APOD: Spiral Galaxy NGC 2841 Close Up (2011 Feb 19)

[Ann]

Thanks for your kind words, Doug!

You asked:

Do astronomers have actual evidence of black holes inside elliptical galaxies ?

The best way to prove the existence of a black hole in the center of any galaxy is to measure the orbits of the innermost stars orbiting the black hole. If they orbit fast enough, and if their orbits are small enough, then Kepler's law about orbits tells us that a huge amount of matter is contained within a very small space inside that orbit. And if so much matter is contained within such a small orbit, then this matter has to collapse into a black hole, according to what astronomers believe they know about the extent to which matter can be compressed without collapsing.

This is an interesting youtube video about the orbits of some stars around the central black hole of our own galaxy:

Click to play embedded YouTube video.

It is hard if not impossible to measure how stars orbit the black holes of galaxies that are millions of light-years away, which is the case of all large elliptical galaxies. Another problem with elliptical galaxies is that they lack the brightest kinds of stars, stars which are found in our own galaxy: the millions of B stars, the thousands of O stars and the few exceptional Luminous Blue Variables (LBVs) and the yellow and red hypergiants. I believe that several of the stars that were observed as they orbited the black hole of our own galaxy are B-type stars. Elliptical galaxies don't have these stars.

No, but elliptical galaxies do have red giant stars, which can become quite bright. They can reach a maximum brightness, which I believe is at least a thousand times the brightness of the Sun. This maximum brightness is called the tip of the red giant branch. Wikipedia writes:

Tip of the red giant branch (TRGB) is a primary distance indicator used in astronomy. It uses the luminosity of the brightest red giant branch stars in a galaxy to gauge the distance to that galaxy. It has been used in conjunction with observations from the Hubble Space Telescope to determine the relative motions of the Local Cluster of galaxies within the Local Supercluster.

Hubble is sharp-eyed enough to spot the brightest red giants in a supergiant elliptical galaxy like M87, which according to Hubblesite is located about 50 million light-years away. So Hubble could theoretically follow the innermost red giants of M87 as they orbit the central black hole of that galaxy. And it is quite possible that Hubble has made at least a preliminary study of the innermost stars in M87. But Hubble is most famous for having spotted a chaotic gas cloud orbiting the central black hole of M87:

http://hubblesite.org/newscenter/archiv ... l/1994/23/

Hubble spotted a gas cloud rapidly orbiting the black hole, rather than individual stars orbiting it (though my guess is that there must be many stars orbiting close to the black hole, too).

As the gas cloud orbits the black hole of M87, the black hole keeps swallowing some of the gas. As gas falls into the hole, some gas is rapidly blasted away in the opposite direction, forming a strong jet:

Click to view full size image

There is actually a jet in the opposite direction too, except that it is only visible as a radio jet:

Click to view full size image

You can see that one of the jets is strongly bent, but there are clearly two of them.

Other elliptical galaxies also show evidence of having piled-up mass rotating around their centers, and the rotation can be measured and the mass of the black hole can be inferred.

Doug, you asked me:

How did you access images at cosmo.nyu.edu website ?

I have no idea how I did it. I just did. I use the web reader Opera, and sometimes it seems able to do things that Explorer is unable to. On the other hand, the opposite is often true, too.

Ann

[dougettinger]

Hello Ann,
I want to confirm that we are thinking the same way about the aforementioned "plumes". Plumes are the aftermath of supernovae and novae. From our vantage point we are looking at the outside envelop of supernova and red giant remnants. These envelops, or bubbles, or plumes still retain their spherical or oblong shapes because they have not been disturbed recently in cosmic time by other star explosions and the perturbations of other passing stars. These plumes are becoming the giant interstellar molecular clouds that will create more stars in the future. I hope that this summary is close to what you are trying to convey; if not, then please explain the differences.

I still wonder about your statements regarding that prolific supernovae and starbursts expelling the necessary gases and dust from a galaxy that can create future stars. Will not the gravity of observable matter and dark matter retain the dust and gases although they may move farther away from the plane of a spiral galaxy's disk ? The dust lanes seen on the edges of the Somberro Galaxy are obviously being retained by gravity forces. Assume that a supernova star is orbiting the galaxy at about 250 km/sec and then its supernova explosion accelerates matter in radial directions at about 1000 to 2000 km/sec. The gravitational attraction of the residual mass of the explosion (the neutron star or black hole) and the initial velocity vector should eventually bend the trajectory of the expelled mass in the direction of the initial orbit of the exploding star and keep the debris close to the plane of the galaxy's disk.

Perhaps you are using inductive reasoning for explaining that the expulsion of dust and gases from a galaxy results in the observed low or reduced star generation of numerous galaxies. Anymore commentary about gas and dust expulsion would be appreciated.

Thanks for explaining evidence of black holes in the center of elliptical galaxies. I was thinking that the stars of an elliptical galaxy hid the center and would not reveal the nature of their central regions. It is truly amazing how you are bringing all the facts together to tell a fairly complete storyline for galaxies, which I believed until now did not exist. You are in a good position to write a paper or dissertation about galaxies. You could especially expand your idea that elliptical galaxies w.r.t. spiral galaxies vary over time in a certain way. I think a new constant of nature is in the making.

Doug Ettinger, Pittsburgh, Pa 3/20/2011

[Ann]

Hello, Doug!

No, I don't think you and I mean the same thing when we talk about plumes. I must point out once again that I got my ideas about plumes from James D Wray, and he may have been wrong about them, for all I know. But to illustrate what James D Wray meant, and what I have picked up from him, let's look at the latest Hubble image of a spiral galaxy, NGC 5584:

Click to view full size image

James D Wray included a picture of this galaxy in his Color Atlas of Galaxies, and this is how he described NGC 5584:

The inner spiral pattern is well established while the outer pattern tends toward plume structure.

