We explore the possibility that "deep mixing" is the reason that there are "too many" red giants in the Galactic globular clusters M5 and M30. Deep mixing has often been invoked to account for the abundance anomalies observed in many bright Galactic globular cluster giants. Here we assume that it also adds fuel to the hydrogen-burning shell of a typical giant in some clusters and returns helium to the envelope above. We use a simple mixing model to estimate how much extra fuel has been added to the hydrogen-burning shell of a typical red giant in M5 and M30 from the number of extra stars in the red giant branch (RGB) luminosity functions for these two clusters. Presumably, extra red giants are common in Galactic globular clusters. M5 and M30 were chosen solely because the available luminosity functions for both were derived from an exceptionally large sample of stars in two colors: the RGB luminosity functions are especially well defined and well normalized to unevolved main-sequence stars.If deep mixing is the reason that there are too many M5 giants, the number of extra giants suggests that the envelope of a typical M5 star should be significantly enriched in helium (Delta Y approximate to 0.07) by the time it reaches the tip of the giant branch. Hence, deep mixing may be (at least part of) the reason that M5 has a surprisingly blue horizontal branch (HB) for a moderately metal-rich cluster. If deep mixing occurs and Delta Y approximate to 0.07, we can account for part, but only a small part, of the very low value of R-and the surprisingly low estimate for Y-that Sandquist et al. found from their B-band luminosity function. Sandquist et al.'s value of the helium indicator delta suggests that the helium abundance is enhanced (Delta Y = 0.04 +/- 0.02) in HB stars at the blue edge of the instability strip.If deep mixing accounts for the "extra" giants that have been observed in M30, Y will be about 0.05 higher in the envelope of a typical M30 HE star than it was when that star left the main sequence. In addition to accounting for the extra M30 giants, deep mixing can give a good account of (1) the lack of a "bump" on the RGB or the presence of a small bump in the wrong place, (2) the surprisingly bright horizontal branch, and (3) the anomalous values of both Delta and R.The abundances of the light elements suggest that deep mixing is much more common in some clusters than in others and more extensive in some stars in a cluster than in others. If deep mixing does add fuel to evolving giants in some clusters, there are many interesting ramifications. For example, all horizontal branches may not have been created equal; some might well be brighter than others. Sweigart and Sweigart & Catelan have recently described how mixing helium out into the envelope of a red giant may solve a number of cluster blue star mysteries. We point out a number of additional mysteries that might be solved by looking at the fuel that is mixed down into the hydrogen-burning shell: e.g., the large range in the values that have been observed for R (the ratio of bright giants to HE stars), the observed range in the helium estimator Delta at a constant metallicity, and the variations observed in the RGB "bumps" among otherwise similar clusters.

By employing two different spectroscopic techniques we have mapped out the variable ISM dust extinction endemic to globular cluster M4. We derive an average E(B-V) reddening of 0.33 +/- 0.01 and R = 3.4 +/- 0.4, both in good agreement with previous studies of M4. For individual stars in the most heavily reddened regions of M4, we find E(B-V) values significantly higher than those inferred by studies of interstellar gas; for heavily reddened regions, the gas measured in column density measurements may not completely trace the dust.

We have serendipitously discovered an extremely lithium-rich star on the red giant branch of the globular cluster M3 (NGC 5272). An echelle spectrum obtained with the Keck I HIRES reveals a Li I 6707 Angstrom resonance doublet of 520 milli-Angstrom equivalent width, and our analysis places the star among the most Li-rich giants known: log[epsilon(Li)] ~= +3.0. We determine the elemental abundances of this star, IV-101, and three other cluster members of similar luminosity and color, and conclude that IV-101 has abundance ratios typical of giants in M3 and M13 that have undergone significant mixing. We discuss mechanisms by which a low-mass star may be so enriched in Li, focusing on the mixing of material processed by the hydrogen-burning shell just below the convective envelope. While such enrichment could conceivably only happen rarely, it may in fact regularly occur during giant-branch evolution but be rarely detected because of rapid subsequent Li depletion.

