# I How close do stars get?

#### zuz

The closest star tp the sun is 4 light years away. Are there any stars that are closer that are not part of a multi=star system? How close close can stars get to each other and still be independent systems?

Related Astronomy and Astrophysics News on Phys.org

Staff Emeritus
"Independent systems" is a statement about velocity and gravitational binding. "Close" is a statement about position. They cannot be directly compared.

#### trurle

Gliese 710 will pass 0.2 light-years from Sun in 1.3 mln. years from now.

#### DaveC426913

Gold Member
The closest star tp the sun is 4 light years away. Are there any stars that are closer that are not part of a multi=star system?
No.

The fact that the Centauri system is a trinary says nothing about the fact that its the closest star system to us.

How close close can stars get to each other and still be independent systems?
In theory, they can get close enough to distort each other and still be independent - if it were a flyby on a hyperbolic path.

#### russ_watters

Mentor
"Independent systems" is a statement about velocity and gravitational binding. "Close" is a statement about position. They cannot be directly compared.
With the exception one little part, they aren't really being compared, they are simultaneous but separate criteria. Reading between the lines, the answer would be there is no limit. A follow up might be; what has been observed?

The difficulty is time: gravitational binding is a long term phenomena whereas a flyby or collision happens fast and is easy to miss.

Staff Emeritus
Reading between the lines, the answer would be there is no limit.
If stars were points, that would be true. You tell me how close you want it, and I'll tell you how fast it needs to be going.

#### russ_watters

Mentor
If stars were points, that would be true. You tell me how close you want it, and I'll tell you how fast it needs to be going.
The OP isn't asking about theory, he's asking about reality. Again, reading between the lines (I shouldn't have to), the answer I'm seeing is "arbitrarily close to zero". So, how close have we observed in reality? I suspect the answer due to the timing issue is not very close at all.

Staff Emeritus
Stars have collided. That's pretty close.

#### russ_watters

Mentor
Stars have collided. That's pretty close.
Have we observed stars that were not gravitationally bound collide? It must be an extremely rare thing to observe.

Still, this violates the OP's criteria, so while I'd be interested to know, it doesn't answer the question.

Staff Emeritus
Blue stragglers (at least some of them) are thought to be collisional. This is a complicated business, but the velocity distributions of at least some subsamples are different than that of typical stars, which is not what you'd expect if the sole formation source was mass transfer from pre-existing binaries.

Does this "count"? Well, Star A is not bound to Star B, but both are bound to their globular cluster as a whole. But I maintain that if that doesn't count, no stars in galaxies count, and that's practically all of them.

mfb

#### sophiecentaur

Gold Member
still be independent systems?
How independent? In fact what do you mean by the word? Orbits (two body) that are circular or elliptical could be called not independent but when the orbits are hyperbolic, the kinetic energy is greater than the potential energy at the closest point. A close approach perhaps but the two could be called independent.
But the point has already been made that stars tend to appear in galaxies or clusters and there is a multi-body solution.

#### DaveC426913

Gold Member
I'd say it's a fair assumption that the OP is considering 'independent' to mean not part of a stable binary or trinary system, like in his example of A. centauri.

#### zuz

Sorry for the discrepancy. Yes, I meant not part of a binary system, and yes I meant what has been observed. I know all stars are moving around in space. I just wondered what was the minimum distance observed between two stars was right now. Also, is the Centauri system moving toward or away from us? Thanks.

Staff Emeritus
In that case, the minimum distance is zero because stars (rarely) collide.

#### DaveC426913

Gold Member
Also, is the Centauri system moving toward or away from us? Thanks.
"... the mean radial velocity has been determined to be around 22.4 km/s towards the Solar System."

That gives you about ... (...furlongs in a cubit ... carry the one...) 57,500 years to cover your roof in reflective paneling.

Last edited:

#### Nugatory

Mentor
Have we observed stars that were not gravitationally bound collide? It must be an extremely rare thing to observe
It’s less uncommon in globular clusters, where the density of stars is much greater.

#### snorkack

In that case, the minimum distance is zero because stars (rarely) collide.
No, because can you point at observed star collision?
Not a blue straggler suspected to have formed by collision some time before observation (and it is argued whether blue stragglers might form by inspiral of binaries previously bound) - no, actual observed collision, constellation, year, date, properties of each precursor and resulting star?

And even that would not be distance between two stars right now because a past observed collision is one star now. Are there any stars observed on collision course that are now unbound (not on inspiral course)?

#### eachus

Globular cluster - Wikipedia

The typical distance between stars in a globular cluster is about 1 light year, but at its core, the separation is comparable to the size of the Solar System (100 to 1000 times closer than stars near the Solar System). Globular clusters are not thought to be favorable locations for the survival of planetary systems.

So looking at nearby globular clusters today, we don't see any stars colliding. But it should be possible to calculate the average time between collisions which result in mergers. Yes, these are thought to be so-called blue stragglers. (The straggler part is that stars that blue and bright could not have formed when the rest of the globular cluster did. Assuming that they were created recently in collisions works.)

Are there any stars observed on collision course that are now unbound (not on inspiral course)?

No. It is very tricky since, observing from Earth we can measure two dimensions very accurately, but the third dimension will have an error of light years. If two stars are orbiting each other, that allows us to determine both that they are bound, and give a better estimate of distance.

Would it be possible to work out all the gravitational interactions in a globular cluster, and predict which stars will collide when? Not today. Quantum computers with millions of qubits might make it possible.

