Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://www.stsci.edu/science/preprints/prep1297/prep1297.ps Äàòà èçìåíåíèÿ: Wed Nov 11 19:42:36 1998 Äàòà èíäåêñèðîâàíèÿ: Sat Dec 22 14:17:04 2007 Êîäèðîâêà: IBM-866 Ïîèñêîâûå ñëîâà: coma

Optical Di#use Light in Clusters of Galaxies
Rosendo VÒÐlchez--GÒomez
Space Telescope Science Institute, 3700 San Martin Drive, Baltimore,
MD 21218, USA
Abstract.
I present here a review of the observed characteristics of the optical
di#use light in clusters, the possible sources of this light and some of
the theories that try to explain the existence of big envelopes around the
brightest cluster galaxies.
1. Introduction
The first reference that we can find in the literature about the di#use light in
cluster of galaxies was given by Zwicky (1951): ``One of the most interesting
discoveries made in the course of this investigation [in the Coma cluster] is the
observation of an extended mass of luminous intergalactic matter of very low
surface brightness. The objects which constitute this matter must be considered
as the faintest individual members of the cluster. [We report] the discovery of
luminous intergalactic matter concentrated generally and di#erentially around
the center of the cluster and the brightest (most massive) galaxies, respectively''.
This is a perfect characterization of the optical di#use light in cluster: extended,
low surface brightness and around the center of the cluster.
Zwicky was trying to settle three of the problems of the extragalactic astroní
omy at that moment: (1) this luminous intergalactic matter can account for the
dark matter needed in Coma if this cluster were virialized; (2) the shape of the
luminosity function (a Gaussian, according to Hubble) is monotonely increasing
with decreasing brightness; and (3) the galaxies extend notably far away from
their centers 1 .
The characteristics of this di#use matter published by Zwicky (1951, 1957,
1959) were qualitative: it has an extension of around 150 kpc, the color index is
rather blue and produces a local absorption of light of the order of six tenth of
a magnitude.
The first published attempt to obtain a value for the surface brightness of
the faint intergalactic matter in Coma corresponds to de Vaucouleurs (1960). He
reported an upper limit of B > 29.5 mag arcsec -2 at #r# # 0 # .9. With this value,
de Vaucouleurs reasons out that ``a stellar population composed exclusively of
extreme red dwarfs of mass M< 0.1 M# and absolute magnitudes M(pg) > +15
1 Baum (1955) claims that ``galaxies blend into one another with no vacant intergalactic gaps in
between''.
1

would, in principle, give an M/L ratio of the order measured in Coma. While
such stars are known to exist in the neighborhood of the sun, it seems very
di#cult to admit that they could populate intergalactic space with the required
density and to the exclusion of all other stars of slightly greater mass'' 2 . Thus,
de Vaucouleurs concludes that the mass of the intergalactic matter is not enough
to account for the mass value estimated through the virial theorem.
The next step in the first studies of the di#use light in clusters corresponds
to Matthews, Morgan & Schmidt (1964). During the analysis of radio sources,
they found near the center of a number of Abell's rich clusters, supergiant D
galaxies with diameters 3--4 times as great as the ordinary lenticulars in the same
clusters. They gave the prefix ``c'' to these very large D galaxies, ``in a manner
similar to the notation for supergiant stars in stellar spectroscopy''. The reason
for this remark is that I believe that there is not a real di#erence between the
detection of intracluster light or the halo of a cD. Whether this di#use light is
called the cD envelope or di#use intergalactic light is a matter of semantics. In
fact, Oemler (1973) in his study of Abell 2670 where he traced a di#use envelope
to almost 1 Mpc says: ``An important question is the relation between this di#use
component and the central elliptical galaxy, the combination of which seems to
produce the cD galaxy''. Nevertheless, there are clusters without a cD in the
center where a di#use light has been detected in its central part, as it is the case
in Cl 1613+31 (VÒÐlchez--GÒomez, PellÒo & Sanahuja 1994a,b)
Before the CCD detectors were widely used, most of the observations and
study of the di#use light in clusters was carried out in the Coma cluster: Abell
(1965); Gunn (1969); de Vaucouleurs & de Vaucouleurs (1970); Welch & Sastry
(1971, 1972); Gunn & Melnick (1975); Mattila (1977); Melnick, White & Hoessel
(1977); Thuan & Kormendy (1977). There are also some studies in Virgo:
Holmberg (1958); Arp & Bertola (1969); de Vaucouleurs (1969). Finally, there
are also studies of the faint envelopes of elliptical and cD galaxies: Arp & Bertola
(1971); Baum (1973); Kormendy & Bahcall (1974); Oemler (1973, 1976). I will
consider here, basically, the problems associated to the use of CCD's in the study
of the di#use light as well as the results obtained with that kind of detectors.
2. Problems and Errors
If we consider that the intracluster light is expected to be extremely faint, about
25--26 mag arcsec -2 in a red filter (if it represents 10 to 25% of the total light
in the center of an intermediate redshift cluster), it is easy to understand how
hard can be to obtain a reliable detection and analysis of the di#use light in
a cluster. We have to be sure that our detection is not the result of spurious
e#ects, such as instrumental scattering or contamination due to bright stars or
faint galaxies. I will comment some of this error sources:
Instrumental Scattering. The di#use light due to the mirrors of the telescope
is the first source of parasitic light. If the cleanliness of the telescope optics is
2 Actually, Boughn & Uson (1997), studying three rich Abell clusters, where they don't detect
any anomalous reddening in the intracluster medium, conclude that no more than 2h -1 % of
the dark matter can be in the form of low mass (# 0.1 M#) subdwarfs or old disk dwarfs.
2