So where are the plumes in this galaxy? Well, James D Wray would have said that the feature that emanates from about eleven o'clock and points straight up is a plume. Note that this feature is as long as a spiral arm. James D Wray would probably also have said that the feature that emanates from about five o'clock and points straight down (and is full of star formation) is a plume. He would have said that they are plumes because they are long and straight and don't follow the spiral pattern.

What you are talking about when you say plumes is supernova remnants, or that is how I read you. Supernova remnants can have all kinds of interesting shapes:

Click to view full size image

Click to view full size image

Click to view full size image

Click to view full size image

However, most of the time a supernova remnant will not become

the giant interstellar molecular clouds that will create more stars in the future

as you put it. The reason is that the gas and dust that the supernova expelled just isn't enough to create many more stars, particularly in view of the fact that most stars that go supernovae have already shed much of their mass through a strong stellar wind before they exploded. Also the gas that does get violently flung out during the supernova explosion must be strongly compressed in order to give rise to new stars. Therefore a supernova that "explodes into nothing" will just leave behind a gas cloud that expands in many directions and gets thinner and thinner. No new stars will get created out of this.

However, supernovae can indeed trigger star formation, but I believe it is necessary for the supernova to explode into a previously existing gas cloud for any new star formation to take place. When the force of the supernova explodes into a "fat" and "thick" part of a pre-existing gas cloud, it is not hard to think that this gas cloud may be compressed to the point that it will start forming new stars.

An interesting exemple is the giant star forming region in nearby galaxy M33:

Click to view full size image

You can see that this star forming region appears to consist of several "walls" or "shells" of dusty gas structures. There is a relatively thick "wall" pretty much in the middle of the picture. There are many more stars to the right of this "wall" than there are to the left of it.

There is a blue "mist" permeating most of the nebula. The blue mist is X-rays, recorded by the Chandra X-ray Telescope. Interestingly, there are enough hot young stars to the right of the middle "wall" to account for all the X-rays there. The fiercely colliding stellar winds from many hot young stars are enough to explain the X-rays.

But what about the region to the left of the wall in the middle? There are definitely fewer hot young stars in that part of the nebula, yet the blue light of X-rays is every bit as strong to the left as it is to the right. What generates the X-rays to the left of the wall?

The answer is almost certainly supernovae. Certainly there are no supernovae exploding there right now, but there must have been several supernovae going off there in the relatively recent past. The energy of those supernovae still show up as a lot of X-rays.

But what is more, I believe that those supernovae helped compress the gas cloud to the right and trigger star formation there. That is why the stars to the right of the wall are younger and much more numerous than they are on the left. Personally I believe that this is a clear case of supernovae triggering star formation.

And if you wonder why there is a "hole" surrounding the mighty cluster to the right of the wall, I'd say that this "hole" was created when all those hot bright stars in the cluster unleashed their fierce stellar winds and blew the gas away.

But the gas didn't just disappear. It was "piled up" instead. Take a look at the brightly salmon-colored wall to the right of the bright cluster (at three o'clock). My guess is that this is a particularly thick wall of gas and dust that was created when the gas was compressed by the stellar winds of the hot stars in the cluster. When the most massive stars start going supernovae, which hasn't happened yet, the gas on the right of the cluster will probably become even more compressed, and even more stars may be born out of of the gas here.

Read about NGC 604 here: http://chandra.harvard.edu/photo/2009/n604/

Ann

P.S. I just thought of something. I'm sure I've read someplace that meteorites in our solar system contain isotopes (or something) that are usually only forged in supernova explosions. A possible interpretation is that our entire solar system was created out of material that had been forged in a previous supernova. Perhaps that supernova compressed and injected new material into a pre-existing gas cloud to the point that this cloud collapsed and gave birth to a new star - and that star was our own Sun, plus the disk of material that gave rise to the Earth and to us! If so, then we are living examples of the starforming (and life-giving) ability of supernovae.

[Ann]

Doug, you asked:

I still wonder about your statements regarding that prolific supernovae and starbursts expelling the necessary gases and dust from a galaxy that can create future stars. Will not the gravity of observable matter and dark matter retain the dust and gases although they may move farther away from the plane of a spiral galaxy's disk ?

A very good question! Indeed, why doesn't the gravity of big galaxies hang on to the gas that has been expelled from them by supernovae explosions? Why doesn't the gravity of the big galaxies make that gas fall back to them again? You are right, that should happen. So why does the gas become lost?

I think that one reason is that all galaxies started out gas-rich but small. Because they were small, their gravity was weak. If there was a flurry of supernovae explosions in a small galaxy, the galaxy's gravity would have been unable to hold on to the gas. This would have happened mostly in the early universe, before galaxies had had time to grow big and massive. A present-day example of what it may have looked like could be nearby galaxy M82:

Click to view full size image

M82 is a small galaxy. Its luminosity, according to the Principal Galaxy Catalog, is only 0.2 of the galaxy of the Milky Way. According to http://www.fas.org/irp/imint/docs/rst/Sect20/A4.html, M82 has a diameter of only about 16,000 light-years, versus about 100,000 light-years for the Milky Way. Because M82 is small, its self-gravity is much lower than the self-gravity of the Milky Way. It has undergone a tremendous central starburst, which has set off a large number of supernovae. Here is the result: huge amounts of gas is being expelled from the galaxy. Will the gas fall back onto the galaxy again? Well, some of it may, and some of it probably will. But much of it probably won't. Where will the lost gas go? My guess is that it will get caught up by the gravitational forces that bind the three galaxies, M82, M81 and NGC 3077:

Click to view full size image

This image shows M81 at bottom left, M82 on the right and NGC 3077 at top left. You can see the tidal tails of galaxies connecting the galaxies. Personally I can easily imagine the gas being expelled by M82 to end up in such tidal tails, rather than falling back onto M82.

