Hubble Discovers a Strange Collection of White Dwarf… Dwarfs
Written by Ian O'Neill
Small helium white dwarfs can be caused by a binary partner (NASA)
A collection of very odd white dwarfs have been discovered in a local globular cluster. Twenty-four white dwarfs (18 of them are new discoveries) have been spotted. Although these degenerate stars aren't exactly an uncommon (they are the small sparkling remnants left over after star death), this particular set are unique; they are made from helium, rather than the "standard" carbon and oxygen. And they are small, even smaller than the smallest dwarfs.
How did this dense cluster of old stars evolve? It turns out their stellar material is being stolen, stifling their development…
"Helium-core white dwarfs have only about half the mass of typical white dwarfs, but they are found concentrated in the center of the cluster," said Prof. Adrienne Cool, from San Francisco State University, in a paper to be published in the Astrophysical Journal in July. "With such low masses, the helium-core white dwarfs ought to be floating all around the cluster, according to theory. The fact that we find them only in the central regions suggests that they have heavy companions — partner stars that anchor them to the cluster center."
Cool and co-author Rachel R. Strickler believe they are seeing a case of stellar plasma theft by companion binary stars in the NGC 6397 cluster, approximately 7,200 light years away. These binary partners not only anchor these strange-looking white dwarfs in the centre of the cluster, they also have a huge role to play during the dwarfs evolution.
Before a white dwarf emerges from a planetary nebula, the parent star will have gone through the red giant phase (a phase our Sun is expected to go through in 4-5 billion years time). If this red giant has a binary partner (which seems to be the case of the 24 white dwarfs in this study), the outer layers of the puffed-up giant will be stripped away by the partner, stifling the red giant's evolution. As mass is lost, the giant never gets the chance to burn helium and then progressively heavier elements such as carbon and oxygen in and around its core. Helium then becomes the key component of these smaller-than-usual white dwarfs.
"This is the first time that helium-core white dwarf stars have been discovered in partnerships with other white dwarfs in a globular cluster," Cool said. "This large sample allows us to answer questions about the mass and nature of the partner stars, and the prevalence of these kinds of binaries in the globular cluster."
Binary stars are known to affect their partners fairly radically, they are even known to slow or even stop the development of black holes, stripping the outer layers of the dying star, stifling black hole development by removing mass from the parent star. However, not all questions have been answered.
From Cool's calculations, 5% of the stars found in NGC 6397 should end their lives as dim helium-core white dwarf stars, but after studying Hubble data, many of these tiny dwarfs are missing. "It's possible that these helium-core white dwarfs cool so slowly that they haven't had time to get very faint yet," Cool said.
There remains the possibility that the oldest binaries containing helium-core white dwarfs have actually been destroyed by interactions with other stars in the cluster. Regardless, this is a fascinating area of study. To understand how these ancient stars evolve will not only aid the development of globular cluster models, but it will provide an invaluable insight to how binary stars influence their partners.
Source: EurekAlert!
Filed under: Astronomy, Hubble
Related stories on Universe Today
Comment policy: Be nice and brief. Don't advertise your stuff, or promote your personal theories. We'll delete any comments that break these policies. Click here for more details.
April 25th, 2009 at 4:52 am
Why do helium core dwarf stars cool so slowly. Do they have a large heat capacity?
Lawrence B. Crowell
April 25th, 2009 at 10:06 am
Lawrence B. Crowell:
Late Stages of Evolution For Low-mass Stars:
April 25th, 2009 at 11:25 am
Analyzing the posted story and the linked report it appears that the presence of heavy companions can only be inferred: "The fact that we find them only in the central regions suggests that they have heavy companions"
A reference to a "partner star" is made, but no mention of any detectable signal from the partner star.
So apprently the "companions" emit no electromagnetic signal to indicate their presence.
What kind of "star" fails to emit any detectable emissions? Neither the paper or the story addresses the reason for why these "heavy companions" are undetectable.
What is the background theory for why helium white dwarfs would be presumed to be more spread out in the cluster? As stated: "With such low masses, the helium-core white dwarfs ought to be floating all around the cluster, according to theory."
Could there be a reason why the these helium-core white dwarf are concentrated in the "central regions" independent of the proposition that of the presence of a "heavy companion" that is undetectable?
Could there be a another unknown characteristic of that region of the cluster that promotes the propagation of these types of stars?
Or is it possible that there is something about these helium-core white dwarfs that we don't fully understand, possibly how they form?
It should be noted that "its [white dwarf] outer layers spilled onto the companion" is only an inference which seems tenuous until an explanation is given for why the "partner star" is undetectable.