not correct enough, some of the results that we could ascribe to the intracluster
light would be masked or spoiled.
Flat Fielding. Our images must be cleaned of any kind of residual ghost image
structures as well as free of fringing. A good level of flattening should be lower
than 0.5%.
Contamination due to Bright Stars. As we are trying to obtain accurate surí
face brightness profiles at 25 mag arcsec -2 and lower, it is necessary an accurated
removal of the halos of stars and bright clusters members. An unaccurate subí
traction can alter the result in more than 0.5 mag arcsec -2 . It is also essential
to check for the possibility of contamination due to halos of stars located outí
side but near the field. Some comments about the removal of the halos can
be found in Gudehus (1989); Uson, Boughn & Kuhn (1991); Mackie (1992);
VÒÐlchez--GÒomez et al. (1994a).
Sky Level. If we want to fix what is the real extension of our intracluster light,
we need a correct determination of the sky level. Thus, we ought to be sure
that we are far enough of the central part of the cluster to reach the end of the
di#use light profile. We can get this either working with a big field (i.e., making
a tessellation of di#erent images as in Scheick & Kuhn 1994) or studying a
relatively distant cluster in order to be sure that all the cluster is inside our
CCD.
Faint Galaxies. We have to correct from the contamination due to the galaxies
fainter than the completeness limit in magnitude for our sample. One possibility
is to extrapolate a Schechter luminosity function fitted to our data, until the
detection limit. But if we use the kícorrection, then, we have to assume a
Hubble type for this galaxies. If we consider that they are E/S0 galaxies we
tend to overcorrect the di#use light in the red filters with respect to the blue
ones (VÒÐlchez--GÒomez et al. 1994a).
Other Sources of Errors. For example, vignetting in the image, incorrect deí
termination of the galactic absorption, statistical errors associate with the meaí
sure, wrong redshift for the cluster, the presence of galactic cirrus as reported
by Haikala & Mattila (1995) or Szomoru & Guhathakurta (1998).
3. Characteristics
I will try to summarize some of the most important characteristics associated
with the di#use light in clusters of galaxies:
Luminosity. It shows a wide range. The intracluster light can represent beí
tween the 10% and the 50% of the total light of the region where it is detected.
Schombert (1988) finds some correlation, but faint, between the luminosity of
the cD envelope and that of the underlying galaxy. This correlation can hint
that the process of formation of the brightest cluster galaxy (BCG) has some
reflection in the origin of its envelope.
3

Color. Di#erent authors have report various results. Valentijn (1983) in B-V
and Scheick & Kuhn (1994) in V -R find blueward gradients that vary between
0.1 to 0.6 mag drop. Schombert (1988) in B - V doesn't find any evidence of
strong color gradients or blue envelopes colors. Finally, Mackie (1992) in g - r
reports a reddening at the end of the envelopes, in one case of the order of 0.15
mag.
Structure. Schombert (1988) and Mackie, Visvanathan & Carter (1990) find a
apparent break in the surface brightness profile of the underlying cD galaxies.
According to Schombert (1988), this break is found near the 24V mag arcsec -2
but there are no sharp changes in either eccentricity or orientation between the
galaxy and the envelope. However, Uson et al. (1991) and Scheick & Kuhn
(1994) don't see such a break in their studies.
Reinforcing the idea of common evolutive processes, Schombert (1988) and
Bernstein et al. (1995) find that the di#use light, globular cluster density and
galaxy density profiles seem to have similar radial structure, proportional to
r -2.6 . However, Cl 1613+31 shows a di#erent profile for the di#use light and
the galaxies (VÒÐlchez--GÒomez et al. 1994a).
Cluster properties and di#use light. Schombert (1988), in one of the most comí
prehensive studies of cD envelopes, finds the following correlations between the
luminosity of the envelope (L env ) and the general properties of the cluster: (1)
There is a clear correlation between L env and cluster richness for compact, reguí
lar clusters; (2) there is no evident correlation with velocity dispersion; (3) there
is a slight correlation with the Bautz--Morgan or Rood--Sastry cluster type; (4)
there is an unambiguous correlation with the Xíray luminosity.
Finally, there are no reports of galaxies with envelopes in the field and the
cDílike galaxies observed in poor clusters dwell in local density maxima, compaí
rable to the central region of rich clusters (West & Van den Bergh 1991). That is,
a cluster or subcluster environment with high local density contrast looks like an
unambiguous requirement for the presence of cD envelopes or intracluster light.
4. Sources for the di#use light
Basically, there are five processes to explain the origin of the intracluster light:
Stars from the outer envelopes of galaxies. Sometimes the extension of the
di#use light is so large (several core radius) that is hard to believe that these stars
are gravitationally bound to any galaxy, and probably, they are stripped material
after the interaction between galaxies. This could be the case in Cl 1613+31
(VÒÐlchez--GÒomez et al. 1994a). Also, it could be that the stars have born directly
in the intergalactic medium from a cooling flow, for example (Prestwich & Joy
1991).
Dwarf galaxies and globular clusters. Part of the light in the intergalactic
medium in distant clusters, where it is not possible to resolve dwarf galaxies
and globular clusters, can have this origin. Nevertheless, Bernstein et al. (1995)
have measure in the Coma cluster a di#use light apart from dwarf galaxies and
globular clusters.
4