We present a chemical composition analysis of 36 giants in the nearby mildly metal-poor ([[Fe/H]] = -1.18) "CN-bimodal" globular cluster M4. The stars were observed at the Lick and McDonald Observatories using high-resolution echelle spectrographs and at the Cerro Tololo Inter-American Observatory using the multiobject spectrometer. Confronted with a cluster having interstellar extinction that is large and variable across the cluster face, we combined traditional spectroscopic abundance methods with modifications to the line depth ratio technique pioneered by Gray to determine the atmospheric parameters of our stars. We derive a total-to-selective extinction ratio of 3.4 +/- 0.4 and an average [E(B-V)] reddening of 0.33 +/- 0.01, which is significantly lower than that estimated by using the dust maps made by Schlegel and coworkers. We determine abundance ratios typical of halo field and cluster stars for scandium, titanium, vanadium, nickel, and europium with star-to-star variations in these elements of less than +/-0.1. Silicon, aluminum, barium, and lanthanum are overabundant with respect to what is seen in other globular clusters of similar metallicity. These overabundances confirm the results of an earlier study by Brown & Wallerstein based on a much smaller sample of M4 giants. Superposed on the primordial abundance distribution is evidence for the existence of proton capture synthesis of carbon, oxygen, neon, and magnesium. We recover some of the C, N, O, Na, Mg, and Al abundance swings and correlations found in other more metal-poor globular clusters, but the range of variation is muted. In the case of Mg and Al, this is compatible with the idea that the Al enhancements are derived from the destruction of Mg-25,Mg-26, not Mg-24. We determine that the C+N+O abundance sum is constant to within the observational errors and agrees with the C+N+O total that might be expected for M4 stars at birth. The asymptotic giant branch (AGB) stars in M4 have C, N, and O abundances that show less evidence for proton capture nucleosynthesis than is found in the less evolved stars of the red giant branch (RGB). Deeply mixed stars of the RGB, subsequent to the helium core hash, might take up residence on the blue end of the horizontal branch and thus fail to evolve back to the AGE, but reasons for skepticism concerning this scenario are noted.

We present new VI photometry for the halo globular cluster M30 (NGC 7099=C2137-174) and compute luminosity functions (LFs) in both bands for samples of about 15,000 hydrogen-burning stars from near the tip of the red giant branch (RGB) to over 4 mag below the main-sequence (MS) turnoff (TO). We confirm previously observed features of the LF that are at odds with canonical theoretical predictions: an excess of stars on subgiant branch (SGB) approximately 0.4 mag above the turnoff and an excess number of RGB stars relative to MS stars. Based on subdwarfs with Hipparcos-measured parallaxes, we compute apparent distance moduli of (m-M)V=14.87 ± 0.12 and 14.65 ± 0.12 for reddenings of E(V-I)=0.06 and 0.02, respectively. The implied luminosity for the horizontal branch (HB) at these distances is MHBV=0.11 and 0.37 mag. The two helium indicators we have been able to measure (R and Δ) both indicate that M30's helium content is high relative to those in other clusters of similar metallicity. M30 has a larger value for the parameter ΔVHBTO than any of the other similarly metal-poor clusters for which this quantity can be reliably measured. This suggests that M30 has either a larger age or higher helium content than all of the other metal-poor clusters examined. The color-difference method for measuring relative ages indicates that M30 is coeval with the metal-poor clusters M68 and M92.

We present spectroscopic evidence for small star-to-star variations in the abundance of the atoms from Ca to Ni (primarily Fe) among three bright giants in the metal-poor globular cluster M92: V-45, XI-19, and XII-8. Although all three have very similar B - V colors and visual magnitudes, weak "iron peak" absorption lines are typically about 15%–30% stronger in the spectrum of star XI-19. A line-by-line analysis shows that the abundances of the "iron peak" elements are very similar in stars V-45 and XII-8; the (logarithmic) mean difference is only 0.01 ± 0.01 dex for 106 lines. The average abundance of the "iron peak" elements is 0.18 ± 0.01 dex larger in XI-19 (121 lines). The very small "internal" error in the mean difference can be traced to (1) comparing the abundances one line at a time in order to avoid errors introduced by uncertain gf-values, (2) averaging equivalent widths for two spectra for both XI-19 and the comparison stars to reduce random errors in the equivalent widths, and (3) using a large number of lines to drive down the error in the mean. The "external" errors must also be small: the very small differences in the positions of the stars in the (V, B - V) color-magnitude diagram imply such similar values of Teff and log g that the line-strength differences must be due to real abundance differences. Comparison of small "iron peak" abundance variations with the larger variations that have been observed in the abundances of the light elements C, N, O, Na, Mg, and Al or the n-capture elements Ba and Eu in metal-poor Galactic globular clusters appears to be possible and may shed some light on the origins of both in the future.