April 25th, 2009 at 2:13 pm
Anaconda:
Behold… The Peculiar Millisecond Pulsars in the Globular Clusters NGC 6397 and NGC 6752
April 25th, 2009 at 2:20 pm
There's more…
An Eclipsing Millisecond Pulsar with a Possible Main-Sequence Companion in NGC 6397
April 25th, 2009 at 2:28 pm
Even more…
Chandra Exposes the Core-collapsed Globular Cluster NGC 6397
April 25th, 2009 at 4:07 pm
@ Ivan3Man:
I appreciate the links, particlularly the full papers.
But the problem is that the observationalists in the profiled paper in the post don't mention detecting any kind of signal. The paper you link on the same cluster, NGC 6397, only talks about one pulsar and that does send out an electromagnetic signal.
Presumably the authors of the profiled paper had access to the paper you link to, yet they make no reference to that paper, or any of the observations in it.
So, we're still dealing with an "inference" of undetectable "partner stars".
So, again, while I appreciate the links you provide, they shed no evidenciary light on the present observations and the conclusion of invisible "partner stars" still seems tenuous until an explanation is given for why the “partner stars” are undetectable.
April 25th, 2009 at 6:30 pm
I wrote a reply, Anaconda, but it disappeared (maybe because it included a URL?).
Anyway, the net is simple: the Cool et al. paper is scheduled to be published, in ApJ, in July, and it seems there is no preprint on arXiv … so no one (other than the authors of the paper) are in a position to comment further on what you write.
I did find that the presented something at the 210th AAS meeting (in May, 2007) which seems to cover much the same topic …
April 25th, 2009 at 8:27 pm
Thanks for the head's up on the preprint, Nereid. My guess is they linked positions of the He dwarfs to radio emissions from the pulsars or X-ray emission from the binary systems. This would mean that EM radiation is emitted by these "invisible' partner stars. (I would guess probably all 3 papers are referenced in the new study, but we'll have to wait to see a copy of it.)
April 26th, 2009 at 4:28 pm
IVAN3MAN, I am pretty aware of the evolution of such stars. It just struck me as odd that these hewhite dwarf stars would cool more slowly.
Lawrence B. Crowell
April 26th, 2009 at 4:28 pm
hewhite, should read helium core
April 27th, 2009 at 12:01 pm
@ Lawrence B. Crowell,
I'm sure that you're aware of the evolution of such stars, I was merely pre-empting the response from the "Electric Sun" nutcases!
April 27th, 2009 at 11:47 pm
Perhaps this is the arXiv preprint?
"Helium-Core White Dwarfs in the Globular Cluster NGC 6397", with authors R. R. Strickler, A. C. Cool, J. Anderson, H. N. Cohn, P. M. Lugger, A. M. Serenelli.
In case I can't post URLs yet, the reference is arXiv:0904.3496v1.
If anyone's interested, we can discuss the preprint, rather than relying upon the UT story …
April 27th, 2009 at 11:48 pm
Here's the URL: http://arxiv.org/abs/0904.3496
April 28th, 2009 at 2:23 pm
@Nereid: Thanks for the link to the preprint version of this paper. The authors state that this is "the first extended sequence of Helium White Dwarfs found in a globular cluster" They also conclude that the Helium Dwarfs near the center of post core-collapse cluster require "heavy White Dwarf companions" to occupy their present position in NGC 6397. BTW, 2 of the 3 papers cited above by IVAN3MAN are referenced in this paper.
April 29th, 2009 at 10:47 am
@Anaconda: I've now read the preprint, and can answer your questions (well, I can have a go at answering them).
But first I should say that full answers would likely be very long, unless you already have a good grasp of the relevant parts of modern astrophysics. Also, I urge you to read the preprint yourself, as Jon Hanford seems to have done, and come back with any questions you may still have after reading it.
"A reference to a “partner star” is made, but no mention of any detectable signal from the partner star.
So apprently the “companions” emit no electromagnetic signal to indicate their presence.
What kind of “star” fails to emit any detectable emissions? Neither the paper or the story addresses the reason for why these “heavy companions” are undetectable."
The preprint spends quite a bit of time on just that; in a nutshell, the expected mass of the unseen partner stars is ~1 sol; if the partners were main sequence stars or neutron stars, they'd be seen in the visual waveband or radio or x-ray, so they are likely other white dwarfs.
"What is the background theory for why helium white dwarfs would be presumed to be more spread out in the cluster? As stated: “With such low masses, the helium-core white dwarfs ought to be floating all around the cluster, according to theory.”"
