NGC 6749 Aquila - Tough Cookie, Soft Heart
Submitted: Tuesday, 27th August 2013 by Dana De Zoysa
NGC 6749 has a reputation as the most difficult NGC globular in the sky. The observing reports one finds on the amateur forums mention aperture requirements of 300mm and upward (way upward). I wondered about its visibility in my 150mm to 200mm medium-aperture scopes. Now I've observed its field across the past three nights for more than an hour each night under excellent skies using 100mm to the 200mm scopes.
There are two issues in positively distinguishing N6749 from adjacent field contamination. First, listed globular visual magnitudes in most charts are misleading. Globular light falls off at a specific exponential rate from core to halo which depends on the cluster's density, from Class I highly concentrated, through Class XII, very loose. A better measure is the half-light radius which is a measure of core concentration. You can look up this data and all the other relevant data in the Harris Catalog.
N6749's half-light radius is 1.10 arcmin across a total diameter of 6.1 arcmin. That indicates a very weak halo concentration and low core visibility. The visual mag in N6749's half-light radius is 11.9, which compares with 11.0 for NGC 7009 at its 1 arcmin half-light radius. N6749 is the loosest globular in the sky at Cl XII, which explains its mean surface brightness of 21.8 (compare w. N7009 at sb 19.0). Both N7009 and Pal 12 are convenient nearby reference points when getting a handle on 6749's observability under a specific session's sky.
Problem #2 is that when using smaller scopes there is a sprinkle of seven mag 11.8 to 14.2 stars within a roughly trapezoidal shape directly adjacent to N6749. It stands out in WikiSky much less clearly as a distinct group than it does at the eyepiece. This group abuts N6749 out to 5 arcmin W and 2 arcmin SW. At lower apertures and magnifications this group blends into a patch which can easily give the impression of a low-contrast extended object. It shows up that way in my 100mm refractor at 60x and there's no way I can ever see a mag 12 GC one arcmin in dia. in a 100mm scope. My 200mm Mak is mounted alongside the 100mm refractor. Switching to it at 169x quickly resolves the patch into distinct stars. Only then does N6749 become visible as a very faint glow just touching the NE corner. It took nearly an hour to confirm the cluster on first night out under approx. mag 7 seeing, and that was only three fleeting glimpses. This was nearly the same visual impression the 100mm refractor gives of the starry patch. The next two nights I could see N6749 as a 1.5 arcmin very faint glow in a 180mm Mak, and more cleanly distinct in the 200mm. As is often the case, once having cleanly ID'd the object, it is much easier on subsequent tries.
What are we looking at?
Galaxies are fearsome places to live. They fatten on dwarf galaxies and nibble globular clusters for snacks. In our eyepieces, globulars have the air of self-contained splendour impervious to time and change. Compared with nearby open cluster cute young things, globulars are remote dowagers. But globulars live in a rough neighbourhood. The Milky Way has consumed roughly 77% of its original globular population through tidal truncation and gravitational shock. Of the 153 globulars catalogued today, only 112 are primordial. The rest are either newbies under 10 billion years old, or were stripped from dwarf galaxies. Yet those 112 are only ~23±6% of the original population when our galaxy was 1 billion years old (see §4.5 of this paper). Where have the 370-odd others gone? Is N6749 destined to join them, a stream of stars dissolving into the Galactic outer halo?
NGC 6749 is an aging halo cluster on an as-yet undetermined orbit which presently is plunging the cluster through the Milky Way. According to the Harris catalog, N6749 is presently 5 kiloparsecs (16,300) light years from the Galactic core and 7.9 kpc (25,750) ly from us. At only 2.1 degrees below the Galactic plane at +32.6° Galactic longitude, it lies in the interarm region between the Scutum/Centaurus Arm and the Sagittarius Arm (of which our Orion Spur is a part). Between Earth and N6749 is a thick 2,000 light-year star forming region where our Galaxy’s bar structure turns sharply left to begin the Sagittarius Arm. After that, light from N6749 passes through an additional 10,000 ly of spiral arm star formation, dust, infalling cloud gas, shock fronts emitting in the IR, plus supernova and magnetic turbulence on its way to our eyepieces. Much of N6749’s original light has been absorbed by dust on the way. N6749 is being tidally disrupted by the passage, but we don’t know how gravitationally weakened the cluster has become because astronomers haven’t yet determined its path and original composition. The fate of N6749 is not unique—NGC 2298 Puppis has lost 85% of its stars to tidal stripping and likewise may not survive much longer.