In the early universe, all galaxies were small, or at least smaller than today's big galaxies. They also contained more gas. There must have been many times when small galaxies underwent violent starbursts, and much of their gas was lost due to a series of supernova explosions. Put it this way: the violence of all the supernova explosions that expelled the gas was stronger than the galaxies' gravity that tried to make the gas come back.

Well, time flies, and I must end this post now.

Ann

[Ann]

When it comes to really big galaxies, and particularly when it comes to really big elliptical galaxies, it would seem that it is not supernovae that are the mains culprits when it comes to removing the gas from the galaxies. It would seem that the monstrous black holes typically found at the centers of huge elliptical galaxies are responsible for driving away most of the gas from the galaxies and preventing the remaining gas from forming stars. Check out this page: http://www.msnbc.msn.com/id/14686275/ns ... nce-space/

Here are a few quotes from that page:

[O]nce a black hole reaches a critical mass and become too large for its host galaxy, it zaps away nearly all the gas needed for young stars to form.

How is that zapping done? Well, here's how:

The removal or ejection of cold gas from the center of galaxies possibly is achieved in part by jets of energy that shoot from black holes.

You saw the picture of giant elliptical galaxy M87 and its jet that I posted earlier. That jet is carrying a lot of gas away from the galaxy. The energy of the jet is enough to carry the gas away far enough so that it doesn't fall back onto the galaxy, in spite of the galaxy's impressive gravity. Take a look at the radio picture of M87 again, the picture that shows the two incredible radio jets:

Click to view full size image

A lot of gas is being carried away from M87 by these jets.

But let's return to the page about black holes and star formation. The black hole doesn't carry all the galactic gas away, but it stirs up and heats the remaining gas so that it can't form new stars:

Galaxies can contain a large amount of hydrogen gas. If that gas is sufficiently cold and dense, clouds of it collapse to form young stars. However, if the gas in elliptical and lenticular galaxies has been heated to very high temperatures, it becomes unavailable as fuel for star-formation.

And it is the the black hole and its various outbursts and jets that heats and stirs up the gas in large elliptical galaxies so that this gas can't "settle down" and cool and start to condense to form a stellar core. The gas becomes too hot and is just whirling about.

Ann

[Ann]

What happens to the gas that has been expelled by the jets of the monstrous black holes of large elliptical galaxies? Does this gas just disappear once it has been expelled? No, it is trapped by the combined gravity of the galaxy cluster, because most large elliptical galaxies are members of galaxy clusters. The gas that is trapped inside the cluster is called the intracluster medium. http://www.daviddarling.info/encycloped ... edium.html writes this about the intracluster medium:

intra-cluster medium (ICM)

The hot (tens of millions of K) and extremely tenuous, X-ray emitting gas that exists between galaxies in clusters of galaxies. In big clusters, the ICM may contain more material than all the galaxies put together(though both are out-weighed 10:1 by dark matter). One of the puzzles in astrophysics has been to explain why this gas hasn't cooled down, because of its X-ray emission, and then condensed to form more galaxies. The solution to this now seems tied to jets emitted from massive black holes that lie at the center of active galaxies at the heart of many galaxy clusters. The black holes swallow up any gas coming close to them and liberate enormous amounts of energy in the process. This energy drives very narrow outflows of gas at velocities close to the speed of light, which reheat the intra-cluster gas. Effectively, the black holes act as thermostats. As hot gas in a cluster cools, it flows to the cluster center and is consumed by the black holes inside active galaxies. Some of the energy from this process drives jets into the cluster gas further out, which heats the remaining gas and drives it away from the cluster center. As the black hole runs out of fuel, it shuts down, ready for the whole cycle to begin again.

Ann

[Ann]

Why is it that large elliptical galaxies have such massive central black holes? M87, for example, has a central black hole that contains 6.6 billion times the mass of the Sun, according to Wikipedia. The Milky Way, which is a good-sized spiral, has a black hole that contains about 4 million times the mass of the Sun, again according to Wikipedia. So the black hole of M87 is more than a thousand times more massive than the central black hole of the Milky Way. Why is it that large elliptical galaxies get such massive black holes in their centers?

Perhaps the answer is the opposite - it is not the elliptical galaxies that grow such truly massive central black holes for some reason, but it is rather the supermassive central black holes, once they has grown past a critical mass, that shut down the star formation of their host galaxies and turns them into either elliptical or lenticular galaxies. A lenticular galaxy is a galaxy with a flat disk but no star formation.

Click to view full size image

A lenticular galaxy, NGC 3115. It has a flat disk and a large halo, but no star formation. Its central black hole is two billion times the mass of the Sun, according to Wikipedia.

Ann

[Ann]

So the black hole of M87 is more than a thousand times the mass of the Milky Way black hole. The NGC 3115 black hole is about 500 times the mass of the Milky Way black hole. Neither of these galaxies has any star formation or any visible dust structures. How massive must a central black hole be before it turns off star formation in its host galaxy altogether?

It is interesting that you mentioned the Sombrero Galaxy, M104. How massive is its central black hole?

According to Wikipedia, the central black hole of M104 may be a billion times the mass of the Sun, which makes it about 250 times the mass of the Milky Way black hole. That is really a lot, but clearly much less than the black holes of M87 and NGC 3115. And while M87 has no disk, no obvious dust structures and no star formation, NGC 3115, with its smaller hole, has a disk, but no dust structures and no star formation. M104 has a very obvious disk and very obvious dust structures. Does it have any star formation?

