Proc. Natl. Acad. Sci. USA Vol. 90, pp. 4779-4781, June 1993 Colloquium Paper

This paper was presented at a colloquium entitled "Physical Cosmology," organized by a committee chaired by David N. Schramm, held March 27 and 28, 1992, at the National Academy of Sciences, Irvine, CA.

Challenges to the standard model of nucleosynthesis GARY STEIGMAN Departments of Physics and Astronomy, The Ohio State University, Columbus, OH 43210

ABSTRACT provides a unique The gap at mass-5 ensured that 4He would be the second most probe of the early evolution of the Universe and a crucial test abundant element (after hydrogen) in the Universe and that of the consistency of the standard hot Big Bang cosmological any heavier elements would be produced only in trace model. Although the primordial abundances of 2H, 3He, 4He, amounts. Since 2H and 3He (as well as 3H) are being burned and 7Li inferred from current observational data are in agree- to 4He, the higher the nucleon abundance the more rapidly ment with those predicted by Big Bang nucleosynthesis, recent are 2H and 3He burned away and the smaller will be their relic analysis has severely restricted the consistent range for the abundance. The magnitude of the nucleon abundance [as nucleon-to-photon ratio: 3.7 s 110 ' 4.0. Increased accuracy measured by the universal ratio of nucleons to (cosmic in the estimate of primordial 4He and observations of Be and background radiation) photons: 1 N/ y; 710o 101°71] is B in Pop II stars are offering new challenges to the standard crucial also to bridging the gap at mass-5 and, therefore, the model and suggest that no new light particles may be allowed mass-7 abundance is sensitive to q1. As a result, the relic (NBBN < 3.0, where N,, is the number of equivalent light abundances of2H, 3He, and 7Li provide both upper and lower neutrinos). bounds to the universal abundance of nucleons (3), A hot dense genesis for the Universe is strongly suggested by 2H, 3He, 7Li: 2.8 -'-110 ' 4.0. [1] the presently observed expansion coupled with the thermal The cosmic background radiation temperature (Tyo = 2.74) at spectrum ofthe cosmic background radiation. Along with the present observed large-scale isotropy and homogeneity, these em- (7) permits us to infer the present density of nucleons pirical data form the basis of the "standard" (i.e., simplest) (in terms of the critical density), hot Big Bang cosmological model. Extrapolation to very QNh2 = [2a] early epochs within the context of this model point to the 0.0037'1lo infant Universe as a primordial nuclear reactor. It is the 0.010 < Nh20.015, [2b] comparison of the predicted abundances of those elements synthesized during the early evolution of the Universe with In Eq. 2, Q-N PN/PC, where PN is the nucleon density and Pc those inferred from current observational data that provides is the critical density and the Hubble parameter is Ho = lOOh one of the very few direct tests of the standard model as well km*s-l Mpc-1 [1 parsec (pc) = 3.09 x 1016 m] (or, H-l = as constraints on alternative cosmologies (1-3). At this sym- 9.8h-1 gigayears). This BBN bound on QkN may be compared posium, Pagel (4) and Schramm (5) have reviewed the current with various estimates of the current mass density thus status-and successes-of the standard model. The consist- connecting directly the early and present Universe. For h c ency between the predictions of standard Big Bang nucleo- 1, QIN ' 0.01, suggesting that some (perhaps most) nucleons synthesis (BBN) and the observational data not only lends in the Universe are "dark" (3). For a lower bound of Ho . support to the hot Big Bang model but also leads to con- 40 km'-s1-Mpc-1 (8), fQN 5 0.09, providing strong support for straints on the baryon density of the Universe (1-5) and on the possibility that the Universe is dominated by nonbaryonic particle physics beyond the standard model (6). matter. As impressive as the successes of the standard