We find significant star-to-star variations in the abundances of both the light elements oxygen, sodium, magnesium, and aluminum, and the neutron capture elements barium and europium in a sample of eighteen bright M15 giants. The oxygen and sodium abundances are strongly anticorrelated as are the magnesium and aluminum abundances. Sodium and aluminum are correlated. Hence, there are M15 giants with large sodium and aluminum ''enhancements'' and smaller oxygen and magnesium ''depletions'' as well as giants with more ''normal'' abundances of these light elements. The correlations are very similar to those found earlier in M13, M10, and M92. The CNO abundance sum is approximately the same in the five M15 giants for which all three elements have been observed. This abundance pattern is most likely the result of proton-capture nucleosynthesis and most easily explained if it takes place within the stars that we have observed. The barium and europium variations are unlikely to be the result of nucleosynthesis within the stars that we have observed. They provide strong evidence for ''primordial'' neutron-capture abundance variations in a cluster other than omega Cen. The range in both [Eu/Fe] and [Ba/Fe] is about 0.6 dex. The abundances are strongly correlated and appear to fall into two groups, one with larger Eu and Ba abundances, the other with values of [Eu/Fe] and [Ba/Fe] that are about 0.35 dex smaller. The mean value of [Ba/Eu] is the same for both groups, and falls between the pure r-process value and the solar system r+s process mix. The light element abundances apparently do not correlate with the neutron-capture abundances even though both vary over a significant (often large) range. We find no evidence for star-to-star variations in either [Fe/H] or [Ca/Fe]. We have looked for small differences in [Fe/H] and [Ca/Fe] between stars with large and small values of [Eu/Fe] and [Ba/Fe]; the average values of [Fe/H] and [Ca/Fe] are the same for both. Hence, we find that there are large variations in the abundances of a number of elements (but not all of them) in M15, a cluster only recently thought to have a uniform composition. The abundances of the light elements (C, N, O, Na, Mg, and Al) are strongly correlated as are the abundances of the neutron-capture elements (Eu, Ba), but variations in the abundances of light and neutron-capture elements seem to be independent of each other, and both appear to be independent of the Fe and Ca abundances. (C) 1997 American Astronomical Society.

Observations of 11 giants in the globular cluster M13 fainter than M(V) = -1.7 have been obtained with the High Resolution Echelle Spectrograph on the Keck I Telescope. When combined with similar data for brighter giants obtained with the 3.0-m Shane Telescope's Hamilton Echelle spectrograph, these data allow examination of the detailed abundances of oxygen, sodium, magnesium, and aluminium over a wide range of luminosity on the red giant branch. We find that oxygen depletions, correlated with sodium enhancements, can be found in giants with luminosities down to nearly the level of the horizontal branch at M(V) = 0.3. Aluminum abundances in M13 giants also span a wide range (similar to 1.4 dex) and are anticorrelated with the observed magnesium abundances. However, field halo giants of comparable metallicity and evolutionary state exhibit (normal) high abundances of O and Mg and low abundances of Na and Al. Stars in M13 with high sodium and aluminum and low oxygen and magnesium abundances are found at all giant branch luminosities, as are stars with more normal light element abundances (as defined by field halo stars of comparable metallicity). The detailed abundance patterns of the light elements oxygen, sodium, magnesium, and aluminum can be explained most easily as a result of deep mixing and proton capture nucleosynthesis, Evidence in support of this explanation is (a) the constancy of the abundances of Al+Mg and C+N+O, (b) the shift in the distribution of sodium abundances toward higher values and of oxygen abundances toward lower values in cluster giants with surface gravities below log g = 1.0. The action of deep mixing, however, is seen not only at the top of the giant branch, but already in some giants even as faint as the level of the horizontal branch. (C) 1996 American Astronomical Society.