Wikipedia says this about star formation in M104:

Based on infrared spectroscopy, the dust ring is the primary site of star formation within this galaxy.

To me this is a very unsatisfying statement, since it says that the galaxy does contain star formation, but it makes no attempt to quantify how much of it there is. My guess is that there isn't much at all. The color indexes of M104 all point to a galaxy that is poor in star formation. The B magnitude of the galaxy is 9.075, but the B-V index is 0.980, which is red for a spiral galaxy. The overall color of the galaxy is comparable to the color of well-known K0 star Pollux in Gemini. This indicates that there aren't many A-type stars like Sirius and Vega in the Sombrero galaxy. The U-B index, which is particularly sensitive to the presence of hot OB type stars, is 0.530, which is also red. It indicates that the Sombrero galaxy has few if any OB stars. Finally, the galaxy's far infrared magnitude, which is particularly sensitive to the presence of extremely young dust-shrouded stars, is 10.925, almost two magnitudes fainter than the B magnitude. This indicates that the dust lane of M104 is mostly devoid of new-born stars.

So in my opinion there is either very, very little star formation in M104, or else there is none at all. Is there any star formation there at all? This picture of M104 does show small bluish patches and clumps in the dust lane, which might mean that there is indeed a tiny amount of star formation in M104 after all: http://apod.nasa.gov/apod/image/0803/M1 ... is2048.jpg

What about the Andromeda galaxy? There can be no doubt that there is more star formation in M31 than there is in M104. According to Wikipedia, the Andromeda Galaxy produces about one solar mass of new stars per year, which is something that the Sombrero galaxy definitely can't match. But our own galaxy is more proficient at making stars than M31, since the Milky Way cooks up 3-5 solar masses of new stars each year. What about the black holes in our galaxy and in M31? The central black hole in the Andromeda galaxy is 1.1–2.3 × 108 M☉ according to Wikipedia, which would make it 25-60 times more massive than the black hole of the Milky Way.

So the bigger the hole, the more disk-less and dust-free the galaxy will be. And the bigger the hole, the fewer new stars will be made by the galaxy.

Ann

[Ann]

Well, you sure got me talking, Doug!

And while I was talking, I found that I was actually changing my own mind about the most important mechanism behind the outflows of gas from galaxies. I know believe - or I would say that I realize - that massive black holes are the most important players here. Interestingly, the supermassive black holes, the ones that turn their host galaxies into ellipticals, don't so much expel the gas from their host galaxies as they make the gas unfit for star formation. There is a lot of gas inside supergiant elliptical galaxy M87, for example. Wikipedia writes about M87:

Between the stars is a diffuse interstellar medium of gas that has been chemically enriched by elements emitted from evolved stars. Any dust formed within the galaxy is destroyed within 46 million years by the X-ray emission from the core

I love the precise "life expectancy" of dust within M87 - 46 million years! Please note that dust is a hugely important "enabler" of star formation in spiral galaxies like the Milky Way. Dust helps gas clouds to "settle down", to cool and to begin to shrink and concentrate, so that the central mass of them can become high enough for stars to form out of the dusty gas. To see how dust enables star formation in the Milky Way, take a look at this picxture of a star forming region in the southern constellation of Corona Australis:

Click to view full size image

You can see a blue patch named NGC 6726 and 6727. These are newborn stars of late class B, wrapped in their dusty birth cocoon and making it shine blue with their own light. IC 4812 is another birthsite for a late class B star, except that this star is surrounded by an even thicker "birth cocoon". NGC 6729 and R Corona Australis are birthsites for even younger stars, which have still not "turned on their stellar engines" of hydrogen fusion. They are still glowing by contraction, as they are in the process of "getting started".

Yes, but look at the very thick dark brown structure to the upper right of the newborn and not-yet-born stars! That is the dust cloud that these stars were born out of. As the stars form, they consume a lot of dust and gas, but much of the gas and dust remains here, as you can see. This is the sort of dust that stars are born out of! But in M87, any dust that is formed is destroyed within 46 million years, which is probably far too short a time for the dust to act as "condensation points" for gas, so that the gas can gather into clouds. Instead the gas in M87 is very turbulent, unable to settle down.

Wikipedia also says about M87:

Gas is infalling into the galaxy at the rate of two to three solar masses per year, most of which may be accreted onto the core region.
...
The regular eruptions prevent a huge reservoir of gas from cooling and forming stars, implying that M87’s evolution may have been seriously affected, preventing it from becoming a large spiral galaxy.

I used to talk a lot about mergers and how they affect the star formation of galaxies, but now I have talked a lot more about supergiant black holes. But I think that the mergers and the supergiant black holes are related. I think mergers, particularly repeated mergers, can created supergiant black holes. Look at famous merging galaxies NGC 4038 and 4039:

Click to view full size image

You can see the two merging galaxies and their very long tidal tails. The "upper" galaxy (NGC 4038) is more gas-rich than the "lower" galaxy (NGC 4039), and the tidal tail that emerges from NGC 4038 is longer than the one that emerges from NGC 4039. It is probable that the shorter tidal tail emerging from the more gas-poor galaxy is mostly made up of stars. But the longer tidal tail clearly consists of both stars and gas. Note the blue "clumps" at the end of the longer tidal tail. These clumps are newborn stars, which have been formed out of gas in the tail, as this gas was "piled up" and compressed by some mechanism.

So gas has been driven out of NGC 4038. But the extremely close encounter between NGC 4038 and 4039 has sent gas flowing in many directions:

Click to view full size image

Take a look at the huge brown dust patch that has formed where NGC 4038 and 4039 have collided and "joined" (on the left in this image). A torrent of star formation is occuring here. But where is the dusty gas going?