Hot intracluster bremsstrahlung radiation in the optical. Woolf (1967), Mattila
(1977) and Bernstein et al. (1995) established that, at least for the Coma cluster,
this contribution is not significant if we take into account the boundaries imposed
by the observations in Xíray and the observed H# intensity.
Light scattered by intergalactic dust. The existence of dust in rich clusters
of galaxies as established by Zwicky (1959) or Hu (1992) would suggest the
production of di#use scattered light. Mattila (1977) makes an estimation of
around 12% of the total surface brightness of the Coma cluster can be due to
the surface brightness of the scattered light with origin in the dust.
The radiative decay of particles. Partridge (1990) considers that the radiative
decay of low mass particle (m # # 4 eV) would produce extragalactic light in the
visible.
The first source seems to be the most important. Scheick & Kuhn (1994),
studying the di#use light ``granularity'' in Abell 2670 established that the luí
minosity of each source is less that 10 4 L# . This suggests that the main origin
of the di#use light is light from stars since the luminosity associated with the
sources is about a factor 100 smaller than the luminosity of the faintest dwarf
galaxies. Similar result is found by Bernstein et al. (1995) for the Coma cluster.
5. Origins of the cD envelopes
There are basically four theories that try to elucidate what is the origin and
evolution of the cD envelopes. None of them o#ers a complete picture of the
problem.
Stripping theory. This theory was initially proposed by Gallagher & Ostriker
(1972). According with this theory, the origin of the envelope is on the debris due
to tidal interactions between the cluster galaxies. These stars and gas are then
deposited in the potential well of the cluster where the BCG is located. This
process begins after the cluster collapse and the envelope grows as the cluster
evolves. The fact that di#erent cD envelopes show di#erent color gradients
can be explained as the result of di#erent tidal interaction histories: in some
clusters the tidal interactions involve mainly spirals, but in others, early type
galaxies are the source material (Schombert 1988). The main problem to this
hypothesis is the di#culty to explain the observed smoothness of the envelopes
as the timescale to dissolve the clumps is on the order of the crossing time of
the cluster (Scheick & Kuhn 1994).
Primordial origin theory. This hypothesis, suggested by Merrit (1984), is simí
ilar to the previous one but, in this case, the process of removing stars from the
halos of the galaxies was carried out by the mean cluster tidal field and took
place during the initial collapse of the cluster. The BCG, due to its privileged
position in relation with the potential well, gets the residuals that make up its
envelope. However, this picture is di#cult to reconciliate with the fact that
there are cD's with significant peculiar velocities (Gebhardt & Beers 1991) as
well as with the smoothness of the di#use light either the envelope is a#xed
5

