Globular clusters: a chemical roadmap between anomalies and homogeneity

0.19 MB·

by  A. Mucciarelli

Mem.S.A.It.Vol.75,282
(cid:13)c SAIt 2008 Memoriedella
Globular clusters: a chemical roadmap between
anomalies and homogeneity
A. Mucciarelli1
4
1
0 DipartimentodiFisica&Astronomia–Universita´degliStudidiBologna,VialeBertiPichat
2 6/2,I-40127Bologna,Italye-mail:[email protected]

n
a
J
Abstract.

7 For several decades, globular clusters have been considered the best example of simple
1 stellar populations, hosting coeval and chemical homogeneous stars. The last decade of
spectroscopicandphotometricstudieshasrevealedamorecomplexviewoftheirchemical

] composition,withahighlevelofhomogeneityintheirironcontentbutstar-to-starvariations
R
insomelightelements.Thiscontributionsummarizesthemainevidenceaboutthechemi-
S calanomaliesinthestellarcontentoftheglobularclusters,discussingalsosomepeculiar
. objectswithintrinsicdispersionsintheirironcontent.
h
p
Keywords.Stars:abundances–Stars:atmospheres–Stars:PopulationII–Galaxy:glob-
-
o ularclusters
r
t
s
a 1. Introduction has conclusively established that the stars in
[
a GC do not share the same initial chemical
Globularclusters(GCs)areusuallyconsidered
1 composition. In fact: (i) some elements have
as the best example available in the nature of
v been observed to vary in all the GCs, like C,
3 simple stellar populations (SSPs), aggregates N,O,NaandAl;(ii)someelementshavebeen
2 of stars with the same age and initial chem- observed to vary only in some GCs, like He,
3 ical composition (see e.g. the seminal paper
Li, Mg, Si, K; (iii) finally, there is a bunchof
4 by Renzini & Buzzoni 1986). GCs as SSPs
strangebeastscharacterizedbya moreor less
. are valuable tools to approach several aspects
1 pronouncedFestar-to-starvariations.
0 of the astrophysics: they allow to check the
4 predictions of the stellar evolution theory, to
1 study the chemical enrichment history of the As a generalgoldenrulewe can define as
v: parentgalaxy,andtoinvestigatetheproperties genuine GCs all the massive, stellar systems
i of unresolved stellar populations(because for thatarehomogeneousintheirFe(andFe-peak)
X GCs in theMilkyWay andinnearbygalaxies content. Generally, all the stars in a GC have
r wehavesimultaneouslytheresolvedandinte- thesameironcontent.Ironisproducedbyboth
a
gratedinformation). Type II and Type Ia Supernovae (SN). Thus,
A large number of photometric and spec- stellar systems with an intrinsic Fe dispersion
troscopicevidencecollectedinthelastdecades intheirstellarcontentdidretaintheSNejecta
in their gravitational well, while systems like
Sendoffprintrequeststo:A.Mucciarelli thegenuineGCsdidnotretaintheseejecta.
Mucciarelli:Globularclusters 283
Willman & Strader (2012) performed a normalized metallicity distributions of the 5
comparison between the intrinsic Fe spread beastsareshowninFig.1andbrieflyexplained
(calculated by taking into account the uncer- asfollows:
tainties in each individual star) of GCs and (1) Omega Centauri — The most famous
other stellar systems, like dwarf spheroidal case of GC-like system with a intrinsic Fe
galaxiesandultra-faintdwarfs,findingthatthe variation is Omega Centauri, whose metallic-
GCs are characterized by intrinsic Fe spreads itydistributionhasbeeninvestigatedbyseveral
smaller than 0.05 dex, while dwarf systems authors (see e.g. Freeman & Rodgers 1975;
have Fe spread larger than 0.2 dex. Thus, the Pancino et a. 2002; Johnson & Pilachowski
level of homogeneity in the Fe content repre- 2010; Marino et al. 2011). Its metallicity dis-
sents the main chemical fingerprint to distin- tribution is very large (covering up to 1.5
guish GCs from other (more complex) stellar dex) and multi-modal, with at least 5 peaks.
systems.Obviously,thisdefinitiondoesnotal- However, the most popular scenario is that
lowtodistinguishbetweenGCsandlessmas- Omega Centauri is not a genuine GC but it is
sivestellarsystemswithoutFespread,likethe the remnant core of a tidally disrupted dwarf
open clusters. In this context, it is important galaxy.