N6749 enjoys great vogue among amateur enthusiasts who struggle past weather, aperture, eyepieces, cold, mosquitoes, and light pollution to claim the pennant of actually having seen it. It is arguably the most difficult NGC globular to log. The cluster makes the task no easier by hiding behind 4.6 visual magnitudes of Galactic obscuration. Its spectroscopic E(B-V) extinction is 1.5, which is the optical B (blue) band at 455nm minus the V (visual) band at 551nm. [To convert an E(B-V) extinction to the visual magnitude you will see in a telescope, multiply the R(B-V) number by 3.1. The Harris listed extinction of 1.5 converts to 4.6 magnitudes as seen visually.]
With clusters, the ‘visual magnitude’ listed in most observer’s guides is misleading—it is the magnitude the cluster’s overall surface brightness would be if concentrated into a single starry point. A cluster’s listed visual magnitude is many faint stars spread out over its visual diameter, brightest at the centre and fading to sky brightness in the halo. A more realistic expectation of what we can see is the globular’s half-light luminosity. N6749’s half-light magnitude is 11.9 across the central 1.1 arcmin of the cluster, which is only 3.25% of the total light-emitting area. The brightest part is in the smallest bit. The listed visual magnitude is 12.4 overall. The cluster’s surface brightness is 21.8, two magnitudes fainter than the cluster’s mag 19.5 horizontal branch (HB) stars. The cluster’s brightest red giant is mag 18.1.
If not for the -4.5 extinction, N6749 would resemble its nearby neighbour in space, NGC 6760 Aquila at vis.mag 9.1 and surf brightness 17.5, or NGC 6712 Scutum beneath M11. All three clusters are physically close to each another, with only 3,500 light years separating them, all on the under side of the Galactic bulge/bar from us. In our telescopes, half a degree away to the NW from N6749 is the glowing H-alpha cloud Sh. 2-72, a sign of how much gas and dust intervenes in the area. The nearest globular to N6749 physically is NGC 6760, which is an easy find in amateur telescopes. N6760 lies 1800 ly closer as seen from Earth, and only 1.8 degrees Galactic latitude lower at -3.9. Despite such a close line-of-sight view, their visual properties are very different. N6760 glows with the combined luminosity of 10,800 Suns; N6749 barely glows with 1965 solar luminosities. This is due mostly to differential dust reddening. Amateurs hunting for N6749 often check N6760 first to gauge whether the seeing conditions are good enough to try N6749.
Unfortunately, for all its fascination to astronomy enthusiasts, N6749 is little studied by the professionals. A literature search on ADS, arXiv, IOP, and Simbad raises few papers, the most recent from Rosino et al 1997, supplemented by Brian Skiff’s 2002 update reviewing nearby Mira variables (none part of the cluster). The globular professionals should consider a major upgrade of the data on this cluster because it is one of the prime examples of what happens to a cluster’s stars after the core has been disrupted and its halo tidally stripped. (At the bottom of this article is a suggested list of spectral and orbital data that would considerably sharpen our knowledge of the cluster.)
Until about six years ago professional astronomers faced the same problem with this cluster that we do: it’s tough to distinguish cluster members from the rich, crowded field surrounding it. There are no hirez color-magnitude plots of 6749 for us to estimate the extent to which bulge field metallicity and star contamination misleads its actual age and distance. The standard CMD published here is so contaminated by bulge stars that we have a poor idea of the cluster’s main-sequence turnoff, the slope of the red giant branch, or the blue/red properties of the horizontal branch. For example, stars that show up as blue stragglers (merged low-mass stars) cannot be distinguished from ‘blends’—several unrelated stars that occupy the same digital pixel. The 1997 Rosino paper estimates the cluster’s red giant branch slope and blue horizontal branch by model-fitting using mathematical models that were excellent at the time but have been much improved since. Besides, hi-rez data-per-pixel provides a more accurate fit than models.