hot Big The relic abundance of4He is very sensitive to the balance Bang model are, eternal vigilance is a prerequisite for the between the universal expansion rate and the weak interac- success of any scientific enterprise. Any model worthy of tion rate, providing a sensitive probe of "new physics" (6). consideration is worthy of challenge. Here, I will comple- For an inferred primordial 4He mass fraction (4, 9), Yp < ment the contributions of Pagel (4) and Schramm (5) by 0.240, there is a very restrictive bound on light weakly examining those challenges to the consistency of (standard) interacting particles (3), BBN that seem most serious at present. As a potentially falsifiable model for the structure and evolution of the AN,= N,-3 '003, [3] Universe, the hot Big Bang cosmology should be the subject of constant scrutiny. where N, is the number of equivalent light neutrinos. If, indeed, the standard (homogeneous, isotropic, . . .) hot The Standard Model Big Bang model provides an accurate description ofthe early evolution of the Universe, the inferred upper and lower Within the context of the standard model, when the Universe bounds to q should approach each other without crossing. was a few minutes old, conditions were right for nuclear Also, the LEP data (10) confirming the standard model result reactions to proceed rapidly, building the lightest nuclides that N -.3.0 [but, recall that LEP is sensitive to very massive (2H, 3He, 4He, and 7Li) (see refs. 1-5 and references therein). particles (sMz/2, where Mz is the Z-mass) that, if present, would play no role in BBN] leads to a lower bound to the The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" Abbreviation: BBN, Big Bang nucleosynthesis; ISM, interstellar in accordance with 18 U.S.C. §1734 solely to indicate this fact. gas. 4779 Downloaded by guest on September 26, 2021 4780 Colloquium Paper: Steigman Proc. Natl. Acad. Sci. USA 90 (1993) predicted primordial abundance of4He (3), Yp . 0.236, which With allowance for theoretical uncertainties, possible sys- must be tested. tematic observational uncertainties, and possible destruction of Li in these stars, Walker et al. (3) inferred (Li/H)p z 2 x Deuterium 10-10, which leads to the bounds on the nucleon abundance, Observations of 2H yield lower bounds to its primordial 7Li: 1.5 ' qlio s 4.0. [8] abundance (11) since 2H is destroyed in the course ofgalactic evolution. Very recent Hubble Space Telescope observa- Recently, a new wrinkle has been added to the quest for tions (12) of the local interstellar gas (ISM) toward Capella primordial Li. Be (16-18) and B (19) have been observed in yield several of the Spite plateau Pop II stars. The best (only) candidate for Be and B production in the early Galaxy is 105(2H/H)IsM = 1.65 - 0.07 [4] (spallation/fusion) nucleosynthesis (20-23). Along with any Be and B produced, 6'7Li will also be This is completely consistent with the earlier Copernicus synthesized by carbon, nitrogen, and oxygen (CNO) spalla- results summarized by Steigman (13) who found an average tion and, particularly, by a-a fusion (22). Although model- 105(2H/H)IsM = 1.6+ 30. It is of interest to compare this dependent uncertainties render very uncertain the prediction interstellar abundance with the presolar value (13). The 95% of absolute abundances of cosmic ray-produced Li, Be, and confidence level (2o-) ranges are B, relative abundances are less uncertain (22). Current Pop II Be and B data (16-19) are consistent with a cosmic ray origin 1.29 s 105(2H/H)IsM < 1.79 [5a] (22, 23) and suggest that perhaps -25% ofthe Li observed in the Spite plateau (15) stars may have a galactic (i.e., nonpri- and mordial) origin. Since there is a "floor" to BBN production of Li [(Li/H)BBN 2 1.1 x 10-10], it is crucial to attempt to 1.85 < 105(2H/H) C) < 3.27, [5b] separate the galactic and primordial contributions to the Li hinting at the possibility of some slight astration of 2H