Do you have adblock enabled?
If you can read this, either the style sheet didn't load or you have an older browser that doesn't support style sheets. Try clearing your browser cache and refreshing the page.

(BBC)   Upper limit placed on star growth at 150 times the size of sun. Red supergiants insist the stuff they took was legal and shouldn't affect their status   ( divider line
    More: Interesting  
•       •       •

5238 clicks; posted to Main » on 09 Mar 2005 at 4:26 PM (12 years ago)   |   Favorite    |   share:  Share on Twitter share via Email Share on Facebook   more»

56 Comments     (+0 »)

Oldest | « | 1 | 2 | » | Newest | Show all

2005-03-09 03:02:00 PM  
Headline made me giggle. Good job.
2005-03-09 03:16:38 PM  
Astronomy rocks!
2005-03-09 03:18:40 PM  
No Star Jones reference? Hmmm, I'm gonna go investigate.

Geology Rocks!
2005-03-09 03:20:49 PM  
nice headline
2005-03-09 03:21:23 PM  
... I could see Star Jones getting bigger after having a kid
2005-03-09 03:22:35 PM  
i tried to read the article, but the whole thinking thing wasnt working... then there was a picture and it all made sense.

Pictures Rock!
2005-03-09 03:28:03 PM  
[image from too old to be available]

Damn it. Too late.
2005-03-09 03:43:21 PM  
That magic number of "150" is not a very hard number, not as the articles make it sound to be. A quote from NASANews report:

"Although Figer did not find any stars larger than 130 solar masses, he conservatively set the upper limit at 150 solar masses."

And as Figer being a responsible, careful observer, he adds:

"Figer cautions the upper limit does not rule out the existence of stars larger than 150 solar masses."

Despite what scientists tell you, there aren't any perfect standard theories that can describe the formation of very massive stars in a self-consisted manner, yet. That's why Figer is cautious on his interpretation of the observations. The bottom line is that he did not find any stars more massive than 130times of the Sun's mass.

Whether Star Jones would be massive enough to collapse onto itself to turn into a degenerate state remains to be seen.
2005-03-09 04:29:58 PM  
I don't think Mark McGwire is THAT big of a star, yet.
2005-03-09 04:31:22 PM  
So... some scientist looked at a single star cluster, added ~15% to the size of the biggest star he could find and declared that no star could exceed that size?

Sounds scientific to me.
2005-03-09 04:31:36 PM  
Figer figgers 'hunred fitty final figger.

I just wanna f it.
2005-03-09 04:36:40 PM  
the ~15% is probably the stat+syst error on the measurement. so, he's just making a confidence-level-type estimate.

so, you're right - it is scientific.
2005-03-09 04:36:56 PM  
hurray for good headlines! oh ya the star thing is cool too.
2005-03-09 04:38:06 PM  
White Rose Duelist

So... some scientist looked at a single star cluster, added ~15% to the size of the biggest star he could find and declared that no star could exceed that size?

Sounds scientific to me.

That's why it's called a theory. Go find a star larger than 150 solar masses and prove him wrong.
2005-03-09 04:40:28 PM  
BTW 150 solar masses is still smaller than [image from too old to be available]

\ducking.... ;-)
2005-03-09 04:40:45 PM  
2005-03-09 04:42:53 PM  
YOU go find it...i'm not gonna. Dont care. Poopiehead.
2005-03-09 04:44:07 PM  
Damn Liberal Media.
2005-03-09 04:46:40 PM  
/searching for the upper limit on ass growth too.
2005-03-09 04:46:42 PM  
Science and sports mixed in one joke? And some people have enough knowledge of both to get it?

