researchers in the field(GP98 is my most cited paper), showing thatthe humble work of reordering and systematiz-ing can be useful too. Ten years later, the H-H cooling rate by GP98 was superseded bya newer calculation by Glover & Abel (2008),based on updated collisional rate coe ffi cients.The last word on the subject is represented inmy opinion by the extensive set of theoreticalcalculations performed by Franc¸ois Lique andcoworkers (Lique et al. 2012; Lique 2015) ofH-H collisional cross sections and rate coef-ficients down to temperatures of 10 K, whoseimpact on the H-H cooling rate still needs tobe fully assessed.In 2013 Francesco and I were invited towrite an Annual Reviews paper on primordialchemistry. For us this was a good opportunityto summarize the latest developments in a his-torical context (Galli & Palla 2013). We real-