Fossil G roups Groups Fossil? Groups? Renato Dupke

Transcription

1 Fossil Groups Fossil? Groups? Renato Dupke Collaborators: Eric Miller (MIT) Claudia Mendes de Oliveira, Rob Proctor, Raimundo Lopes de Oliveira, (IAG-USP), Pieter Westera (National Observatory Brazil, UFABC) Eli Rykoff (UCSB), Jimmy Irwin (Univ. Alabama) + others

2 OUTLINE 1- Motivation 2-Groups? 3- Old? 4-Cool cores (or absence thereof) 5- A check for winds 6-New samples

3 Fossil (old) Groups Dominated by a cd 2 magdifference between the 1 st &2 nd rank galaxies (in R-band) within 0.5 r 200. Extended X-ray emission (L X, bol >10 42 h erg/s). Discovered > 15 years ago (Ponman et al. 1994) Originally thought to be the cannibalistic remains of galaxy groups that lost energy through dynamical friction (e.g. Mulchaey & Zabludoff 1999). Expected large times involved in dynamical friction and the observed lack of X-ray substructures imply that FGs formed early and were undisturbed for a very long time (Ponman et al. 1994; Jones et al. 2000; Vikhlinin et al. 1999).

4 GROUPS? Khosroshahi et al Mendes de Oliveira et al confirm high σ s with many more galaxies Group Nspec σ <T> L X M dyn Name Km/s kev erg/s M sun RX J (25) 773±214(750±150) 2.63± RX J ± ± RX J (22) 419±187(565±77) 1.16± RX J (25) 694±120(584±121) 4.0± RX J (22) 721±150(635±164) 2.85± NGC ± ± ESO ± ±

5 T X of the FG s IGM is more similar to that of clusters, sometimes in excess of 4 kev (e.g. RX J ; Khosroshahietal. 2006) - possible 6.5±1.4 kev Cheshire Cat! z~0.42 (Irwin & Dupke in prep.) z=0.426 & Lensed at 0.97 e 1.4. Very concentrated. M(<62kpc) ~ 3.3x10 13 M sun Poor (25 galaxies) group. ΔV>1100km/s, possibly unbound or >14Mpc los FGs. If one of the BGs is a fly-by, extended X-ray halo centered on only one BG If FGs, separated by 14 Mpc, then two separate regular X-ray halos around each BG. if 2 FGs interacting, X-ray irregular.

6 T X of the FG s IGM is more similar to that of clusters, sometimes in excess of 4 kev (e.g. RX J ; Khosroshahi et al. 2006) - possible 6.5±1.4 kev Chestshire Cat! z~0.42 (Irwin & Dupke in prep.) Answer: None of the above Answer: None of the above. X-center in between the 2 BGs, smooth, regular - L X ~4.4x10 44 erg/s Given the high T, collision + BG displacement?

7 σ in FGs consistent with T X measured, at least for the few FGs with relatively good X-ray data (Mendes de Oliveira 2006; Cypriano et al. 2006; Mendes de Oliveira et al. 2006, 2009). Not?atypical? location in the L X -T X relation (e.g. Khosroshahi et al Fig1). deeper gravitational potential wells, more typical of poor clusters. L X -T X relation for FGs, from Khosroshahi et al Clusters are red+blue, groups are green and FGs are the data points.

8 σ in FGs consistent with T X measured, at least for the few FGs with relatively good X-ray data (Mendes de Oliveira 2006; Cypriano et al. 2006; Mendes de Oliveira et al. 2006, 2009). Not?atypical? location in the L X -T X relation (e.g. Khosroshahi et al Fig1). L X -T X relation for FGs, from Khosroshahi et al Clusters are red+blue, groups are green and FGs are the data points. deeper gravitational ti potential ti wells, more typical of poor clusters.

9 Are They Old? X-ray derived mass profiles high values of the concentration ti parameter c 200. Given the correlation found between c 200 and formation epoch in N-body simulations of ΛCDM cosmologies (Wechsler et al. 2002), FGs should be very old (z formation >1.5). 3 FG with resolved T(r) & 2 OLEGs (diamonds) are compared with non-fgs clusters (open squares). From Khosroshahi, Ponman & Jones (2007). Δm 12,R of each simulated FG as a function of formation time defined as the epoch in which the group assembled 50% of fthe system s final mass Recent numerical+hydro simulations correlation between formation epoch and Δmag, the older groups having higher magnitude differences (e.g. D onghia et al. 2005). The latter suggest a typical FG formation age of Gyr (0.75<z<1.3) as opposed to regular groups (~ Gyr).

10 Are They Old? X-ray derived mass profiles high values of the concentration ti parameter c 200. Given the correlation found between c 200 and formation epoch in N-body simulations of ΛCDM cosmologies (Wechsler et al. 2002), FGs should be very old (z formation >1.5). NFW+STARS fittings for NGC6482 (Humphrey et al. 2006). Adding the stellar mass component allows the DM halos to be substantially less concentrated, since less DM is required in the center of the halo. Much better agreement with the results of N-body simulations than those obtained with the NFW model. There is still a slight trend toward more concentrated halos than CDM.