to the cD or fixed and the cD is moving through it. Moreover, if the origin of
the di#use light is primordial, how can we explain the observation of blue color
gradients in some envelopes, supposed little activity after virialization?
Cooling flows. Fabian & Nulsen (1977) proposed that the radiative cooling of
hot Xíray gas can produce an increase of the densities around the BCG until
star formation takes place. But this process is confined to the first one hundred
kpc from the center of the cluster (Prestwich & Joy 1991), insu#cient to explain
the big envelopes of several hundred of kpc observed. Moreover, a blue gradient
color is expected if recent star formation is taking place.
Mergers. Villumsen (1982, 1983) found that after a merger with the BCG, and
under special conditions, it is possible to form an halo similar to that present in
cD galaxies since there is a transfer of energy to the outer part of the merger,
resulting an extended envelope. Although this theory reproduces the profile
observed for the envelopes, it is not possible to accomplish for the luminosities
and masses seen for the di#use light. However, in poor clusters where there
are cDílike galaxies without a clear envelope this mechanism can play a more
important role (Thuan & Romanishin 1981; Schombert 1986).
6. Conclusions
After this review, it is clear that it is necessary to carry out a more systematic
study of the di#use light in clusters to obtain a better comprehension of the origin
and evolution of its properties and its relation with the global characteristics of
the cluster.
Acknowledgments. I would like to thank R. PellÒo and B. Sanahuja for
their help and comments in my study of the di#use light in clusters. I am
grateful to Kalevi Mattila for enabling my participation in this conference. I
thank also STScI and IAU for financial support.
References
Abell, G. O. 1965, ARA&A, 3, 1
Arp, H., & Bertola, F. 1969, Astrophys. Letters, 4, 23
Arp, H., & Bertola, F. 1971, ApJ, 163, 195
Baum, W. A. 1955, PASP, 67, 328
Baum, W. A. 1973, PASP, 85, 530
Bernstein, G. M., Nichol, R. C., Tyson, J. A., Ulmer, M. P., & Wittman, D.
1995, AJ, 110, 1507
Boughn, S. P., & Uson, J. M. 1997, ApJ, 488, 44
Fabian, A. C., & Nulsen, P. E. J. 1977, MNRAS, 180, 479
Gallagher, J. S., & Ostriker, J. P. 1972, AJ, 77, 288
Gebhardt, K., & Beers, T. C. 1991, ApJ, 383, 72
Gudehus, D. H. 1989, ApJ, 340, 661
6

Gunn, J. E. 1969, BAAS, 1, 191
Gunn, J. E., & Melnick, J. 1975, BAAS, 7, 412
Haikala, L. K., & Mattila, K. 1995, ApJ, 443, L33
Holmberg, E. 1958, Medd. Luns. Astron. Obs. II, No. 136, 63
Hu, E. M. 1992, ApJ, 391, 608
Kormendy, J., & Bahcall, J. N. 1974, AJ, 79, 671
Mackie, G. 1992, ApJ, 400, 65
Mackie, G., Visvanathan, N., & Carter, D. 1990, ApJ, 73, 637
Matthews, T. A., Morgan, W. W., & Schmidt, M. 1964, ApJ, 140, 35
Mattila, K. 1977, A&A, 60, 425
Melnick, J., White, S. D. M., & Hoessel, J. 1977, MNRAS, 180, 207
Merrit, D. 1984, ApJ, 276, 26
Oemler, A. 1973, ApJ, 180, 11
Oemler, A. 1976, ApJ, 209, 693
Partridge, R. B. 1990, in The Galactic and Extragalactic Background Radiation,
S. Bowyer & C. Leinert, Dordrecht: Kluwer, 283
Prestwich, A. H., & Joy, M. 1991, ApJ, 369, L1
Scheick, X., & Kuhn, J. R. 1994, ApJ, 423, 566
Schombert, J. M. 1986, ApJS, 60, 603
Schombert, J. M. 1988, ApJ, 328, 475
Szomoru, A., & Guhathakurta, P. 1998, ApJ, 494, L93
Thuan, T. X., & Kormendy J. 1977, PASP, 89, 466
Thuan, T. X., & Romanishin, W. 1981, ApJ, 248, 439
Uson, J. M., Boughn, S. P., & Kuhn, J. R. 1991, ApJ, 369, 46
Valentijn, E. A. 1983, A&A, 118, 123
Vaucouleurs, G. de 1960, ApJ, 131, 585
Vaucouleurs, G. de 1969, Astrophys. Letters, 4, 17
Vaucouleurs, G. de, & Vaucouleurs, A. de 1970, Astrophys. Letters, 5, 219
VÒÐlchez--GÒomez, R., PellÒo, R., & Sanahuja, B. 1994a, A&A, 283, 37
VÒÐlchez--GÒomez, R., PellÒo, R., & Sanahuja, B. 1994b, A&A, 289, 661
Villumsen, J. V. 1982, MNRAS, 199, 493
Villumsen, J. V. 1983, MNRAS, 204, 219
Welch, G. A., & Sastry, G. N. 1971, ApJ, 169, L3
Welch, G. A., & Sastry, G. N. 1972, ApJ, 171, L81
West, M. J., & Van den Bergh, S. 1991, ApJ, 373, 1
Woolf, N. J. 1967, ApJ, 148, 287
Zwicky, F. 1951, PASP, 63, 61
Zwicky, F. 1957, Morphological Astronomy, Berlin: SpringeríVerlag, 48
Zwicky, F. 1959, in Encyclopedia of Physics, S. FlØugge, Berlin: SpringeríVerlag,
53, 398
7