The 1997 Rosario study was based on a 1024x1024 chip at 0.37 arcsec/pixel resolution in a 1.54-meter Danish telescope in Chile. The data set is a constrained sample of six red giants and six blue horizontal branch stars. Thirteen years later, in 2008 the first Multi-Conjugate Adaptive Optics Demonstrator (MAD) went into service at the 8-meter Melipal telescope in Chile. Adaptive optics with 0.001 arcsec resolvability, combined with a diffraction grating of 41,000 lines/inch, became the first really hi-rez spectrograph system devoted mainly to cluster studies. Melipal’s present infrared-band detectors capture 2048 x 2048 pixel data at a scale of 1 pixel = 0.028 arsecs across a 2 x 2 arcmin field. That’s 13 times the 1997 resolution, and many more bandwidths are now available. Remember when your PC was running an Intel 486 and was the fastest thing known in a buff gray box? That's roughly what a 1997 CMD means today. N6791’s metallicity was measured at -1.61 (1/40th the Sun’s) compared with the average bulge metallicity at that latitude, which close to Solar. Good as the Ortolani and Barbuy team was in 1997, they couldn't noodle out whether the effective reddening defined N6749 as a bulge globular, or the blue tail on its horizontal branch tagged it as a old halo object undergoing a disc crossing.
N6749’s aging horizontal-branch stars indicate that it is a mix of the two: an old halo globular presently passing near the bulge. No orbital vector been determined, so we don’t know if the passage is grazing or a deep plunge. But N6749 is clearly being disrupted by this particular bulge crossing. It is the least-concentrated globular in the sky, at 0.79. For our visual comparison, nearby N6760’s concentration is 1.65, tipping it as a Class IX cluster.)
N6749 is a good study in globular cluster dissolution consequent to a tidal encounter. Its core radius is 0.62 arcmins while its half-light radius is 1.10 arcmin across a total diameter of 6.1 arcmin. The half-light radius has a surface brightness of 21.8 mag sq.arcsec. At Class XII, N6749 is the loosest globular in the sky. With a diameter of 47 light years, its half-mass density is 3.3 solar luminosities per cubic light year. If you lived in N6749’s core, the sky would be full of mag 3 and fainter stars—a little like life in a sky filled with so many Alpha Hercules- or Aries-like stars that you couldn’t see much else.
When a cluster as loosely bound as N6749 undergoes a tidal crossing, its outer stars evaporate into the bulge at 3 to 5 escapees per ten million years. This is easily interpreted as gravitational stripping. But it can be more accurately understood as an example of a cluster’s thermal equilibrium out of balance with its surroundings. The life cycle of star clusters is a cluster’s attempt to balance thermal differences. When a globular first forms, its first generation of stars undergoes a 100 million year cycle of thermal rebalance from the cold dark cloud state to a hot cluster state. Early supernovae of stars over 8 solar masses occurs within 6 Myr. A red-giant-to-white-dwarf cycle involving 2.2 to 8.0 solar-mass stars lasts roughly 5 to 7 billion years, but peaks starting 200 million years and slowly tapers off. Finally comes a very long senescence of sub-2 solar-mass stars radiating smaller amounts of energy up to the present age of the universe; this last is also the era when a cluster is the most dynamically (gravitationally) stable.
Even if there is a second star-forming generation starting at about 200 Myr made out of the unused gas of the first—which is frequent among globulars—the cluster has ejected nearly all its star-making gas via solar winds within the first half-billion years its life. Thereafter, the cluster is in a constant struggle with thermal equilibrium to stay together. Initially, more massive stars migrate to the centre to form binaries—core collapse occurs, lasting roughly 1-2 billion years. Core collapse is an attempt to eject excess kinetic heat. It loosens the cluster’s hold on peripheral stars in an attempt to establish thermal equilibrium. Then the process goes the other way: there is a drop in thermal energy in the outer regions as stars drift away. In the centre, third-star encounters with binaries eject the least-massive star, which lowers thermal energy and brings the core to re-expand. In any warm object, cooling in the outer regions moves surplus interior heat outward. Core re-expansion is a long-term process lasting 5 to 8 Gyr. The cluster steadily loses mass because of stars escaping into the galaxy while its aging core population losing mass by radiation and stellar winds. (Movie stars aren’t the only stars to have problems after sticky affairs.)