Click to view full size image

The gas appears to be flowing straight into the center of NGC 4038. There is no evidence of a supermassive black hole in NGC 4038. Clearly, however, if huge amounts of gas are going to flow straight into the core of this galaxy, a very large black hole could well form there.

NGC 4038 and 4039 are an isolated pair. Once they are done merging with each other, they can't get any huge helpings of gas from other galaxies. But what if they had been inside a cluster? When they had merged with each other, quite possibly forming a large spiral instead of an elliptical galaxy, they might soon collide with another galaxy, and that way they may get another large helping of gas settling onto their common core. And then they may collide with another galaxy... and then another... and they may get just more and more gas flowing into their core, until the black hole there became supermassive.

All the mergers that happen inside clusters of galaxies may well explain why some galaxies in clusters grow absolutely supermassive black holes. And if the black holes get massive enough, the outbursts from them may shut down star formation not only in the galaxies actually hosting the black holes. The violent outbursts from the supermassive black hole may create such an "intra-galactic bad environment" for star formation that no new stars may be able to form in most of the galaxies in the cluster.

Ann

[dougettinger]

Hello Ann,

You are correct. We are thinking about different structures in the galaxy when we are referring to "plumes". My questions about plumes are in regards to structures situated well inside the the central regions of the disk that look like bubbles. I am not referring to the "plumes" that Wray refers which are clouds of stars that are spun off the ends of the spiral arms. Also, these bubbles or plumes that I refer to would be recent supernova remnants that have not appreciably dissipated and are still opaque. Examples that you have shown for the most part are well dissipated and one can clearly see through them.

Doug Ettinger, Pittsburgh, PA 3/23/2011

[Ann]

Let's return to the question about NGC 2841 itself, then.

Click to view full size image

You are referring to some of the dust structures here, and you are asking if many of them are made by supernovae. You are also asking if they are hiding a higher than average amount of star formation.

I don't think so. The reason why I don't think so is the B and infrared magnitudes of this galaxy.

Typically, when a galaxy has a lot of star formation, it is also quite dusty. This dust gets warm during star formation, because it is warmed by the newborn stars. Richly starforming galaxies are bright at far infrared magnitudes, and they are brighter in far infrared light than they are in blue light.

Consider M82:

Click to view full size image

Astronomers are sure that M82 underwent a violent starburst in its center due to interactions with its larger neigbour, M81. Many of the massive young stars then evolved to a stage when they went supernova more or less simultaneously. All those supernova explosions distorted M82's disk, apparently destroying its spiral pattern, and set off an enormous outflow of gas from the galaxy's poles. All the supernova explosions also created a lot of dust, which was heated both by the supernovae themselves and possibly by ongoing central star formation.

The B magnitude of M82 is 9.161, but its far infrared magnitude is 5.584. So M82 is three and a half magnitudes brighter in the far infrared than in blue light.

The relationship between the B magnitude and the far infrared magnitude is very different in NGC 2841. Its B magnitude is 10.040, and its infrared magnitude is 10.871, more than half a magnitude fainter in the far infrared than in blue light. So the dust in NGC 2841 is not particularly warm. There is not a greater than average amount of star formation going on in that galaxy, but rather the opposite - there is a bit less star formation there than in many spiral galaxies.

There is most definitely star formation going on in NGC 2841, because all the pink emission nebulae visible in the Hubble image made that clear. Obviously the dust in the galaxy helps gas in there to cool and condense enough to set off star formation. Certainly there are many pockets of dust in NGC 2841 which are warmed by the thermal emission of newborn stars. And undoubtedly a lot of that starbirth-enabling dust was created by supernovae in NGC 2841.

I compared NGC 2841 with M82, but we may compare it with M82's larger neighbour, M81, too. Like NGC 2841, M81 has a large yellow bulge. Unlike NGC 2841, M81 has long elegant spiral arms. But guess what? The color properties of M81 and NGC 2841 are quite similar. Like NGC 2841, M81 is more than half a magnitude fainter in the far infrared than in blue light. In relation to how much blue light they produce, M81 and NGC 2841 have similar amounts of star formation.

Click to view full size image

M81 by Hubble.

My point is much of the dust in NGC 2841 was probably made by supernovae, but few of the actual structures in the dusty disk of NGC 2841 are likely to have been made by supernovae.

Ann

[Ann]

I would like to show you another galaxy and describe its color properties. The galaxy is NGC 3310:

Click to view full size image

NGC 3310 is a very blue galaxy. It is also a very ultraviolet galaxy. It is also a richly starforming galaxy.

The B magnitude of this galaxy is 11.078. Its B-V index is 0.350. That is really blue for a galaxy. It means that the galaxy's overall color is considerably bluer than the color of the Sun. The galaxy has a large population of blue A stars, which outshine the yellow stars in the galaxy's core.

NGC 3310 is also a very ultraviolet galaxy. Its U-B index is -0.430. That's really very ultraviolet for a galaxy. It means that NGC 3310 has a large population of O and B stars.

NGC 3310 also has a lot of star formation, because its far infrared magnitude is one and a half magnitude brighter than its blue magnitude.

NGC 2841, by contrast, is a relatively red galaxy, both when it comes to its B-V index, its U-B index and its "blue minus far infrared" index. Its overall color is redder than the Sun, so its yellow stars outshine its blue ones. Its U-B index shows that it hasn't got many O and B stars, and its "blue minus far infared index" shows that it isn't very prolific when it comes to star formation.

Ann

[owlice]

I'm unhappy with the color balance of this picture, but Hubble often produces images where the color is not rendered in such a way that it reflects what our eyes would see, if they were a lot more sensitive to faint light than they are.