2005-03-09 04:46:59 PM  
Late, but I can't stop myself:

In other news, mathematicians baffled by Star Jones exceeding growth limit by factor of 10.
2005-03-09 04:47:18 PM  
noooo, sorry ironazis, I was thinking "sarcasm" but somehow it came out all wrong. Please do not send me away :(
2005-03-09 04:48:07 PM  
I do think this is an interesting theory, though I am not sure I agree with it. Even assuming that we expected to find stars of a mass of 500 or greater than our sun, that does not mean that we were correct in assuming that. And also, simple propability tells us that, most likely, in the millions of other gallexies out there, there must be larger stars; we only examined the one...just some simple thoughts.
2005-03-09 04:48:39 PM  
Oh, and in my defence, the terms are almost interchangable in my native language. The term "kaldhaedni" or "cold humour" conveys both a sense of irony and sarcasm.
2005-03-09 04:51:44 PM  
morrisonsl - what on earth is your native language? i tried to guess from that word and had no clue whatsoever....

Origins - star formation is not something we know nothing about...
there are good reasons to expect that there are limits (on both sides) due to the physics involved....
2005-03-09 04:53:17 PM  
2005-03-09 04:55:42 PM  

nature article (M circle dot = solar mass)

There is no accepted upper mass limit for stars. Such a basic quantity eludes both theory and observation, because of an imperfect understanding of the star-formation process and because of incompleteness in surveying the Galaxy. The Arches cluster is ideal for investigating such limits, being large enough to expect stars at least as massive as approx 500 solar masses (approx 500 M circle dot; based on a typical mass function), and young enough for its most massive members to still be visible. It is also old enough to be free of its natal molecular cloud, it is at a well-established distance, and it is close enough for us to discern its individual stars. Here I report an absence of stars with initial masses greater than 130 Mcircle dot in the Arches cluster, whereas the typical mass function predicts 18. I conclude that this indicates a firm limit of 150 Mcircle dot for stars; the probability that the observations are consistent with there being no upper limit is 10^-8.

Theory provides little guide in determining the most massive star that can form. Pulsational instabilities were once thought to destroy stars more massive than 95 M circle dot; however, these pulsations may be damped. Radiation pressure, and/or ionizing flux, inhibit accretion for stellar masses greater than 60 M circle dot, but direct collisions of protostellar clumps may overcome these effects. Although stellar evolution models have been computed for massive stars covering a large range in mass, up to 1,000 M circle dot, no such stars have ever been observed. Indeed, some of the most massive candidates have proved to be systems of multiple stars.

Stars generally form with a frequency that decreases with increasing mass for masses greater than approx1 Mcircle dot, that is, d(log N)/d(log m) = Gamma, where m is the initial stellar mass, N is the number of stars per logarithmic initial mass interval, and Gamma is observed to be -1.35. For stellar clusters young enough not to have lost members to supernovae, the distribution of stars is populated to the point where the mass function predicts one star, within the uncertainties of low number statistics. Therefore, stars with mass M > 150 Mcircle dot can only be observed in very massive clusters with total stellar mass >10^4 M circle dot. This requirement limits the potential sample of stellar clusters that can constrain the upper mass limit. Only a few clusters in the Galaxy satisfy this requirement, and all are located in the Galactic Centre.

To investigate the possibility that stars with M > 150 Mcircle dot exist, imaging data were obtained using the Near-Infrared Camera and Multi-Object Spectrometer instrument on the Hubble Space Telescope in a programme to measure the mass functions of the most massive young clusters in the Galaxy, near the Galactic Centre. Intervening dust prevents observations of these clusters at optical or ultraviolet wavelengths, so images were obtained in near-infrared wavelengths (see Supplementary Fig. 1). Nearby control fields were also imaged to estimate the number of field stars that contaminate observations in such a densely populated region.

Photometry for stellar sources in the images was extracted, and corrected for the absorbing effects of dust by comparing the observed colours to those expected for the appropriate spectral types; note that intrinsic colours of massive stars on the main sequence at infrared wavelengths differ by only a few per cent. The dereddened fluxes were then converted into bolometric fluxes by accounting for the distance to the Galactic Centre, and the Geneva stellar evolution models were used to infer initial masses for each star (see Fig. 1). Although these models have associated errors, note that the Arches stars are relatively unevolved; indeed, only the brightest dozen or so members show evidence of chemical enrichment by nucleosynthetic processes18. Some of the brightest stars in the cluster (three to ten, depending on cluster age within a range of 22.5 Myr and the coefficients in the extrapolation law) extend just above the 120 Mcircle dot limit of the massflux relation; I estimate masses for them that do not exceed 130 Mcircle dot through an extrapolation of this relation (see Supplementary Fig. 2).