in the observed in Pop II stars. This could provide a crucial test of =4.6 gigayears since the formation of the solar system. the consistency of the standard hot Big Bang cosmology. Recently, Steigman and Tosi (14) have followed the evo- lution of 2H (and 3He) using models for the chemical evolu- Challenges tion of the Galaxy. In all models, there is little destruction of 2H between solar system formation and the present epoch Reanalysis of the 2H and 3He observations in the context of (less than a factor of 2.3), consistent with the results in Eq. models for the chemical evolution of the Galaxy (14) have 5. Steigman and Tosi (14) use solar system observations and narrowed the window ofconsistency. The inferred primordial their evolution models to set upper and lower bounds to abundances of 2H, 3He, and 7Li are still in concordance with primordial 2H, 3.9 s 105X2p < 9.0, which leads to constraints the predictions of standard BBN (3) but for a much more on the nucleon abundance, 3.4 ' q10 s 5.6, that are more restricted range ofnucleon abundance. Comparing Eqs. 6 and restrictive than previous constraints (3). Note, in particular 8, we have that the restrictive new upper bound to 2H [X2p < 9.0 x 10-5 2H, 3He, 7Li: 3.7 ' 7110 s 4.0. -+ (2H/H)p , 5.9 x 10-5] raises the lower bound to 110 (from [9] 2.8 to 3.4; see Eq. 1). Fortunately, the tighter constraints from 2H (and 3He) can be Helium-3 tested. Solar system and ISM observations combined with the bounds on 2H astration (14) lead to a predicted upper Although all 2H cycled through stars is destroyed, some 3He bound to primordial deuterium: (2H/H)p s 6 x 10-5. Future survives stellar processing. This result was exploited by Yang UV observations in metal-poor extragalactic systems could et al. (1) to derive an upper bound to primordial 2H plus 3He test this prediction (and, in the process, provide a less from solar system observations. Current data (3, 13) provide model-dependent estimate of the primordial abundance). the upper bound [(2H + 3He)/H]p s 1.0 x 10-4, which is What of4He and the Steigman, Schramm, and Gunn bound responsible for the lower bound to iq in Eq. 1 (qlo . 2.8). As (6) to NV? Pagel (4) has reviewed the observational situation we have just seen, the results of Steigman and Tosi (14) raise (see also ref. 9). At present, although the third decimal place this lower bound. They (14) have also used their models to is surely uncertain, the primordial 4He mass fraction is track the evolution of 3He, which, as anticipated, hardly bounded from above (95% confidence level?) by changes from its primordial value. For the (95% confidence level) range of the primordial mass fraction of 2H plus 3He YpBS 0.240. [10] consistent with solar system data, Steigman and Tosi (14) find 6.3 < 105X23p s 11, which corresponds to 3.7 s q10 s 5.7. If, indeed, the new lower bound to i1 from 2H + 3He (14) is Thus, this (galactic chemical evolution) model-dependent correct (j1102 3.7), then for a neutron lifetime of .882 sec (3), approach has raised the previous lower bound to 7, consid- the minimum BBN 4He mass fraction (forN, . 3) is predicted erably narrowing the "window of consistency." to be 2H,3He: 3.7 s ho<5 6 [6] yRBNyP 0.240. [11] Lithium-7 Thus the window appears to be closed. That is (to the extent that the third decimal place may be trusted), there is no room The warmer Pop II stars are observed to have the same Li for any additional light particles, abundance within a very narrow range (the "Spite plateau"; ref. 15). For several dozen such stars (3, 13), AN, = NV-3 s 0. [12] 10'0(Li/H)p,p 11 = 1.2 -+- 0.2. [7] This, truly, is a challenge of the standard model. Downloaded by guest on September 26, 2021 Colloquium Paper: Steigman Proc. Natl. Acad. Sci. USA 90 (1993) 4781 Conclusions 1. Yang, J., Turner, M. S., Steigman, G., Schramm, D. N. & Olive, K. A. (1984) Astrophys. J. 281, 493-511. The standard model of cosmology is testable. BBN and the 2. Boesgaard, A. M. & Steigman, G. (1985) Annu. Rev. Astron. observed abundances of the light elements provide a unique Astrophys. 