The point of Hubble is NOT to produce pretty pictures, but science, and color is used as a tool in these images. Hubble is not up there to satisfy your, nor anyone's, notions of "what our eyes would see, if they were a lot more sensitive to faint light than they are." That's an unreasonable expectation of this scientific instrument.

[Ann]

Yes, I was just going to remove that comment about the color of the picture of NGC 3310.

I put my statement badly. Instead, let's take a look at what filters were used to produce the image. The filters were F336W (U), F439W (B), F814W (I) and F300W (Wide U), F814W (I). In other words, the galaxy was imaged through two ultraviolet filters, one blue filter and two infrared filters. Although I can't find it in the "Fast Facts" box, it seems obvious that the image taken through the ultraviolet filters was colored blue, the image taken through the blue filter was colored green, and the picture taken through the infrared filters was colored red. With this in mind, it becomes easier to read the image. The outer, greenish-looking parts of the galaxy are dominated by the blue light of A-type stars. The very red center is dominated by the light of the usual suspects of galactic bulges, that is, yellowish K- and M-type stars. The very white arms contain large numbers of O-type stars, but they also contain a lot of A stars. But they are also rich in star formation and dust, which will produce a lot of dust-reddened, infrared light.

Ann

[dougettinger]

Ann, thanks again for your very detailed replies. My question is still lingering. What really are the dust structures that pervade NGC 2841 ?

Also, how do X-ray emisions from a galaxy core destroy the dust formed in the galaxy within 46 million years ? What is the physical mechanism at work ? Let's say the dust is molecules of CO2 or CH4; what happens to these molecules ?

Doug Ettinger, Pittsburgh, PA 3/23/2011

[Chris Peterson]

Also, how do X-ray emisions from a galaxy core destroy the dust formed in the galaxy within 46 million years ? What is the physical mechanism at work ? Let's say the dust is molecules of CO2 or CH4; what happens to these molecules ?

The statement (which is from Wikipedia) is incorrect, or at least incomplete.

X-rays can sputter atoms from the surface of interstellar dust, and can also cause the dust to explode into smaller components. It is recognized that interstellar dust is always being created, and is always undergoing a process of destruction- and x-rays are a key component of the destructive process. That said, the reference to 46 million years seems unsupported, and the process is much more complex than stated. The rate of dust loss is determined by the type of dust (as you suggest), by the distance of the dust from the source, by the dust replenishment rate, and other factors. Also, much of the dust "destruction" isn't really true destruction (reduction to atoms), but simply breaking up of grains, or redistribution away from energy sources.

[dougettinger]

What seems to be supported by mathematics and theory per Wikipedia is that X-rays from the core of a typical galaxy and various overlays of explosions from supernovae produce high frequency X-rays that can breakdown dust in a typical giant IMC in 40 to 50 millions years if it has not already established some density gradients and early protostar disks.

So any dust and gas clouds created by supernovae and novae must start forming protostars or protostar disks inside of 50 million years or explosion remnants will be useless for forming new star births. (?)

Doug Ettinger, Pittsburgh, PA 3/24/2011

[Ann]

My guess is that you are right about that, Doug.

I've been thinking more about what you said about the bubble-like dust structures in NGC 2841. I found this old Hubble image of NGC 2841, where bubble-like dust structures are quite obvious:

http://en.wikipedia.org/wiki/File:NGC_2 ... ikiSky.jpg

In the right-hand part of the upper left quadrant of the Hubble image, there is a squarish dust structure which appears to be filled with a grouping of brighter than average stars. My guess is that this dusty square is the remnant of a dusty gas cloud that created the bright stars that we now can see. There were several OB stars among them, and their stellar winds blew the dust away, making it pile up many light-years away from the stars, until the dust was stopped because it collided with other dust structures that had been piled up by the stellar winds of other clusters and associations. You can see that there appear to be clusters or associations of stars both to the upper left and to the lower right of the squarish dust structure.

So I think that star formation, and certainly also supernovae, created these bubble-like structures of dust by compressing it and making it collide with other dust structures in NGC 2841.

Ann

[dougettinger]

My guess is that you are right about that, Doug.

I've been thinking more about what you said about the bubble-like dust structures in NGC 2841. I found this old Hubble image of NGC 2841, where bubble-like dust structures are quite obvious:
http://en.wikipedia.org/wiki/File:NGC_2 ... ikiSky.jpg
So I think that star formation, and certainly also supernovae, created these bubble-like structures of dust by compressing it and making it collide with other dust structures in NGC 2841.
Ann

Yes, I believe these bubble-like structures are the result of the build-up and combination of star death remnants and perhaps their compression by the solar winds of recent protostars. This is further evidence for me that the debris, dust and gases, from supernovae do stay within the confines of the galaxy disk. You have mentioned that very active and starburst galaxies can evacuate their dust required to make new stars which may be special cases.

You have also mentioned previously that setting a galaxy at 15 billion years of age and estimating that blue massive stars live for 15 million years produces 1000 succeeding generations of O and B type stars. For this to occur the remnants must obviously re-combine to form more O and B type stars. What I believe happens, and I think this is behind all your ideas, too, is that more and more less massive, long-lived stars are formed after each successive generation thereby taking away more of the available dust and gases for future succeeding generations of O and B type stars. Eventually, the remnants of dying stars are not sufficient to be combined and/or compressed with enough other remnants to create anymore blue stars or even other stars. All that remains are yellow and red stars that keep living their long lives.

Of course, this is a very general concept that can be affected by galaxy mergers, other evolutionary processes, and the emission of gases by large central black holes in the form of jets. I am trying to make your arguments more congruent. I would appreciate any further comments.