The most recent observations of the abundances of oxygen, sodium, magnesium, and aluminum provide strong support for the idea that "deep mixing" occurs within M13 giants. In deep mixing scenarios, material from the envelopes of globular cluster giants is assumed to be mixed well into the hydrogen burning shells of some (but not all) stars. There the abundances of a number of light elements are reshuffled by proton capture before the material is returned to the envelope above. The recent observations also put strong constraints on the conditions under which the reshuffling occurs. Shetrone's (1995, 1996) estimates of magnesium isotope ratios provide especially powerful constraints on the reshuffling temperature for the brightest aluminum-rich M13 giants. We find that proton capture nucleosynthesis can gie excellent account of the abundances of aluminum and the isotopes of magnesium in the brightest aluminum- rich giants of M13 -- and of the abundances of oxygen and sodium as well. In order to match the observations, the abundances in about ninety percent of the material that makes up the envelopes of these stars would have to have been reshuffled by proton capture at a temperature near 70 million kelvins (MK). The maximum temperatures in the hydrogen shells of canonical models for bright M13 giants is only ∼ 55 MK. We briefly discuss ways in which the observations and hte models might be reconciled. Our "best bet" is that hydrogen burning occurs intermittently (at higher temperatures) in some M13 giants rather than steadily (at lower temperatures) as it does in canonical models. We also find that recent rates for the Ne-Na cycle reactions (El Eid & Champange 1995) suggest solutions to two earlier puzzles: the shape of the Na-O abundance anticorrelation observed in globular cluster giants and the Na-N correlation observed in field halo giants.

There is substantial evidence that most of the material in the envelopes of some globular-cluster giants is mixed so deeply into the interior that the abundances of many of the light elements, including magnesium and aluminum, are reshuffled by proton capture before the material is returned to the envelope above. In order to understand these abundance changes we must know all of the appropriate proton-capture rates. We have recalculated the thermonuclear reaction rates, N-A[sigma upsilon](T), for the Mg-24(p,gamma)Al-25 reaction using the canonical 57 temperatures from Caughlan and Fowler (At. Data Nucl. Data Tables, 40, 283, 1988) in the range from T9=0.004 to T9=10.0, where T9 is the temperature in units of 10(9) K. We have done these calculations using a range of values for the unmeasured total width, Gamma(1), of the lowest resonance, (1/2)(+), in the p+Mg-24 system. This resonance is located at 2486 keV excitation in Al-25 and there is a large uncertainty in the value of Gamma(1). The rates currently used are based upon a value of Gamma(1)=75 meV, whereas the most recently accepted experimental value is an upper limit for Gamma(1) of 32 eV. For values of T9<0.07 the rates are quite sensitive to this uncertain width. We discuss the consequences of this uncertainty and the need for experimental data to determine Gamma(1). We have also found that for T9>5.0, the recalculated rates are somewhat higher than those in Caughlan and Fowler (1988). We present our rates in tabular form.

Sodium abundances have been derived for 130 giant stars in the globular cluster M13 from spectra of the Na I doublet at lambda lambda 5682 and 5685 Angstrom obtained using the KPNO 4-m Hydra fiber positioner and bench spectrograph. Magnesium abundances have also been obtained for the brightest 98 stars in the sample from the nearby Mg I line at lambda 5711 Angstrom. The stars observed in M13 range from the faintest at M(upsilon)similar or equal to+1.0 and log g similar or equal to 2.6 up to the tip of the giant branch, and include 18 stars on the asymptotic giant branch. Among the lower luminosity giants, the sodium abundances have a large star-to-star range, approximately from -0.3 less than or equal to[Na/Fe]less than or equal to+0.5. However, the sodium abundances of the most luminous giants (M(upsilon)<-1.7) are usually high; typically, [Na/Fe]greater than or equal to+0.3, with a much smaller star-to-star scatter. The asymptotic giant branch stars have smaller sodium abundances on average than do the red giant branch tip stars. The spread in [Na/Fe] ratios is larger in M13 than it is among halo field giants of comparable metallicity; M13 contains many more stars with high [Na/Fe] ratios than can be found in the field, even at relatively low luminosities on the giant branch. Magnesium is uniformly overabundant ([Mg/Fe]similar or equal to+0.3) in all stars with a low sodium abundance, but the [Mg/Fe] ratio ranges from approximately -0.3 to +0.3 in stars with a high sodium abundance, These sodium and magnesium abundance variations in M13 are discussed in the context of proton capture and deep mixing hypotheses. In addition to the CN and ON hydrogen burning chains previously discussed in the literature, the NeNa and MgAl burning chains have also contributed to the abundance mixture observed in M13 giants. At least some of the products of proton capture chains have been produced in situ in the giants, and brought to the surface, most probably via deep mixing. Evidence in support of the occurrence of proton capture nucleosynthesis and deep mixing among M13 stars includes (1) the absence of sodium-poor stars at the red giant tip, (2) the fact that asymptotic branch stars have lower sodium abundances on average than do stars near the red giant tip, and (3) the existence of a positive correlation between sodium and nitrogen abundances as well as a partial anti-correlation of sodium and magnesium abundances. (C) 1996 American Astronomical Society.