torecallthatalltheGCsproperlyobservedso (2) Terzan 5 — Another case of mult-
farwithlargesamplesofhigh-resolutionspec- modal and broad Fe distribution is Terzan 5,
tra exhibit also star-to-star variations among a GC locatedin theinnerbulgeand knownto
the light elements (C, N, O, Na, Mg and Al), harbor two red clumps in its color-magnitude
known since forty years ago. Even if the pre- diagram (Ferraro et al. 2009). These two
cise mechanism able to produce this kind of red clumps are associated to two different
chemical variations are still debated and un- Fe abundances at about [Fe/H]=–0.30 and
der scrutiny (see Section 3.1), these chemical [Fe/H]=+0.30 dex (Ferraro et al. 2009). The
anomalies are observed only in GCs (and not same Fe difference has been observed among
in openclustersorinfield stars),andtheyare the giant stars (Origlia et al. 2011), with the
consideredastypicalfeatureoftheGCs. detection of an additional, third component
at [Fe/H]∼–0.8 dex (Origlia et al. 2013). The
metallicitydistributionofTerzan5turnsoutto
2. Strangebeasts
be very large (about 1.5 dex) with at least 3
Uptonow,only5(outof∼150)GalacticGCs distinctpeaks(see also the contributionby D.
exhibit intrinsic Fe spreads, namely Omega Massaritothisconference).Thestrikingchem-
Centauri,Terzan5,M54,M22andNGC1851. icalsimilaritybetweenTerzan5andthebulge
Throughout this contribution, I refer only to seems to suggest that Terzan 5 could be the
the GCs forwhichchemicalabundancesfrom remnant of one of the pristine fragment that
high-resolution spectroscopy and large sam- contributedtoformtheGalacticbulge(Ferraro
ple of stars are available. Other 2 GCs have etal.2009).
been proposed to have a Fe spread, namely (3) M54 — A different case is provided
NGC 5824 and NGC 3201. The analysis of by M54, a massive GC immersed in the nu-
NGC 5824presentedby Saviane et al. (2012) cleus of the Sagittarius remnant. Carretta et
is basedontheCa II triplet(asaproxyofthe al. (2010a)analyzedwith FLAMES@VLT76
metallicity)butchemicalanalysisbasedonthe stars of M54, finding a broad (∼0.7 dex) but
directmeasurementofFelinesarestilllacking. uni-modaldistribution.
Recently,Simmereretal.(2013)foundanap- (4)M22—Thisclusterhasbeensuspected
preciablespreadamongthestarsofNGC3201, to harbor an intrinsic Fe spread since thirty
by using high-resolution spectra of 24 giants. years ago (Pilachowski et al. 1982) even if
However,otheranalysisalwaysbasedonhigh- other analysis have ruled out such a variation
resolutionspectra(Carrettaetal.2009;Munoz (Cohen1981;Gratton1982).Recently,Marino
et al. 2013) seem to contradict this finding, et al. (2009) and Marino et al. (2011) per-
withoutclearhintsofintrinsicFespreads.The formedthe analysisof high-resolutionspectra
284 Mucciarelli:Globularclusters
for a totalof35 starsof M22,findinga broad stars.CN-strong/CH-weakstars,observedonly
but unimodalmetallicity distribution (ranging in GCs, can be interpreted as having some
from[Fe/H]∼–2dexto∼–1.6dex).Also,M22 amountofCNOprocessedmaterialintheirat-
isknowntoharborasplitatthelevelofthesub- mospheres. The use of high-resolution spec-
giantbranch,with two distinctbranchesasso- troscopyforgiantstarsinGCs hasallowedto
ciatedtodifferentabundancesofs-processele- link these CN/CH anomalies to other chemi-
ments(Marinoetal.2012). cal anomalies. CN strength is correlated with
(5)NGC1851—Based on the analysisof Na and Al (Cottrell & Da Costa 1981) and
124stars,Carrettaetal.(2010b)claimasmall anti-correlated with O (Sneden et al. 1992).
intrinsic Fe dispersion for this cluster. When Also, anti-correlationsbetween O and Na ex-
the metallicity distribution is roughly divided ist,asdiscoveredbytheLick-Texasgroup(see
intwogroups,themetal-poorstarsturnoutto Fig.16inIvansetal.2001,forasummary).