This is N6749’s biography. It is also every globular’s biography. Their fate is played out with many variations and endings, all caused by the cluster’s relationship with its host galaxy and its own thermal deterioration. (We use the term ‘host galaxy’ rather than ‘parent’ galaxy because the clusters usually form before the galaxy itself.) Today, being disrupted by a galactic crossing, N6749 hasn’t enough mass to keep itself together. Its likely fate is to be stretched into a stream to eventually join all the other disrupted globular stars in the Galactic halo. Such is the current fate of the Sagittarius Dwarf and Canis Major Dwarf galaxies, plus perhaps 375-odd other globular clusters. But while the Sagittarius Dwarf bequeathed the Milky Way 12 globulars (of which we can readily spot four in our backyard Palomars), and the Canis Major Dwarf gave over 7 globulars to the Milky Way, N6749 will end up banished via dispersion to the Galactic halo.
How can we be so certain this will happen to N6749? Despite the low resolution of the data we have, N6749 clearly evidences three clues to its origins and evolution: (a) a blueish main-sequence turnoff point; (b) a steep red giant branch; and (c) an extended blue horizontal branch. Let’s look at these in turn.
A blue main sequence turnoff occurs at the point where a low-mass star’s hydrogen fusion flickers off and the star undergoes a gradual collapse. The star’s core is helium ‘ash’ from the hydrogen burning; a helium core is incompressible and uniformly dense, so it just gets hotter as the envelope contracts. When the core reaches 20 million degrees K, a thin hydrogen shell surrounding the inert helium core ignites, converting hydrogen to helium via the CNO cycle. The star cools and reddens as it turns rightward off the main sequence. In stars like the sun, the main sequence lasts roughly 7 billion years. The shell contraction and ignition phase leading to the red giant phase lasts another 1-2 Gyr. The red giant phase itself takes up another billion years. Most of N6749’s remaining main sequence stars are well under the Sun’s mass. Sub solar-mass stars become brighter and bluer on the color-magnitude diagram than more massive stars like the Sun, and hence appear more blue before their central fire turns off and the stars cool but radiate the same brightness.
The slope of a globular’s red giant branch (RGB) is more vertical if it has a low metals content. Metals add opacity to a star’s luminous envelope. If a cluster has low metallicity and hence little opacity, it appears brighter at any given colour temperature. Hence a low-metallicity (low opacity) cluster contains many stars brighter for their colour, which reveals itself as a steeper-sloped red giant branch. Clusters with RGBs angled lower toward the right have higher metals and thus more opacity; they appear dimmer at a given temperature; they appear more red.
The horizontal branch is the most finely tuned and delicately balanced region of the color-magnitude diagram. Only former red giants of between 0.5 to 1.2 solar masses make the cut. These have 100 million K helium-burning cores of 0.47 to 0.51 solar mass burning at over 100 million K with a thin 20 million K hydrogen-burning shell just above it. Stars with more than 1.21 solar masses arrive at the red end of the horizontal branch turnoff point, but soon re-ascend upward and redward through the AGB asymptotic giant branch. They end their days in the brief splendour of a planetary nebula followed by an interminably long dwindle from white dwarf to dark cinder.
N6749’s horizontal-branch is important because N6749’s color-magnitude plot has no apparent red horizontal branch (hence no AGB stars) but a significant blue horizontal branch (see Fig. 3 here). These are properties of an old, metal-poor halo globular whose remaining low-mass main sequence comprises stars under 0.8 solar mass. The blue horizontal branch of all globular clusters that have one (47 Tucanae, for all its visual glory, doesn’t) is composed of stars with cores between 0.49 and 0.55 solar masses. Stars with 0.49 core masses will fizzle into helium-rich white dwarfs and go quietly with neither a bang nor a whimper. Stars whose cores are 0.52 solar mass will undergo five to seven ‘blue loops’ of wildly fluctuating brightness and color change over million-year cycles before also fading into helium dwarfs. Stars with 0.55 or greater core mass sit briefly at the red end of the horizontal branch, then rise into the AGB.