3/24/2011

[Ann]

You have also mentioned previously that setting a galaxy at 15 billion years of age and estimating that blue massive stars live for 15 million years produces 1000 succeeding generations of O and B type stars. For this to occur the remnants must obviously re-combine to form more O and B type stars. What I believe happens, and I think this is behind all your ideas, too, is that more and more less massive, long-lived stars are formed after each successive generation thereby taking away more of the available dust and gases for future succeeding generations of O and B type stars. Eventually, the remnants of dying stars are not sufficient to be combined and/or compressed with enough other remnants to create anymore blue stars or even other stars. All that remains are yellow and red stars that keep living their long lives.

Exactly!

In his book "Planet Quest", Ken Croswell made the point that contrary to what we think, our Sun is a relatively unusual kind of star. What makes it unusual is its mass and luminosity. Ken Croswell says that 80% of the stars in our galaxy are M-type red dwarfs, which typically contain perhaps half as much mass as the Sun. 9% of all stars are K-type dwarfs, and 5% are white dwarfs. All these stars, 94% of all stars in our galaxy, are fainter and less massive than the Sun!

What is more, it is likely that all these stars will basically hang on to almost all their mass for several times the current lifetime of the universe - indeed, when it comes to the M-type dwarfs, I'm not entirely sure that they are ever going to evolve enough to give back a significant amount of their mass to the galaxy. (And it is indeed likely that our galaxy will be destroyed long, long before even the first M-type dwarf that was ever made by our circa 14 billion-year-old universe has evolved to the point that it leaks any appreciable amount of gas.)

So when an M-type dwarf is made, the gas that went into making it will stay inside the red dwarf for many, many times the current age of the universe. For all intents and purposes, it will be locked up in there forever.

And the M-type dwarfs are incredibly numerous. They are staggeringly numerous if you count the number of M-type dwarfs in large elliptical galaxies. Very recently it was found that there are so many red dwarfs inside huge ellipticals that the total star count of the entire universe has to be multiplied by a factor of three!

So the small M-type dwarfs are in their own little way the most common "bottomless holes" of the universe, objects that consume nourishment and never give it back. Imagine if we had some sort of little rodent or something here on Earth which sapped nourishment from the soil in order to get born. Once born they never ate, but they also never "defecated" and never died. What nourishment they had consumed to get born remained within them, and there was progressively less and less nourishment for other organisms to live on.

Ann

[dougettinger]

Ann, thanks for providing a very convincing timeline or history for the observable galaxies. I did not think there would be any ideas of a galaxial history in my lifetime. What remains is one of the bigger mysteries of galaxies. How did globular clusters, some the oldest structures in the universe, form ? Why do they exist spherically around both elliptical and spiral galaxies ?

3/26/2011

[owlice]

Globular Clusters: Ancient Stellar Relics

How Did Globular Clusters Form?

How, Where and When did the Globular Clusters form?

[Ann]

Thanks, Owlice!

It is not hard to imagine that present-day mergers may create "super star clusters" that may evolve into future globulars. Owlice's third link showed Hubble pictures of massive young clusters in NGC 4038, which have formed out of the collision between NGC 4038 and 4039. Another example is Stephan's Quintet:

Click to view full size image

The colliding yellow galaxies on the right, NGC 7318a and 7318b, are surrounded by a huge "ring" of violently expelled gas which has fragmented and formed large numbers of bright blue clusters. Some of these clusters are undoubtedly very massive. (The galaxy at lower left is a foreground object.)

Bright and massive clusters can form also in galaxies which are not colliding or merging. A favorite example of mine is the super star cluster in NGC 6946. A portrait of this galaxy was Astronomy Picture of the Day on January 25, 2005:

http://apod.nasa.gov/apod/ap050125.html

The super star cluster stands out at eleven o'clock in the image above. Here is a closeup of the cluster:

http://www.astro.uu.nl/siu/people/N6946PC.GIF

While the super star cluster is a unique object in NGC 6946, the galaxy is forming stars at a high rate, and it is producing a very remarkable number of supernovae. My software mentions six supernovae in NGC 6946 between 1917 and 1980, but I'm sure there have been at least two more since 1980. Typcially for a galaxy which is currently or has recently produced a lot of stars, NGC 6946 is quite dusty. Its far infrared magnitude is two magnitudes brighter than its blue magnitude, which is quite a lot for a face-on galaxy. The APOD caption suggested, as you could see, that a possible reason for the high rate of star formation of NGC 6946 might be the accretion of gas clouds from the surrounding region.

Our closest galactic neighbour, the Large Magellanic Cloud, also has a at least one super star cluster, R136:

Click to view full size image

The Large Magellanic Cloud is forming large numbers of stars. The more richly starforming a galaxy is, the greater the chance that it will form at least one really massive cluster. Sidney van den Bergh wrote in the paper that Owlice gave us a link to:

These data show that the fraction of the U light of galaxies that is generated
by young clusters is proportional to the rate of star formation per unit area to the power
~1.2. This result suggests that the present specific cluster frequency in galaxies will, to a
large extent, be determined by the peak rate of star formation per unit area during their
evolutionary history. The high S values observed in some central galaxies of rich clusters
might therefore be due to a high surface density of star formation early in their history.
By the same token the observation that elliptical galaxies have higher mean S values than
spirals may be interpreted to mean that the peak rate of star formation in E galaxies was
higher than that in disk galaxies.

Here van den Bergh says that richly starforming galaxies will stand a better chance to form a really large cluster. He also says that the peak rate of star formation in elliptical galaxies was higher than in disk galaxies, which agrees with what I said earlier about many elliptical galaxies having started out as ULIRGs, Ultra Luminous Infrared Galaxies, which were once extremely dusty because of their incredible star formation.