11 6/43 fossil groups are selected: NGC 741 ESO RXJ 6/43 fossil groups are selected: NGC 741, ESO , RXJ , NGC 3402, ESO and ESO

12 Inconsitencies: Are They Really Old? The cooling time of FGs is observed to be significantly below the Hubble time (e.g., RX J , ESO , Sun et al. (2004); NGC 6482, Khosroshahi et al. (2004, 2006), but they typically lack the expected large well-defined cooling core profiles, indicating a more recent formation time. NGC 6482, Khosroshahi et al. (2004)

13 However, some show mini cool cores ESO Sun et al. (2004)

14 What should we expect for old and undisturbed? radiative cooling only. Ettori & Brighenti 2008

15 Isophotal shapes of earlytype brightest group or cluster galaxies, Fossil groups (crosses), early-type yp BGGs(triangles) and BCGs (circles). From Khosroshahi, Ponman & Jones (2006). disky isophotes. secondary gas infall (Khochfar & Burkert 2005), Spirals took part in the merging. star formation bursts (e.g. van Dokkum et al. 1999), metal rich SNII-driven galactic winds or superwinds (e.g. Strickland et al. 2004, Heckman et al. 1990). deposit metals and energy into the central gas. change the chemistry of the IGM

16 SN Ia/SN II pollution in the ICM shows often radial gradients, SN Ia/SN II ratio is higher in the central region Rasmussen & Ponman 2007 [Si/Fe] vs. radius. Gray lines represent groups, and black lines represent clusters. From Finoguenov, David & Ponman SN II-powered protogalactic winds will tend to disperse metals into the ICM Ram-pressure stripping will distribute the SN Ia polluted ejecta in a more centralized way and is a slower and continuous process that should the central SN Ia Fe mass fraction with time.

17 The Sample

18 Individual Profiles

19 Joint Results Improved statistics gained by joining all FGs shows a general central enhancement of α-elements and Fe, suggesting an central decline of SN Ia dominance. Error-weighted average over all the ratios shown, we find that, despite the overall dominance by SN Ia of ~99 ± 1% characteristic of the central regions of groups and poor clusters of galaxies, there is a significant decline of the SN Ia Fe mass fraction towards the center of FGs (85 ± 2%).

20 f 0.2 is the Fe mass fraction injected by SNe II normalized by 0.2 energy needed to keep the IGM from cooling ΔE The average supernovae rate for this secondary wind would be large (near 10 SNu), larger, but on the same order as that of M82.

21 Westera et al. 2010, Cuisinier et al A&A 455, 825; Lisker et al AJ 132, 2432 No good fit for young populations <0.5Gy in the BGs of our sample, in the typical young+ intermediary+old fit. However, best fit when two populations (0.5-2 Gy) and 4-10Gy are fitted. RX J

22 Westera et al. 2010, Cuisinier et al A&A 455, 825; Lisker et al AJ 132, 2432 RX J RX J RX J RX J RX J RX J

23 Westera et al. 2010, Cuisinier et al A&A 455, 825; Lisker et al AJ 132, 2432 RX J RX J RX J

24 Age Old Age Young Name E(B V) Mo (Gyr) Mi (Gyr) RX J RX J RX J RX J RX J RX J RX J RX J RX J RX J Could AGNs do the same job? With viscosity bubble gas lifting If so, wouldn t it dilute the SN Ia enhancement in ALL groups and clusters?

25 Small sample + typical photon poverty limits our ability to give more definitive answers. To address these issues, we have constructed a sample of 15 fossil groups from the maxbcg optical cluster survey. SAMPLE SELECTION The 15 candidate FGs were selected from the maxbcg cluster survey (Eli Rykoff s talk yesterday) using the following criteria: 0.09 < z < 0.15 L BCG > 9x10 11 L Δi > 2.0 between BCG and second-ranked Stacking L X α R, L BCG α R, L BCG α Δi Chandra snapshot observations Deep XMM follow-up Optical follow-up (Magellan, Gemini)

26 NEW SAMPLE

27 11 have extended emission detected at greater than 90% confidence, with 9 of these (at greater than 3-σ) allowing reliable temperature determination.

28

29

30 Summary 1- The large velocity dispersion of FGs - have deep gravitational potential typical of poor clusters in agreement with the relatively high T X measured. 2- Typical lack of cooling cores (or just deci cool cores), reduction of the SN Ia Fe mass fraction and presence of a ~Gyr-old stellar population in the cds is consistent with a scenario where SN II powered winds resulting from wet merging erase the original central SN Ia Fe mass fraction dominance. 3-Reconciliation with the theoretical models is possible if the halo that will eventually become a fossil group is formed earlier than those that will become galaxy groups and clusters, but the BGG of fossil groups are formed later than the central galaxies of groups and cds. A similar scenario has been put forward recently (Diaz-Gimenez, Muriel and Mendes de Oliveira 2008) in a study of the properties and merging history of the bright galaxies in simulated fossil groups extracted from the Millenium Simulation Galaxy Catalogue. 4- Our large (1 st ) sample of real FGs with short (chandra) medium (XMM) exposures 4- Our large (1 ) sample of real FGs with short (chandra), medium (XMM) exposures and optical data will help to onstraint significantly the nature of these systems.