Figure 1 Observed frequency distribution and inferred masses of stars in the Arches cluster versus brightness. Full legend

High resolution image and legend (65k)

The initial masses I estimate here agree with those inferred through wind/atmosphere modelling of high-resolution spectral observations to within a few per cent (ref. 3). Others have also applied the same technique to construct mass functions from infrared observations of massive young clusters, showing that these determinations are consistent with those estimated from optical observations19. In addition, several groups have found good consistency in physical properties inferred from optical and infrared analyses for massive stars at all stages of evolution20-22.

Figure 2 shows the resultant initial mass function of the Arches cluster, assuming an age of 2 Myr, for stars within a projected radius of 0.5 pc, and solar metallicity2, 18. Although the cluster is the densest in the Galaxy3, the data do not suffer from incompleteness due to crowding or sensitivity for the four highest mass bins in the figure. The small amount of background contamination was removed by subtracting the number of stars observed in nearby fields; this resulted in the subtraction of a total of seven stars from the upper four populated mass bins. The frequency distribution generally decreases with increasing mass, and is fitted by two lines through the four most massive populated bins, which contain 39 stars. One line has a slope of Gamma = -0.9, appropriate for the most recent determinations2, 23, and the other has a slope equal to the Salpeter value that is observed for most clusters. For both slopes, there appears to be a deficit of expected very massive stars with masses beyond approx130 Mcircle dot; variations in assumed age (plusminus 0.5 Myr), mass-loss rates and metallicity do not change the result. I estimate cumulative errors of approx10%, and conclude conservatively that there is an upper mass cut-off of approx150 Mcircle dot (see Supplementary Fig. 3 for the effects of mass loss on the most massive stars).

The observed deficit of stars is significant. If there is no upper mass cut-off, then the odds of identifying no stars beyond the observed limit are 10-8 if 18 are expected, and 10-14 if 33 are expected, assuming Poisson statistics. In addition, the maximum predicted stellar mass is at least approx5001,100 Mcircle dot, values that are far beyond the masses inferred from the observations. I performed a Monte Carlo simulation of model systems to predict (as a function of cut-off mass) the probability that a cluster with the mass of the Arches cluster could have no stars with initial masses greater than 130 Mcircle dot (see Supplementary Fig. 4). In this simulation, I added uncertainties due to differential extinction, photometric error, average cluster age, a spread of ages for individual stars, and error in estimating the average cluster age. The simulation predicts few systems with no stars having initial masses greater than 130 Mcircle dot for cut-offs of 150 Mcircle dot or greater (Supplementary Fig. 4).

Clearly, the cluster age (tau) is an important quantity for the analysis. If the cluster is too old, tau > 3 Myr, then its most massive members would no longer be visible (that is, they would have progressed to supernovae), and the observations would then simply reveal an apparent cut-off due to the natural effects of stellar evolution. If the cluster is too young, tau 1 Myr, then the models would predict much higher initial masses for the brightest members; however, note that even younger ages would still require a firm upper mass cut-off, albeit at somewhat higher masses than predicted by the best estimated age. Analyses indicate that the cluster has an age of 22.5 Myr (refs 2, 3, 18). A younger age is inconsistent with the nitrogen-enriched atmospheres revealed in the spectra of the most massive stars in the cluster18. The fairly narrow age range is required by the observed heavy nitrogen enrichment in the brightest stars together with relatively weak observed nitrogen content in the atmospheres of slightly lower mass stars3, 18. An older age is inconsistent with the evolutionary status of the most massive stars in the clusterthey have not evolved to advanced stages, such as the carbon Wolf-Rayet phase3, 6. In addition, the lack of any supernova remnants in the cluster argues for an age less than 3 Myr. Indeed, if massive stars filling the apparent deficit were formed and evolved to supernovae, one would expect that a supernova remnant would have been formed at least every 50,000 yr for the past 0.5 Myryet none are observed. In summary, stars with masses above approx150 Mcircle dot should still be visible if they were formed, given my estimate of the age for the Arches cluster.