As for the Large Magellanic Cloud, it has apparently never been a very early ULIRG. It does not show any signs of very early extreme star formation. Sidney van den Bergh has shown earlier that the LMC experienced some star formation very long ago, about twelve billion years or so, but then it was more or less quiet for billions of years. However, a few billion years ago it started forming stars again, and about two billion years ago this star formation seems to have peaked, according to van den Bergh, although star formation is still vigorous in the LMC. What happened to this galaxy to turn it into such a star factory so suddenly? Clearly the LMC is interacting quite strongly with the SMC, the Small Magellanic Cloud, and it could be that this interaction started when the star formation of the LMC picked up. But the Magellanic clouds are obviously affected by their close proximity to the Milky Way, too. But the Magellanic galaxies appear not to have been "original" satellites of the Milky Way, and they may not be true satellites of the Milky Way even today. They are apparently passing us by at a high speed, too high for them to be gravitationally bound to us.

Anyway the interaction between the LMC and both the SMC and the Milky Way has definitely made star formation inside the LMC increase. And because it is forming so many stars, it can also find a few truly massive clusters. Wikipedia writes about R136, the most massive young cluster of the LMC:

One of the stars, R136a1, is the most massive star found to date with 265 solar masses,[7] as well as the most luminous at 10,000,000 times the brightness of the Sun.[8] R136 produces most of the energy that makes the Tarantula Nebula visible. The estimated mass of the cluster is 450,000 solar masses, suggesting it will probably become a globular cluster in the future.[9]

So the estimated mass of the cluster is 450,000 solar masses. While impressive, that is not an overwhelmingly large mass after all, not compared with old globular clusters. Wikipedia says that Omega Centauri, the largest globular of our own galaxy, contains an estimated mass of 5,000,000 solar masses. Of course, Omega Centauri may be a special case, since it may not be a "true" single-generation globular cluster, but instead it might be a captured dwarf galaxy. If so, what is left of that dwarf galaxy now is the bulge, which would now orbit the Milky Way mimicking as a globular cluster. So Omega Centauri might not be fully comparable to other globular clusters. However, a "true" and massive MIlky Way globular like M13 contains an estimated 600,000 solar masses, and M13 is one of those truly ancient globulars, only slightly younger than the universe. When R136 gets to be thirteen billion years old, it will most certainly have lost much of its mass. So while R136 is dazzlingly bright by today's standards, M13 was much more dazzling and fantastic at the dawn of time when it was born.

I'll be back with more here, I think.

Ann

[Ann]

Okay, my point is that very many globular clusters seem to have formed more or less at the dawn of time. These clusters are "blue", which means that they contain so few heavy elements that stars in them experience an evolutionary phase called "the blue horizontal branch", which is a phase when the stars are moderately bright, far brighter than the main sequence stars of comparable age, and they are also blue or bluish. In more metal-rich clusters (clusters containing a higher amount of elements heavier than hydrogen and helium) stars never reach the "blue horizontal branch", but stay on the "red horizontal branch" instead. As I said, clusters which lack a blue horizontal branch are more metal-rich, but that should also mean that they are younger. Either that, or else they were formed by some other mechanism, or during different circumstances, than the metal-poor clusters.

Here is a Hubble image of Messier 13, which is an age-old globular, perhaps 13 billion years old, with a lot of medium-bright blue horizontal branch stars: http://www.spacetelescope.org/static/ar ... w1011a.jpg

Here is a Hubble image of 47 Tucanae, the second brightest globular of the Milky Way: http://imgsrc.hubblesite.org/hu/db/imag ... ll_jpg.jpg

47 Tucanae is estimated to be 10 billion years old, according to Wikipedia. It is definitely mor metal-rich than M13, and it lacks blue horizontal stars.

I thought it was most interesting that a person called R. Cen proposed that the blue globular clusters may have formed during the epoch of reionization, when ionization fronts collided with the halos of proto-galaxies. What is appealing with this scenario is that it proposes a mechanism which affected most of the universe at more or less the same time (and very early on), which would explain why so many globular clusters are about twelve billion years or more, and why these very old globulars have such similar properties. It is perhaps surprising that these very old globulars contain any metals at all. But perhaps it is not surprising at all, if the reionization happened as a result of the formation of the earliest stars, the presumably super-bright Population III stars. If some of these stars had already exploded as supernovae when the ionization fronts really got going, then that might explain why the gas that was available for star formation at that time was already somewhat "polluted" with metals.

Ann

[dougettinger]

I now have a vision of globular clusters being formed. Thanks to you and Jabberwren, I have read Owlice's references and articles about metallicity and reionization. After the Big Bang and the inflationary period baryons were created because considerable cooling occurred causing electrons and protons to combine and form clouds of molecular H and He. Density gradients occurred due to shrinkage and condensation on a galaxial scale. The largest clouds shrunk to create massive stars and reionization. The stars formed elliptical and spiral galaxies at the centers of the largest clouds; these core galaxies shrunk faster than their less dense neighboring, perimeter clouds that were still gravitationally bound. These smaller less dense clouds of gas remained gravitationally in their radial positions w.r.t to the central core galaxy and started to shrink on their own thereby creating globular clusters of stars.

These globular clusters are prone to have less generations of blue, massive stars because of their original low density of gas and smaller size. Hence, globular clusters have lower metallicity than the stars in the core galaxies that have processed as many as 1000 successive generations of O and B type stars over 13 billion years thereby manufacturing more metals and Population I stars. During galaxial collisions over the succeeding 13 billion years the core galaxies became more perturbed and continued to evolve while leaving many surrounding globular clusters undisturbed. Also, globular clusters were stripped from other core galaxies that had close encounters with other core galaxies. Globular clusters for the most part continued to be less disturbed and show the same characteristics that they possessed near the end of the reionization period 12 to 13 billions years ago.

Ann, I hope you like my summary. I think you are ready to write your paper.

03/27/2011