Because their properties are so finely tuned, both blue and red HB stars are standard candles by which the distances and dust reddening of globulars can be calculated. The envelope mass of a red horizontal branch star with a core mass of 0.55 solar is about the same as the Sun’s total mass and its surface temperature is 5,500 K. A hot ‘blue subdwarf’ horizontal branch star with a core mass of 0.49 solar has an envelope only 1/100th the mass of the Sun. The envelope is so thin its bright helium-burning core shines through with surface temperatures between 25,000 and 40,000 K. They are as hot as young O and B supergiants, but only one-fifth as bright. The HB is so delicately balanced that a minuscule 0.06 of a solar mass in the core makes the difference between a slowly fading helium dwarf and an incandescent hottie that plummets into obscurity.
Hence all star clusters need some kind of thermal energy source to sustain their cohesion. The volume shape of a cluster’s binding energy is called a gravitational well. The amount of evaporation or expansion is set by the cluster’s half-mass density and the number of its stars. Globular clusters born and residing in the bulge have come to terms with their crowded surroundings (~10 stars per cubic 3.26 light years) in part by circling at nearly the same speed as the bulge (averaging ±26 kms/sec faster than the bulge’s rotation rate), and by collapsing their gravitational well and thermal density into a tighter, more stable form within their first 5 billion years. That they have acquired considerable coping skills is confirmed by the fact that some bulge globulars are located inside the Galaxy’s bar structure.
The Milky Way’s bar first formed about 7 Gyr ago. Since then the Milky Way has formed a bar at least twice and seen it dissolve again. Bars channel enormous quantities of star-forming gas into a galactic bulge, which is why many active galactic nuclei (AGN) are spirals with bars. Bulge globulars have learned to live with this by exchanging modest numbers of outer stars for a denser core. The process is a little like local tribal chieftains paying a prudent tribute to a major warlord just to leave the tribe alone; but while tribes can always have more babies, globulars can’t.
The term ‘evaporate’ is misleading. A cluster’s outermost stars don’t just drift off in any direction; they can escape only through their L1 and L2 Lagrange points. These lie in front of and behind the cluster with respect to the massive Galactic centre. (The L3 and L4 points on either side of the cluster are like the Trojan objects 60 degrees ahead and behind Jupiter on its orbit, safe harbours within which objects are undisturbed by tidal fields). And even arriving at the L1 and L2 points, a star’s path has to be aligned within a 7-degree cone aimed straight at the Lagrange point; if the star isn’t headed in exactly the right direction at the cluster’s escape velocity, the cluster’s own gravitation and the Galaxy’s tidal pressure will loop the star back toward the cluster centre. (You can see the process in action here.) Even after escape the star’s path is constrained to the looping paths of Lissajous orbits.
The process is not all-loss and no-gain. As they plow through a galaxy, dense clusters can also gain field stars, but no Lagrange points are involved. Stars drifting into the gravitational well of an advancing globular will veer around the core and enter its tidal tail. A loftier term for tidal tails is ‘structural overdensity’—hard to visualize but more accurate to calculate. Structural overdensities obey a specific set of rules which predict when, where, and in which vectors stars will fall into a cluster’s energy field, and also how a tidal tail can lose stars back into space at any point on the cluster’s orbit. In some cases we can determine which halo stars escaped from which globular by their metals contents.
N6749’s iron-to-hydrogen metallicity ratio is -1.65, roughly 1/45th that of the Sun. Iron in a star’s envelope originates largely from Type 1a supernovae, originating from primordial stars from 2.2 to 8.0 solar masses. Most other chemicals, and all heavy elements, are made in Type II supernovae from young >8 solar mass stars exploding before their 6-millionth birthday. N6749’s inferred spectra is largely hydrogen and helium with only traces of carbon, oxygen, magnesium, aluminium, and iron. All this points to a very old cluster whose bright massive stars shed their gases and exploded so long ago their remnants have subsided into the Galactic medium.
As you gaze at it today, you are witnessing the slow demise of once-serene grande-dame who lived on an isolated country estate but has now fallen into the midst of a tough, teeming city.
More data is needed
The existing spectography of N6749 is so imprecise that astronomers do not know whether it is a primordial one-generation globular, or contains two early star generations like most globulars. We need today’s hi-rez spectroscopy to determine (a) its true 3-D path through space, (b) whether N6749 has a bimodal main sequence, and (c) the slope of its red giant branch. Although all these details won’t help we amateurs spot it more easily, all astronomy’s understanding of the cluster would be better if we had more of the following data:
Links and Resources