The observed upper mass limit is on the low side of the estimated masses of a few massive stars in the Galaxy, although it still falls within the error bars of these estimates. It is important to note the large errors in such estimates. For instance, many of these estimates rely on stellar wind/atmosphere models that do not model the effects of increased opacity produced by metals in stellar winds (that is, line-blanketing). With more modern models, new mass estimates are smaller by up to a factor of two. In addition, mass estimates often suffer from uncertainties in distance, reddening and photometry. The typical build-up of errors can easily result in an uncertainty of a factor of two in flux, and of a similar factor in mass estimate. As an example, consider Pismis 24-1, which is estimated to have a mass of 210290 Mcircle dot (ref. 24). The build-up in errors for this star, from effects described above, produces at least a factor of two variation in flux estimates, and the original mass estimates were produced without the use of line-blanketing. Once these combined effects are included, the true mass of this star may well be below 100 Mcircle dot. Note that uncertainty in distance is the next obstacle to making accurate mass estimates once line-blanketing is included; however, the distance to the Galactic Centre is very well known (to within 6%), and the Arches cluster is physically connected to phenomena known to be produced in the Galactic Centre3.

If there are stellar systems more massive than the limit, then perhaps they are binaries, or products of mergers of lower-mass stars. Indeed, the Pistol star, with an inferred initial mass of approx150250 Mcircle dot (ref. 13), is surrounded by Wolf-Rayet and red supergiant stars that are older than the expected lifetime of such a star25. This apparent paradox may be reconciled if the star is actually multiple, or if it has recently experienced a rejuvenation through a merger with another star26. High-spatial-resolution imaging suggests that the Pistol star is not binary to within a limit of 110 AU (ref. 13), yet massive binaries can have components with orbits on yet-smaller scales14.

An upper mass cut-off of approx150 Mcircle dot was found for the cluster R136 in the low-metallicity environment of the nearby galaxy, the Large Magellanic Cloud27. This result relies on an apparent deficit of ten stars with masses beyond this limit, based on the assumption that R136 has a total stellar mass of 5 times 104 Mcircle dot; however, this high cluster mass includes stars that span a range of ages, including those that exceed the age at which a massive star is expected to evolve to become a supernova. This has the effect of increasing the base of lower-mass stars from which to extrapolate an expected number of higher-mass stars, thus inflating an apparent deficit if those stars are not seen. Using a lower estimate of the cluster mass, 2 times 104 Mcircle dot, in stars sufficiently young for the present analysis, I estimate that the true deficit beyond 150 Mcircle dot in R136 is roughly four starsthat is, the result in the present work is more statistically significant by this measure. If the deficit of massive stars in R136 is real, then it represents another measurement of the upper mass cut-off.

Surprisingly, the cut-off may be similar in environments that span a factor of three in metallicity, although metal content is often cited as a proxy for the source of opacity that limits the infall of material and eventual build-up of massive stars. This result implies that the process that limits the mass of a star is independent of metallicity, at least in the range of metallicities primarily found within the Galaxy and the nearby Large Magellanic Cloud.

Figure 1 Observed frequency distribution and inferred masses of stars in the Arches cluster versus brightness.
[image from too old to be available]

Figure 2 Frequency distribution versus mass for stars in the Arches cluster extracted from data in Fig. 1. Full legend

[image from too old to be available]
2005-03-09 04:55:53 PM  

I agree with you; our scientists know what they are doing, know a hell of a lot more about it than me, and have good reason to think this. Ahhh...but still...I have a hard time convincing myself that this particular assumption is correct, given the vastness of space...the universe as we know it could look, act, and be completely different in parts of the universe that we haven't even look at yet (flow of time sequence comes to mind, but that's another topic all together...). Surely there are stars that are much larger than we expect. Afterall, we are only human.
2005-03-09 04:58:38 PM  
BillCosby - major cool post - thanks!

_yes, i've already admitted i'm a geek.
__proud of it.
___but do NOT call me a nerd. them's fightin' words
2005-03-09 04:59:42 PM  
Makes sense to me. There is an upper limit to the mass of a planetary body. Past a certain point it's gravity will turn it into a star. If a star passes a certain mass (~150 solar masses?) shouldn't it's own gravity turn it into a black hole?
2005-03-09 05:00:57 PM  
Origins_Demise Surely there are stars that are much larger than we expect.