bemorecentrallyconcentratedwithrespectto Currently,theNa-Oanti-correlationisrec-
the metal-richones, suggesting a kinematical ognized as a typical feature of all the old and
differencebetweenthetwo groupsofstars(in massive GCs, as widely demonstrated by the
fact, Carretta et al. 2010b, proposed a merg- homogeneous survey of more than 1000 gi-
ing as possible origin of NGC 1851). Note ants in 19 GCs performed by Carretta et al.
that otherworksdid notfoundhintsof intrin- (2009). These anomalies have been observed
sic inhomogeneity (Yong & Grundahl 2008; alsoinold,massiveextra-galacticGCs,likein
Villanovaetal.2010)evenifbasedonsmaller Fornax (Letarte et al. 2006) and in the Large
samples. Also, the analysis by Willman & MagellanicCloud(Mucciarellietal.2009).
Strader (2012) suggests that the observed Fe A similar feature, often observed in some
spread is fully compatible with those in other GCs, is the anti-correlation between Al and
GCs.LikeM22,alsoNGC1851exhibitsasplit Mg. The label of ”anti-correlation” could be
initssub-giantbranch,likelyexplainablewith improper, because most of the GCs show a
adifferenceinthetotalC+N+O.TheC+N+O large Al spread coupled with an unique value
in NGC 1851 is still a controversialand open of Mg (in a given cluster), while the Mg-
issue. Yong & Grundahl (2008) measured a poor stars seem to be very rare. Up to now,
spreadintheC+N+Ocontentofabout0.6dex GC stars with [Mg/Fe]<0 have been detected
among4brightgiantsinNGC1851.Thesere- only in NGC 2808 (Carretta et al. 2009),
sults have not been confirmedby the analysis in NGC 1786 (Mucciarelli et al. 2009) and
of15giantsbyVillanovaetal.(2010),finding NGC2419(Mucciarellietal.2012).
nodifferencesintheirC+N+O. All the spectroscopic evidence collected
so far (i.e. CN-CH anti-correlation, CN bi-
modality, Na-O anti-correlation, Mg-Al anti-
3. Elementsthatvary
correlation) are commonly interpreted as the
signature of material processed through the
3.1.C,N,O,Na,Mg,Al
high temperature extension of the proton-
The first evidence of inhomogeneity in the capture reactions (like NeNa and MgAl cy-
chemical content of GCs has been provided cles). Several theoretical models have been
from the study of CN and CH features in the proposedinordertodescribetheformationand
brightest giants by using low-resolutionspec- early evolution of GCs, for instance D’Ercole
troscopy. Osborn (1971) detected for the first et al. (2008), Decressin et al. (2010), Bekki
time 2 CN-strong stars in M5 and M10, and et al. (2011), Conroy & Spergel (2011) and
following studies (Norris et al. 1981; Smith Valcarce & Catelan (2011). However, the ba-
& Norris 1982; Briley et al. 1993; Martell & sicideabehindthesemodelsisthattheclusters
Smith2009;Pancinoetal.2010)haverevealed formedwithachemicalcompositionthatwell
thattheGCsshowabimodalityintheirCNab- resemblesthatobservedinthefieldstars.Then,
sorption and anti-correlationbetween CN and they experiencea periodof star formationac-
CH strengths, both among giant and dwarf tivity,withnewstarsbornfromagaspolluted
Mucciarelli:Globularclusters 285
by ejecta from the first generation stars that NGC 2808(Piotto etal. 2007;Dalessandroet
have processed material through the hot H- al.2011),NGC2419(diCriscienzoetal.2011)
burning.Differentkindsof polluterstars have andNGC6397(Miloneetal.2012).