As you note, it's not that He WDs are special; any kind low-mass star would be expected to be not centrally concentrated in a globular cluster like NGC 6397. The reason is simple: cooling. These globular clusters have essentially zero net angular momentum and are quite compact (so not easily shredded by tidal encounters with their parent galaxies). However, the stars behave pretty much as collisionless points constrained only by their mutual gravitation; except for the fact that many were born as binaries. In a 'hard' collision between a binary and a single star, on average, the least massive star gets a kick, and the heaviest one loses speed wrt the centre of mass of the cluster. Thus low-mass stars tend to get less centrally concentrated, over time, and high-mass ones more. There is a great deal of experimental and observational backing for these general findings.
"Could there be a reason why the these helium-core white dwarf are concentrated in the “central regions” independent of the proposition that of the presence of a “heavy companion” that is undetectable?
Could there be a another unknown characteristic of that region of the cluster that promotes the propagation of these types of stars?
Or is it possible that there is something about these helium-core white dwarfs that we don’t fully understand, possibly how they form?"
Yes, of course. That's true of just about every area of active research, not only in astronomy! And Cool et al. consider several other mechanisms.
April 29th, 2009 at 10:50 am
@ Nereid:
Thank you for the link and I note the additional information provided by Jon Hanford.
I allow that Hanford has identifed the operable phrases and linked them together: “the first extended sequence of Helium White Dwarfs found in a globular cluster” They also conclude that the Helium Dwarfs near the center of post core-collapse cluster require “heavy White Dwarf companions”.
But the use of the word "require" reaffirms my position that there is only an inference of "heavy white Dwarf companions".
These supposed objects have no directly detectable electromagnetic spectrum signal.
Nereid, do you disagree my analysis?
April 29th, 2009 at 11:15 am
@ Nereid:
Nereid presents my [Anaconda's] statement: “A reference to a “partner star” is made, but no mention of any detectable signal from the partner star. So apprently the “companions” emit no electromagnetic signal to indicate their presence. What kind of “star” fails to emit any detectable emissions? Neither the paper or the story addresses the reason for why these “heavy companions” are undetectable.”
And Nereid responds: "The preprint spends quite a bit of time on just that; in a nutshell, the expected mass of the unseen partner stars is ~1 sol; if the partners were main sequence stars or neutron stars, they’d be seen in the visual waveband or radio or x-ray, so they are likely other white dwarfs."
Nereid, it appears that the direct response to my statement would be: "Yes, you are right, there are no detectable signals from the supposed companion stars, it is only an inference."
Nereid, do you disagree with this proposed concise answer?
Is there evidence that white dwarfs can be undetectable? What is this evidence?
Nereid states: "Yes, of course. That’s true of just about every area of active research, not only in astronomy!"
I appreciate your answer — perhaps, you would extend the same courtesy to theories and hypothesis that rely on the Fundamental Force of electromagnetism.
April 29th, 2009 at 5:53 pm
@Anaconda: the direct answer to your direct question ("Nereid, do you disagree with this proposed concise answer?") is yes, I disagree.
If you are interested in focussing on the logic chain that lead Cool et al. to conclude that the 24 He WDs in NGC 6397 likely have WD companions (i.e. are WD-WD binaries), I'd be happy to walk you through that, per the preprint. Fair warning: it will, necessarily, involve quite a lot of material, and the restrictions of these UT comments will likely make that a slow process.
On the other hand, if you are looking for a Yes/No answer to the question of whether the ACS instrument on the HST recorded photon detections that could, plausibly, be attributed to WD companions to the 24 He WDs the Cool et al. paper discusses, then the simple answer is Yes. However, such a tiny bandwidth approach may be unsatisfying to you (it certainly is to me!).
And independent of either of the above, you seem to be conflating "detectable" with "detected". In more detail: with some work, it should be relatively straight-forward to write a clear, quantitative, testable hypothesis concerning the 'heavy' partner stars of the 24 He WDs in NGC 6397 that are the topic of Cool et al.'s preprint. From that – or those – hypotheses it should be quite straight-forward to write a proposal for observing time on the HST (or some other facility), and to expect that if such a proposal were to be granted that the hypothesis would be thus tested.
But there's an even deeper issue that I feel is lurking here, to do with "only an inference".
So let me address this head on, with a concrete example: if you take a spectrum of a planetary nebula, you will almost always find two prominent green lines. These are commonly referred to as the 495.9 and 500.7 nm [OIII] lines. To your way of approaching astronomy, are such designations only inferences? If not, why not?
April 29th, 2009 at 10:40 pm
@ Nereid:
Regarding your example: No, I would say those are detected observations that are measured and are not "inferences". The relevant question is what is the analysis and interpretation on what physical element they correlate to.
With all due respect, I disagree with your assessment that the authors' conclusions are not an inference.
You offer no direct observation & measurement of physically detectable signal from the "star partners".
I appreciate your offer at detail, but that is not necessary. Just state what direct observation & measurement of the large companion white dwarfs were reported by Cool, et al.