Yeah, but the act of being a star (being all bright and everything) will tend to blow away the outer layers of the star. Our sun loses 4 million tons of mass each second! So, if you had a star bigger than 150 solar masses, it wouldn't stay that way for very long.

This isn't a simple issue of not having found anything bigger. The issue is that nothing bigger is stable.
2005-03-09 05:05:09 PM  


Google seems to think so...
2005-03-09 05:05:43 PM  
God damn, the headline writers are ON FIRE today!
2005-03-09 05:22:38 PM  
does anyone have any rudimentary conception of probability theory? Do you have to poll every single person to see if they like Bush? You only have to look at a representative subset (such as the star cluster),and extrapolate probability. That's what error margins in surveys are for.

The guy basically says that 130 solar masses is the result, give or take a known margin of error. This allows him to claim, correctly, that the chances of a star being above 150 solar masses are vanishingly small (although of course not impossible).
2005-03-09 05:24:49 PM  
I wish our sun was a blue-white giant. That would look way cooler than what we have now :( Can someone do something about this?
2005-03-09 05:26:42 PM  
What, did we just conveniently forget about Chandra's Limit?
2005-03-09 05:32:45 PM  
[image from too old to be available]
2005-03-09 05:35:48 PM  
I think Kirstie Alley will disagree with you on that.
2005-03-09 05:36:24 PM  

I wish our sun was a blue-white giant. That would look way cooler than what we have now :( Can someone do something about this?

If the sun was a blue-white giant, we'd be dead.
2005-03-09 05:36:46 PM  
Those wacky scientist. What will they come up with next?

Everybody knows stars are little holes in the canopy of the sky.
2005-03-09 05:37:37 PM  
Why is our moon the only moon in the solar system without a cool name? Titan, Io, those are good names, let's name ours!
2005-03-09 05:39:32 PM  

I thought your supposed to entertain me with family values and kid talk? VERY dissapointing.
2005-03-09 05:39:50 PM  


The article specifically mentions black holes, although not by name.

2005-03-09 05:46:55 PM  

Why is our moon the only moon in the solar system without a cool name? Titan, Io, those are good names, let's name ours!

Do you not consider Luna to be a cool name?
2005-03-09 05:47:35 PM  
I'm pretty sure the Chandrasekhar's limit only dealt with the the mass of a white dwarf toward the end of a star's life and with whether or not the collapse of said dwarf would lead to the creation of a singularity based on that mass
2005-03-09 05:48:55 PM  
There is no limit on the mass of a singularity, however, to our knowledge
2005-03-09 05:51:02 PM

/too lazy to link
2005-03-09 05:58:29 PM  
That's fantastic. Have they figured out how to reverse entropy yet?
ZAZ [TotalFark]
2005-03-09 06:00:01 PM  
What he actually measured was the maximum brightness of a star in the cluster. I'm not convinced we know the mass-luminosity relationship well enough to support his conclusion at the stated confidence level. The author acknowledges this source of uncertainty: The observed upper mass limit is on the low side of the estimated masses of a few massive stars in the Galaxy, although it still falls within the error bars of these estimates. It is important to note the large errors in such estimates. For instance, many of these estimates rely on stellar wind/atmosphere models that do not model the effects of increased opacity produced by metals in stellar winds (that is, line-blanketing).
2005-03-09 06:04:44 PM

/too lazy to link

Chandrasekhar's Limit

Doesn't pop.
Chandrasekhar's Limit
Displayed 50 of 56 comments

Oldest | « | 1 | 2 | » | Newest | Show all

This thread is archived, and closed to new comments.

Continue Farking

On Twitter

Top Commented
Javascript is required to view headlines in widget.
  1. Links are submitted by members of the Fark community.

  2. When community members submit a link, they also write a custom headline for the story.

  3. Other Farkers comment on the links. This is the number of comments. Click here to read them.

  4. Click here to submit a link.