beenproposed,i.e.theasymptoticgiantbranch The direct measurement of the He abun-
(AGB)starsandthefast-rotatingmassivestars. dance in GC stars is a quite hard task, first
Inbothcases,ashort(<100Myr)phaseofstar- of all because of the small number of avail-
formationshouldoccurintheearlystageofthe able He diagnostics. The chromospheric line
lifeofGCs. at 10820Å can be observedin giant stars but
Generally speaking, we used to separate it needs very high signal-to-noise and resolu-
thestarsinaGCintwoclasses,afirstgenera- tion spectra, and it is also heavily sensitive to
tionincludingstarswithchemicalabundances the adopted modeling of the stellar chromo-
similartothoseobservedinfieldstarsofsimi- sphere.ThislinehasbeenusedtoinfertheHe
larmetallicity(thus,Na-poor/O-rich/CN-weak abundance in giant stars of Omega Centauri
stars), and a secondgenerationincluding Na- (Dupree et al. 2011; Dupree & Avrett 2013)
rich/O-poor/CN-strongstars.Obviously,thisis and NGC 2808 (Pasquini et al. 2011). The
abrutalbutefficientclassification.Considering latter provided a differential analysis between
our current knowledge and understanding of twogiantstarsofNGC2808withdifferentNa
the GCs, we cannot exclude that the star- abundances.Theiranalysispointsoutadiffer-
formation in a GC occurs in a continuos way enceinYbetweenthetwostarsofatleast0.17,
and not with discrete bursts (as the classifica- withtheNa-richstarsbeingmoreHeenriched
tion in first and second generations seems to thantheNa-poorones.
suggest). This scheme is contradicted by the Only a few photospheric lines can be
Na-O anticorrelation, that appears to be con- detected among the horizontal branch stars,
tinuous (with the only exception of M4, see with effective temperaturesbetween ∼9000K
Marinoetal. 2008),whileotherfindings(like and ∼12000 K (the latter corresponding to
the CN-bimodabilityand the red giant branch the Grundahl Jump, Grundahl et al. 1999).
splitting observed in the color-magnitudedia- Villanova et al. (2009) performed a chem-
gramsincludingtheU-bandfilter)seemtosup- ical analysis in 4 horizontal branch stars
portthisscenario. in NGC 6752, finding an average value
Note that none of the current models are <Y>=0.25,withvaluesof[O/Fe]and[Na/Fe]
able to fully reproducethe observedchemical compatible with those observed in the so-
patterns,requiringsomefinetuninginmostof called first cluster generation. Also, the anal-
themainparameters.Currently,anumberofis- ysis of 6 HB stars in M4 by Villanova et
suesremainopenandunsolved,likethenature al. (2012) provides a higher He abundance,
ofthepolluters,theneedofdilutionofpollut- with an average value <Y>= 0.29, coupled
ing gaswith unprocessedmaterialor the rela- with high values of [Na/Fe] and low val-
tive fraction between first and second genera- ues of [O/Fe]. Recently, Marino et al. (2013)
tionstars. measured the He abundance in 17 horizon-
tal branch stars of NGC 2808, finding an av-
erage value of <Y>= 0.34. These first find-
3.2.Helium
ings strongly support the scenario where sec-
The CN-strong and Na-rich/O-poor stars are ondgenerationstarsarealsoenrichedinHe.
expected to be also enrichedin helium, what-
ever the polluter stars are, being helium the
3.3.Lithium
main product of the H-burning. The occur-
rence of (mild or strong) enhancement in Y Lithiumremainsakey-elementinthestudyof
(upto Y∼0.4)hasbeenproposedas thecause the formation and evolution of GCs but also
of the main sequence splitting and the pe- an unresolved riddle. It is destroyed at tem-
culiar horizontal branch morphology in the peratures of about 2.5·106 K, thus it cannot
cases of Omega Centauri (Bedin et al. 2004), survive at the typical temperatures of the hot
286 Mucciarelli:Globularclusters
H-burning, larger than ∼107 K. A reasonable areunusualforGCstarsbuttheseMg-poor/K-
expectationis thatthe secondgenerationstars rich stars represent∼40%of the studied sam-
should be depleted in Li. Also, some kind of ples. Also, a Mg-K anti-correlation is clearly
correlations/anti-correlations between Li and detected, suggesting that these anomalies can
the elementsinvolvedin the chemicalanoma- be explained within the self-enrichment sce-
lies are expected, in particular Li-Na anticor- nario.Interestinglyenough,thefractionofMg-
relations and Li-O correlations. Thus, the Li poor stars well resembles that proposedby di
abundancecouldrepresentaformidablesmok- Criscienzoetal.(2011)fortheextremeHe-rich
ingguntodisentanglethenatureofthepolluter populationwithY=0.4,accordingtothemor-
starsabletocreatethesecondgenerationstars. phologyofthehorizontalbranchofNGC2419.