#### DaveC426913

Gold Member
No what?

No, the minimum distance is not zero? Or no, stars don't occasionally collide?

because can you point at observed star collision?
So what?
Not having observed such an event doesn't negate anything Vanadium asserted.

Not sure what your case is here.

Staff Emeritus
can you point at observed star collision?
Nope. I can't point to a living Tyrannosaurus Rex either. However, I can point to their remnants.

#### Janus

Staff Emeritus
Gold Member
No, because can you point at observed star collision?
Not a blue straggler suspected to have formed by collision some time before observation (and it is argued whether blue stragglers might form by inspiral of binaries previously bound) - no, actual observed collision, constellation, year, date, properties of each precursor and resulting star?

And even that would not be distance between two stars right now because a past observed collision is one star now. Are there any stars observed on collision course that are now unbound (not on inspiral course)?
As of present, we have catalogued ~200 million of the ~200 billion stars in our galaxy alone, or roughly 1/1000 them. Not seeing a remnant of a star collision in those catalogued stars merely puts an upper limit on the number of star collisions that could have occurred in the galaxy but does not eliminate their existence entirely. (so for example, if there is, on average, 1 collision per 100 million years or so, then our galaxy might have 130 collision remnants and the odds of one of those being in our 200 million star sample are fairly low. )

Lack of evidence is not always evidence of lack.

Last edited:

#### snorkack

There have been a number of observed inspirals in a few years. One of them formed a kilonova. That was two neutron stars merging.
Do collisions between main sequence stars cause any brightening?

Staff Emeritus
There have been a number of observed inspirals in a few years.
But they were gravitationally bound to each other, and furthermore, don't seem to address your earlier point of "no". Could you answer Dave's question?

Staff Emeritus
Not seeing a remnant of a star collision in those catalogued stars merely puts an upper limit on the number of star collisions that could have occurred in the galaxy but does not eliminate their existence entirely.
I agree with that, but am not sure it's the very best example. First, how would we know a given star had been in a collision? It's not like it would have a dent in it or would be patched with Bondo. Second, if one compares the volume swept out by stars over a couple of billion years to the volume of the disk, it's much less. So we don't expect any collisions in the disk. Maybe the core, but things are both more uncertain calculationally, but also harder to see.

All of these problems go away when one looks at blue stragglers (BSSs). These stars are in globular clusters (GCs) and they are brighter and bluer than their neighbors, and by implication, younger. GCs are very dense collections of very old stars - star formation has stopped billions of years ago. There is both strong evidence for them being collision remnants and weak evidence. The stronger evidence is that if they were main sequence stars, they are much younger than their GC. BSSs are of varying calculated ages, so are not the product of some "microburst" of star formation. For at least some of them, their velocity distribution differs from stars in the cluster, as if they had an event that transferred a lot of momentum to them in the past. The weaker evidence is that they tend to be about twice as large as their neighbors and rotate faster: just what you expect from collisions, but also a purely statistical measure. One can find heavy stars and fast rotating stars in our own galaxy, where we don't think they have collided. So I cast this argument as weaker, since it is purely statistical.

#### stefan r

Gold Member
As of present, we have catalogued ~200 million of the ~200 billion stars in our galaxy alone, or roughly 1/1000 them. Not seeing a remnant of a star collision in those catalogued stars merely puts an upper limit on the number of star collisions that could have occurred in the galaxy but does not eliminate their existence entirely. (so for example, if there is, on average, 1 collision per 100 million years or so, then our galaxy might have 130 collision remnants and the odds of one of those being in our 200 million star sample are fairly low. )

Lack of evidence is not always evidence of lack.
Can we determine the cause of a collision when we look at remnants? For example CK Vulpeculae, can we be sure that was a binary merger? I realize that the odds of a binary merging are much much higher than a random impact. What would look different if the rare event did take place?

...
Do collisions between main sequence stars cause any brightening?
Yes, a great deal of brightening. Then the source should be snuffed out by dust and clouds.

There should be a very large range of possible outcomes from a hyperbolic collision. If you really lined them up "dead nuts" core-to-core it should make an exceptionally large explosion. Fusion already happens in the cores and perfect collision will add temperature and pressure. You could also have a situation that would be a near miss if stars were billiard balls. Suppose, for example, two G-stars have their center of masses exactly 2 solar radii away from each other at perihelion. They would still exchange matter and some material would be flung out of orbit. Or flung into orbit and then blown out on the solar wind. The impact would be effected by the star's rotation rate, direction of travel, and composition. Even if two stars completely missed and did not exchange material there should be some strong tidal effects. Just elongating a star should make it brighter because of increased surface area.

The outer surface of a star is much hotter than the interior. Anything that causes extreme turbulence and mixing should bring hot material outward. Gasses cool when they expand like in red giants but that is still an increase in brightness.

If we just dropped a 1 kg brick into the Sun it should be moving at 617 km/s. Energy is 1.9 x 1011 J. That is less than the core of nuclear bomb but would still be very hot. 10s of millions of degrees K. Any fusion that takes place during the impact would add energy. Meteors are white hot even though they are just falling into Earth's gravity.

"How close do stars get?"

### Physics Forums Values

We Value Quality
• Topics based on mainstream science
• Proper English grammar and spelling
We Value Civility
• Positive and compassionate attitudes
• Patience while debating
We Value Productivity
• Disciplined to remain on-topic
• Recognition of own weaknesses
• Solo and co-op problem solving