If you can't report a direct observation & measurement, then it is an inference based on circumstantial evidence, aka, indirect evidence based on assumptions.
The relevant question then becomes, do the assumptions reflect reality?
The position, relative to the over all cluster, of the observed white dwarfs is only an inference that they have undetectable companions.
I'll throw this out there — binary white dwarf pairs seems like a possibility.
Then you have to ask: Why is the smaller of the pairs observable and the larger of the pairs unobservable?
April 30th, 2009 at 1:30 am
Thanks for the reply, Anaconda, I think it helps get at least the two of us closer to a base of common understanding that we can build on, to have a meaningful conversation.
However, you still seem to be conflating "detectABLE" and "detectED", and now also observABLE and observED.
But first, let's make sure we are on the same page wrt inferences, and in particular "the analysis and interpretation on what physical element they correlate to".
I'll continue with the 495.9 and 500.7 nm [OIII] lines.
What "physical element" do these correlate to? Why a pair of atomic transitions of doubly ionised oxygen (that's what the "OIII" is shorthand for), and forbidden transitions to boot (that's what the square brackets are shorthand for).
Now these transitions have never been observed in any lab, here on Earth, coming from a plasma which contains oxygen in a vessel in said lab.
So how come the lines are designated [OIII], if no one has produced them in any controlled experiment in any lab here on Earth?
Once I understand the extent to which you are prepared to accept things from contemporary astronomy as observations, vs inferences, I can re-phrase my answers to accord with your standards.
Now to get there we may have to take a detour, and spend some time on your simple, yet profound, question "do the assumptions reflect reality?" But not wrt the Cool et al. paper, rather wrt contemporary astronomy in general.
April 30th, 2009 at 6:34 am
Anaconda, here is an example, from our recent comments, of us talking past each other:
===================================
YOU: Nereid presents my [Anaconda's] statement: “A reference to a “partner star” is made, but no mention of any detectable signal from the partner star. So apprently the “companions” emit no electromagnetic signal to indicate their presence. What kind of “star” fails to emit any detectable emissions? Neither the paper or the story addresses the reason for why these “heavy companions” are undetectable.”
And Nereid responds: “The preprint spends quite a bit of time on just that; in a nutshell, the expected mass of the unseen partner stars is ~1 sol; if the partners were main sequence stars or neutron stars, they’d be seen in the visual waveband or radio or x-ray, so they are likely other white dwarfs.”
Nereid, it appears that the direct response to my statement would be: “Yes, you are right, there are no detectable signals from the supposed companion stars, it is only an inference.”
Nereid, do you disagree with this proposed concise answer?
===================================
ME: the direct answer to your direct question (”Nereid, do you disagree with this proposed concise answer?”) is yes, I disagree.
===================================
YOU: With all due respect, I disagree with your assessment that the authors’ conclusions are not an inference.
You offer no direct observation & measurement of physically detectable signal from the “star partners”.
===================================
So how is this talking past each other?
Let me explain.
You offered only "AGREE" and "DISAGREE" as possible answers to your question, and as it is worded the only possible choice (of three – I could have said "I don't know" or similar) is DISAGREE.
Then you ran with my one word answer and gave it a meaning quite different than what I intended, and quite different from the concise statement you asked me to respond to.
You see your concise statement contains the word "detectable", so IF you assume there are partner stars to the observed He WDs, THEN light (electromagnetic radiation) from them should be detectABLE.
Further, I can't see how your 'with all due respect' comment follows, logically, from my one word answer, especially as the chain that lead us here contains your "So apprently the “companions” emit no electromagnetic signal to indicate their presence"; this is surely a conclusion too far (Cool et al. certainly don't even hint at this extreme inference, and I think I've been clear that it is an invalid conclusion).
Do you see how communication has broken down?
April 30th, 2009 at 6:55 am
Anaconda, to help me understand how you parse detection from direct observation from inference, in astronomy, I've prepared a series of examples. If you could please indicate the extent to which there is detection, direct observation, indirect observation, and inference in each I'd appreciate it.
Of course, my list is very terse, and so you may not grasp what an item is about; if so, please ask for clarification.
-> DLA systems (objects detected by the absence of photons)
-> transiting exoplanets (ditto)
-> exoplanets discovered solely by doppler shift methods (detection vs inference)
-> objects discovered solely by microlensing (ditto)
-> 10 TeV gamma-ray emission by the Crab nebula (detection/observation/inference wrt IACTs)
-> energy and source direction of UHECRs (detection/observation/inference distinctions)
-> proper motion and parallax of Barnard's star as determined by HIPPARCOS (ditto)
-> CMB angular power spectrum as determined by WMAP (ditto)
-> galactic 408 MHz emission as synchrotron radiation (ditto).