Theobservationalevidencecollectedsofar Ventura et al. (2012) identified the Mg-
providesusascenarionoteasytoexplain,giv- poor, K-rich stars as the signature of an ex-
ing us more questions and doubts than an- treme nucleosynthesis driven by AGB and
swers. Some GCs show clear or weak hints super-AGB stars, because K can be produced
of Li spread and anti/correlations with Na or by proton capture on Argonnuclei during the
O, like 47 Tucanae (Bonifacio et al. 2007), normal nuclear reactions that create the ob-
NGC6752(Shenetal.2010),NGC6397(Lind servedchemicalanomalies.
etal.2009)andM4(Monacoetal.2012).On The first investigations about possible K
theotherhand,otherworksseemtoruleoutLi spreadsamongthestarsofotherGCshavenot
spreads,like in M4(D’Orazi& Marino2010; shown chemical patterns similar to those ob-
D’Orazietal.2010;Mucciarellietal.2011). servedinNGC2419(Carrettaetal.2013).
However,the small variationof Li associ-
ated in some cases to large variations of Na
4. Somethoughtsaboutglobular
and/orOremainsanopenissue,andthecurrent
clusters
theoretical models are not able to reproduce
this finding. The models where the chemical The deep investigations about the chemical
anomaliesaredrivenbyAGBstarsareableto composition in the GCs performed in the last
explainaLiproductionthroughtheCameron- decade (and gathered by the massive use of
Fowlermechanism,butitneedsahighdegree high-resolution spectrographs) has for some
of fine-tuning,in orderto produceexactlythe aspects changed our view of the GCs, unveil-
same amount of Li previouslydestroyed. The ingaquitecomplexformationprocess.
modelsbasedonfast-rotatingmassivestarsdo Only to provide a simple, mental scheme
notincludeLiproductionmechanisms,invok- toputsomeorderamongthisevidence,wecan
ing dilution processes to partially explain the dividetheGCsinthreeclasses:(1)thegenuine
very similar Li abundances between first and GCs, characterized by homogeneity in their
secondgenerations. Fe content, that show a uni-modal and nar-
row metallicity distribution; (2) the GCs with
large but uni-modal metallicity distribution,
3.4.Potassium
like M54, M22 and NGC1851, even if an en-
Potassium is anewentryamong the elements largeof their studiedsamplesis mandatoryin
observedtovaryonlyinsomeGCs.Infact,the ordertounveilpossiblesecondarypeaks;(3)a
only cluster observedso far harboringa large thirdgroupofsystems,includingonlyOmega
dispersion of K is NGC 2419 (Mucciarelli et Centauri and Terzan 5 with broad and multi-
al. 2012; Cohen & Kirby 2012). This clus- modal metallicity distribution, that cannot be
ter shows two distinct groups of stars, the considered genuine GCs, but stellar systems
firstcharacterizedbynormalvaluesof[Mg/Fe] undergoingcomplexchemicalenrichmenthis-
and[K/Fe](compatiblewiththoseobservedin tories.
otherGCs),thesecondwithextremevaluesfor Can we continue to use genuine GCs as
boththeabundanceratios,reaching[Mg/Fe]∼– simple stellar populations, in spite of their
1.4 dex and [K/Fe]∼+2.0 dex. These values chemical anomalies and the proposed self-
Mucciarelli:Globularclusters 287
enrichment scenario? A reasonable answer is Carretta, E., Gratton, R. G., Bragaglia, A.,
yes(butwithsomecautions).Infact:(1)They D’Orazi, V., Lucatello, S., Sollima, A., &
are homogeneous in Fe and almost any other Sneden,C.,2013,ApJ,769,40