23, 319-378. probe of the early Universe and a crucial test of the standard 3. Walker, T. P., Steigman, G., Schramm, D. N., Olive, K. A. & model. In particular, 2H, 3He, and 7Li provide upper and Kang, H.-S. (1991) Astrophys. J. 376, 51-69. lower bounds to the nucleon abundance that, at present, are 4. Pagel, B. E. J. (1992) Proc. Natl. Acad. Sci. USA 90, 4789- very restrictive (see Eq. 9). The dynamics of stars, galaxies, 4792. 5. Schramm, D. N. (1992) Proc. Acad. and clusters a means Natl. Sci. USA 90, 4782- provides for determining the present 4788. density ofnucleons. How do they compare? Returning to Eq. 6. Steigman, G., Schramm, D. N. & Gunn, J. E. (1977) Phys. 2 but using the tighter constraints in Eq. 9, the allowed range Lett. B 66, 202-204. for the present density of nucleons is only slightly modified 7. Mather, J. C., Cheng, E. S., et al. (1990) Astrophys. J. Lett. but tightly constrained, 354, L37-L40. 8. Sandage, A. & Tammann, G. A. (1990) Astrophys. J. 365,1-12. 0.014 s fQNh s 0.015. [13] 9. Olive, K. A., Steigman, G. & Walker, T. P. (1991) Astrophys. J. Lett. 380, L1-L4. Thus, the case for baryonic dark matter is strengthened; most 10. LEP Collaborations (1991) preprint. of the nucleons in the Universe are not shining. Since the 11. Epstein, R. I., Lattimer, J. M. & Schramm, D. N. (1976) upper bound hasn't changed, nucleons still fail-by a wide Nature (London) 263, 198-200. margin-to close the Universe and the higher lower bound 12. Linsky, J. L., Brown, A., Gayley, K., Diplas, A., Savage, strengthens the case for nonbaryonic dark matter dominating B. D., Ayres, T. R., Landsman, W., Shore, S. N. & Heap, S. the Universe. (1993) Astrophys. J. 402, 694-709. Still, the standard hot Big Bang model is facing some 13. Steigman, G. (1991) in Primordial Nucleosynthesis and Evolu- serious challenges. New estimates (14) of a lower primordial tion ofEarly Universe, eds. Sato, K. & Audouze, J. (Kluwer, abundance of 2H has driven the lower bound to q very close Dordrecht, The Netherlands), pp. 3-21. to the upper bound. Will they cross like ships in the night? 14. Steigman, G. & Tosi, M. (1992) Astrophys. J. 401, 150-156. Detection of Be and B in Pop II stars (16-19) suggests that 15. Spite, F. & Spite, M. (1982) Astron. Astrophys. 115, 357-366. some of the Li observed in these stars is not primordial. Will 16. Rebolo, R., Molaro, P., Abia, C. & Beckman, J. E. (1988) the eventual upper bound to primordial Li fall below the Astron. Astrophys. 193, 193-201. minimum abundance predicted by BBN? Consistency of the 17. Ryan, S. G., Bessel, M. S., Sutherland, R. S. & Norris, J. E. standard model awaits the outcome of these confrontations. (1990) Astrophys. J. Lett. 348, L57-L60. Finally, 4He, the second most abundant element in the 18. Gilmore, G., Edvardsson, B. & Nissen, P. E. (1991) Astrophys. J. 378, 17-21. Universe, is also the most accurately measured. Is its abun- 19. Duncan, D. K., Lambert, D. L. & Lemke, M. (1992) Astro- dance known sufficiently accurately to decisively challenge phys. J. 401, 584-595. the standard model? 20. Reeves, H., Fowler, W. A. & Hoyle, F. (1970) Nature (Lon- The present turbulent cosmological scene is evidence of don) 226, 727-729. healthy science. The standard model can be and is being 21. Walker, T. P., Mathews, G. J. & Viola, V. E. (1985) Astro- tested. We all await anxiously the test results. phys. J. 299, 745-751. 22. Steigman, G. & Walker, T. P. (1992) Astrophys. J. Lett. 3S5, I am grateful to my colleagues H. S. Kang, K. A. Olive, B. E. J. L13-L16. Pagel, D. N. Schramm, M. Tosi, and T. P. Walker for valuable 23. Walker, T. P., Steigman, G., Schramm, D. N., Olive, K. A. & discussions. This work is supported by the Department of Energy. Fields, B. (1993) Astrophys. J., in press. Downloaded by guest on September 26, 2021