We explore the conditions under which large aluminum enhancements might be produced by very deep mixing and proton-capture nucleosynthesis in bright globular cluster giants. We find that: (1) The initial abundances of neutron-rich magnesium seed nuclei must be surprisingly large for stars that belong to an old, metal-poor population. [Al/Fe] overabundances near 1 .0 dex suggest that the initial 25Mg/24Mg ratio must be almost four times as large as it is in the Sun. If 25Mg and 26Mg isotopes make equal contributions to the aluminum overabundance, both ratios must be about twice as big as they are in the Sun. (2) Cluster giants with large aluminum enhancements (-1.0 dex) produced by very deep mixing should show smaller but observable Mg depletions (-0.2 dex) as a result of the destruction of 25Mg and 26Mg. (3) Large aluminum enhancements are very likely to be accompanied by significant hydrogen depletions and helium enhancements throughout the stellar envelope. Adding hydrogen from the envelope to the hydrogen burning shell may have an observable impact on giant-branch evolution; a helium-enriched envelope will surely have an important impact on the star's later location on the horizontal branch.

view Abstract Citations (100) References (48) Co-Reads Similar Papers Volume Content Graphics Metrics Export Citation NASA/ADS Oxygen Abundances in Halo Giants. VI. M10, A Globular Cluster with Metallicity Similar to M3 and M13 Kraft, Robert P. ; Sneden, Christopher ; Langer, G. E. ; Shetrone, Matthew D. ; Bolte, M. Abstract We have obtained high resolution echelle spectra of 15 bright giants of the intermediate metallicity globular cluster M10, in order to compare the oxygen and sodium abundance variations of this cluster with those of similar metallicity clusters M3 and M13. M10, like M13, has a diagram with an extremely blue horizontal branch and, relative to M3, exhibits the "second parameter effect." As in earlier papers of this series we have first determined abundances of the Fe-peak and α element groups. These element abundances reveal no surprises. For iron, <[Fe/H]> =-1.52±0.02, with little star-to-star scatter (σ=0.07). The α elements are enhanced by typical factors for metal-poor stars, [/Fe]=+0.26±0.04. The Fe-peak elements track Fe quite closely, [/Fe]=-0.08±0.07. As in all other intermediate and low metallicity globular clusters, the sodium and oxygen abundances of M10 giants are anticorrelated. While the general form of the O vs Na anticorrelation trend for M10 is indistinguishable from that defined by M3 and M13, the extent of the O and Na abundance excursions indicates that M10 is an intermediate case between the other two clusters. No M10 giants display the extreme O depletions exhibited by the M13 red giant tip stars, but the average O abundance of M10 is lower than that of M3, and the average Na abundance is higher. This suggests that the variations of O and Na in globular clusters do not by themselves tightly constrain the variations in horizontal branch morphologies of the clusters. Publication: The Astronomical Journal Pub Date: June 1995 DOI: 10.1086/117471 Bibcode: 1995AJ....109.2586K Keywords: GALAXY: HALO; GLOBULAR CLUSTERS: INDIVIDUAL: M10; GALAXY: ABUNDANCES full text sources ADS | data products SIMBAD (22) Related Materials (5) Part 1: 1991AJ....102.2001S Part 2: 1992AJ....104..645K Part 3: 1992AJ....104.2121S Part 4: 1993AJ....106.1490K Part 5: 1994AJ....107.1773S