elements, with the exceptionsof C, N, O, Na, Cohen,J.G.,1981,ApJ,247,869
MgandAl.Thuswecancontinuetousethem Cohen,J. G.,&Kirby,E.N.,2012,ApJ,760,
as tracers of the chemical composition of the 86
host galaxy, by focusing our attention on the Conroy,C.,&Spergel,D.N.,2011,ApJ,726,
elements (the majority) that do not show in- 36
trinsicstar-to-starvariations.(2)Theyarestill Cottrell, P. L., & Da Costa, G. S., 1981, ApJ,
single-age stellar systems. The star formation 245L,79
in the GCs occurs within short timescales (< Dalessandro, E., Salaris, M., Ferraro, F. R.,
100Myr).Thistimescaleismuchsmallerthan Cassisi, S., Lanzoni, B., Rood, R. T., Fusi
the GC age and the age differences between Pecci,F.,&Sabbi,E.,2011,MNRAS,410,
GC sub-populations with different chemical 694
compositions(intermsoflightelements)can- Decressin,T.,Baumgardt,H.,Charbonnel,C.,
not be appreciated at the level of the turnoff. &Kroupa,P.,2010,A&A,516,73
(3)Theirintegratedcolorscanbe still usedto D’Ercole, A., Vesperini, E., D’Antona, F.,
calibrate and study unresolved stellar popula- McMillan, S. L. W., & Recchi, S., 2008,
tions, like the GCs outside the Local Group, MNRAS,391,825
because the effect of the chemical anomalies di Criscienzo, M., D’Antona, F., Milone,
ontheintegratedcolorsisnegligible(withthe A. P., Ventura, P., Caloi, V., Carini, R.,
onlycautiontoexcludecolorsincludingUfil- D’Ercole, A., Vesperini, E. & Piotto, G.,
ters,heavilyaffectedbyCandNvariations,see 2011,MNRAS,414,3381
Sbordoneetal.2011). D’Orazi,V.,&Marino,A.F.,2010,ApJ,716L,
AretheGCsstrictlyspeakingsimplestellar 166
populations? The answer is no, but they have D’Orazi, V., Lucatello, S., Gratton, R. G.,
neverbeenconsideredso.AretheGCsthesim- Bragaglia, A., & Carretta, E., 2010, ApJ,
plestsimplestellarpopulationsavailableinthe 713L,1
Universe?Theanswerisclearlyyes,whatever Dupree, A. K. , Strader, J., & Smith, G. H.,
complextheirformationscenariois. 2011,ApJ,728,155
Dupree, A. K., & Avrett, E. H., 2013, ApJ,
Acknowledgements. This research is part of the
773L,28
project COSMIC-LAB (web site: www.cosmic-
Ferraro, F. R., Dalessandro, E., Mucciarelli,
lab.eu) funded by the European Research Council
A., Beccari, G., Rich, R. M., Origlia,
(undercontractERC-2010-AdG-267675).
L., Lanzoni, B., Rood, R. T., Valenti, E.,
Bellazzini,M.,Ransom,S.M.,&Cocozza,
References
G.,2009,Nature,462,483
Freeman,K.C.,&Rodgers,A.W.,1975,ApJ,
Bedin,R.L.,Piotto,G.,Anderson,J.,Cassisi,
201L,71
S., King,I.R., Momany,Y.,& Carraro,G.,
Gratton,R.G.,1982,A&A,115,171
2004,ApJ,605L,125
Grundahl, F., Catelan, M., Landsman, W. B.,
Bekki,K.,2011,MNRAS,412,2241
Stetson,P.B.,&Andersen,M.I.,1999,ApJ,
Briley, M. M.., Smith, G. H., Hesser, J. E., &
524,242
Bell,R.A.,1993,AJ,106,142
Johnson, C. I., & Pilachowski, C. A., 2010,
Bonifacio, P., Pasquini, L., Molaro, P.,
ApJ,722,1373
Carretta, E., Francois, P., Gratton, R. G.,
Ivans.I.I.,Kraft,R. P.,Sneden,C.,Smith,G.
James, G., Sbordone, L., Spite, F., &
H.,Rich, R. M.,&Shetrone,M.,2001,AJ,
Zoccali,M.,2007,A&A,470153
122,1438
Carretta,E.etal.,2009,A&A,505,117
Letarte, B., Hill, V., Jablonka, P., Tolstoy, E.,
Carretta,E.,etal.,2010,A&A,520,95
Francois, P., & Meylan, G., 2006, A&A,
Carretta,E.,etal.,2010,ApJ,722L,1
288 Mucciarelli:Globularclusters
453,547 Piotto,G.,Bedin,L.R.,Anderson,J.,King,I.