We present oxygen, sodium, silicon, calcium, scandium, titanium, vanadium, iron, and nickel abundances for bright giants in M71, a relatively metal-rich globular cluster having a space motion like that of the thick disk. The abundances were obtained from analysis of Lick Observatory Hamilton Echelle spectra centered on the [O I] doublet. We find [[Fe/H]] = -0.79 +/- 0.01 from ten giants covering a range of 400 K in effective temperature. Iron peak and traditional a elements show no significant variations from one star to another. Oxygen and sodium appear to vary, but the range is barely significant, and is smaller than found earlier in M92, M13, M3, and M5. Two of the three stars with low oxygen have higher than average sodium abundances, but one low oxygen star also has relatively low sodium. Whether there exists an anticorrelation between [Na/Fe] and [O/Fe] is much less clear in the case of M71 than in the halo clusters studied in earlier papers of this series. For M71 giants, the combination of literature and present results confirms the existence of first dredge-up convective envelope mixing of C-12 --> C-13 and C --> N, but evidence for very deep mixing to O-depleted and Na-enhanced burning layers remains marginal. Primordial abundance fluctuations probably also play a role in the M71 light element abundances. Although most traditional alpha elements are slightly overabundant relative to Fe as expected, M71 giants are unusual in that the mean [Ti/Fe] ratio of +0.5 appears to be quite large, a property shared in common with a sample of giants in the region of the galactic bulge [McWilliam & Rich, preprint (1993)].

Abundances of Y, Ba, Ce, and Nd are presented for giants in the globular clusters M5, M3, M13, M92, and M15. The spectra for these analyses are the same Lick Hamilton echelle spectra that have been used previously to demonstrate the existence of large, anticorrelated abundance variations of oxygen and sodium among giants in each of these clusters. The four heavy elements studied here display no such variations. They correlate neither with Na/Fe nor O/Na abundance ratios, and the observed star-to-star scatters within each cluster are not in excess of observational uncertainties. The abundance ratios among these elements and with respect to the Fe-peak elements agree well with field stars of the same metallicity range. The elements Y, Ba, Ce, and Nd are formed exclusively by neutron capture nucleosynthesis events; thus their near constancy as Na/Fe varies by 0.5-0.8 dex within a cluster, and the constancy of Sc (whose abundance is also sensitive to neutron capture events), places rather severe constraints on possible neutron capture scenarios for the production of Na. This (negative) result strengthens the case for globular cluster Na enhancements via proton capture syntheses.

We have analyzed high resolution echelle spectra of nine bright M13 giants which, when added to those we have analyzed previously by similar techniques, bring our M13 sample size to 22. The sample is 88% complete from the red giant tip (at M(bol)0 = -3.6) to a point one bolometric magnitude fainter, and is presumably representative down to M(bol)0 = -2.1. We find that the brightest M13 giants are predominantly super oxygen poor ([O/Fe] approximately - 0.4 to - 0. 8) and correspondingly sodium rich ([Na/Fe] approximately +0.2 to +0.4). We argue that these super 0-poor stars are ''first ascent'' giants. The most straightforward way to understand (1) the decline in the average oxygen abundance as M13 stars approach the tip of the giant branch and (2) their anticorrelated oxygen and sodium abundances, is that they are the results of very deep mixing. In the case of the super 0-poor stars this idea requires that 90 percent of the oxygen atoms of the envelope be transformed into nitrogen. Since there are significant variations in the oxygen abundances of stars at the same point on the giant branch, such mixing must vary stochastically from star to star. Variable deep mixing is most likely to be the result of variable internal rotation. We have added Ni and the alpha elements Si, Ca, and Ti to our list of abundance determinations for all 22 M13 giants and seven giants in M3. We find: [Fe/H] is identical for the two clusters, the mean values of [Sc/Fe], [V/Fel, and [Ni/Fe] are close to zero for both clusters, and the alpha elements are overabundant by similar amounts ([el/Fe] approximately 0.2 to 0.3) in all stars. Despite the wide ranges in the observed 0 (greater-than-or-equal-to 1.0 dex) and Na (approximately 0. 8 dex) abundances, we find no correlation between the abundances of 0 or Na and the abundances of any of these other elements.