Lind,K.Primas,F.,Charbonnel,C.,Grundahl, R., Cassisi, S., Milone,A. P., Villanova,S.,
F.,&Asplund,M.,2009,A&A,503,545 Pietrinferni, A., & Renzini, A., 2007, ApJ,
Marino,A.F.,etal.,2011,ApJ,731,64 661L,53
Martell, S. L., & Smith, G. H., 2009, PASP, Renzini, A. & Buzzoni, A., 1986, Spectral
121,577 EvolutionofGalaxies,122,195
Milone,A.P.,Marino,A.F.,Piotto,G.,Bedin, Saviane, I., da Costa, G. S., Held, E. V.,
L.R.,Anderson,J.,Aparicio,A.,Cassisi,S., Sommariva,V.,Gullieuszik,M.,Barbuy,B.,
&Rich,R.M.,2012,ApJ,745,27 &Ortolani,S.,2012,A&A,540,27
Marino,A.F.,Villanova,S.,Piotto,G.,Milone, Sbordone,L.,Salaris,M.,Weiss,A.,&Cassisi,
A.P.,Momany,Y.,Bedin,L.R.,&Medling, S.,2011,A&A,534,9
A.M.,2008,A&A,490,625 Shen, Z.-X., Bonifacio, P., Pasquini, L., &
Marino, A. F., Milone, A. P., Piotto, G., Zaggia,S.,2010,A&A,524L,2
Villanova, S., Bedin, L. R., Bellini, A., & Simmerer, J., Ivans, I. I., Filler, D., Francois,
Renzini,A.,2009,A&A,505,1099 P.,Charbonnel,C.,Monier,R.,&James,G.,
Marino,A.F.,etal.,2011,A&A,532,8 2013,ApJ,764,7L
Marino,A.F.,etal.,2012,A&A,541,15 Smith,G.H.,&Norris,J.,1982,ApJ,254,594
Marino,A.F.,etal.,2013,MNRAS.tmp.2693 Sneden, C., Kraft, R. P., Prosser, C. F., &
Monaco, L., Villanova, S., Bonifacio, P., Langer,G.E.,1992,AJ,104,2121
Caffau, E., Geisler, D., Marconi, G., Valcarce,A.A.R.,&Catelan,M.,2011,A&A,
Momany,Y.,&Ludwig,H.-G.,2012,A&A, 533,120
539,157 Ventura, P., D’Antona, F., Di Criscienzo, M.,
Mucciarelli, A., Origlia, L., Ferraro, F. R., & Carini, R., D’Ercole, A., & Vesperini, E.,
Pancino,E.,2009,ApJ,695,134 2012,ApJ,761L,30
Mucciarelli, A., Salaris, M., Lovisi, L., Villanova, S., Piotto, G., & Gratton, R. G.,
Ferraro,F.R.,Lanzoni,B.,Lucatello,S.,& 2009,A&A,499,755
Gratton,R.G.,2011,MNRAS,412,81 Villanova, S., Geisler, D., & Piotto, G., 2010,
Mucciarelli, A., Bellazzini, M., Ibata, R., ApJ,722L,18
Merle,T.,Chapman,S.C.,Dalessandro,E., Villanova, S., Geisler, D., Piotto, G., &
&Sollima,A.,2012,MNRAS,426.2889 Gratton,R.G.,2012,ApJ,748,62
Munoz,C.,Geisler,D.,&Villanova,S.,2013, Yong,D.&Grundahl,F.,2008,ApJ,672L,29
MNRAS,433,2006 Willman,B.&Strader,J.,2012,AJ,144,76
Norris,J.,Cottrell,P.L.,Freeman,K.C.,&Da
Costa,G.S.,1981,ApJ244,205
Origlia, L., Rich, R. M., Ferraro, F. R.,
Lanzoni, B., Bellazzini, M., Dalessandro,
E., Mucciarelli, A., Valenti, E., & Beccari,
G.,2011,ApJ,726L,20
Origlia, L., Massari, D., Rich, R. M.,
Mucciarelli,A.,Ferraro,F.R.,Dalessandro,
E.,&Lanzoni,B.,2013,ApJ,779L,5
Osborn,W.,1971,Obs,91,223
Pancino, E., Pasquini, L., Hill, V., Ferraro, F.
R.,&Bellazzini,M.,2002,ApJ,568L,101
Pancino, E., Rejkuba, M., Zoccali, M., &
Carrera,R.,2010,A&A,524,44
Pasquini, L., Mauas, P., Kaufl, H. U., &
Cacciari,C.,2011,A&A,531,35
Pilachowski, C., Leep, E. M, Wallerstein, G.,
&Peterson,R.C.,1982,ApJ,263,187
Mucciarelli:Globularclusters 289
Fig.1.Normalizedmetallicitydistributionforthe5GC-likesystemswithanintrinsicFespread:Omega
Centauri (Johnson &Pilachowski 2010), Terzan5(Massariet al.,inprep.),M54(Carrettaetal.2010a),
M22(Marinoetal.2009,2011)andNGC1851(Carrettaetal.2010b).