Want a planet? You might want to avoid lithium

By Phil Plait | November 11, 2009 11:20 am

A science joke:

A woman is out walking and sees a kid on his hands and knees looking at the sidewalk. She asks the boy what he’s doing, and he says, "Looking for a quarter I lost." She asks him where he lost it, and he points across the street. Quizzically, she asks, "Then why are you looking here?" He replies, "The light’s better over here."

What’s this got to do with astronomy? I’m glad I asked.

protostellar_diskAstronomers took a sample of 500 stars, 70 of which are known to have planets orbiting them, while the rest have no planets detected. They examined the spectra of the stars, looking specifically to see how much lithium was present. What they found, with good statistical significance, is that stars with planets had far less lithium than stars that did not have planets.

Somehow, having planets means a star loses its lithium. How the heck does that happen?

A brief digression. Lithium is a weird element. It’s the third lightest after hydrogen and helium, and unlike every other element after it on the periodic table, we don’t think it’s made inside stars. It’s too fragile; the nuclei get smashed up easily, and so it doesn’t last long in the cores of stars. That means that as far as we can tell, all the lithium in the Universe was created in the Big Bang.

Just because it gets wrecked in the cores of stars does not mean they have no lithium at all. Lithium created in the Big Bang would have been in clouds where stars formed, and if a lithium nucleus can avoid the core of the star by staying nearer the surface, it can survive. The Sun has lithium in it, for example, but at far less abundance (<1%) than what you see out in gas clouds. That means the Sun has destroyed a lot – but not all – of its lithium supply.

When astronomers look at other stars like the Sun, the amount of lithium they possess varies wildly. But now it appears that the amount of lithium in a Sun-like star depends on whether it has planets or not. Stars without planets have, on average, 10 times the lithium as stars with planets in the sample.

Weird.

It's possible to think of simple ways that a planet could affect the lithium abundance of a star. Maybe the gravitational tugging of the planet helps mix up the star's interior, letting the lithium get close enough to the core to get destroyed. Shortly after the star and planets form, the planets can migrate slowly toward the star over long periods of time, which might affect how rapidly the star rotates. That in turn will affect how deeply the star's outer convection layer can penetrate the interior (bringing lithium down with it, destroying the element; in fact, this is one scenario proposed by the team that made this discovery.

Or maybe it’s something else. Or maybe there is a third thing we haven’t thought of yet, something that both destroys lithium and allows the star to make planets. The presence of planets and depletion of lithium might be related, but not directly.

It’s a mystery, but astronomers love mysteries. More observations will no doubt uncover more clues, give us more data we can analyze to uncover yet more correlations.

And that brings me back to my joke at the start. The press release for this news story makes an interesting statement:

This finding does not only shed light on the lack of lithium in our star, but also provides astronomers with a very efficient way of finding stars with planetary systems.

I disagree with the philosophy of this conclusion. Sure, if you want to find stars with planets, it might make sense to concentrate on stars with depleted lithium abundance. But I think that’s not a great idea: you’re only looking where the light’s good. When planets were first discovered around sun-like stars, we were all surprised to find them very close to their parent stars, orbiting in days, not years. The reason they’d been missed for so long is that no one had thought to look for them in orbits that small! We’d been looking where the light was good (literally) and not where the planets really were.

I’m not saying it’s wrong to only look to lithium-poor stars when seeking planets, but I am saying that if you’re a planet hunter, you might want to open your criteria a bit. This lithium finding is very interesting, and may well play out to be a hard-and-fast law, but I think it’s still a little early to rule anything out just yet.

As always, the Universe knows what it’s doing. It’s our task to figure out just what that is.

Image credit: ESO/L. Calçada

CATEGORIZED UNDER: Astronomy
MORE ABOUT: lithium, planets, stars

Comments (58)

  1. IVAN3MAN AT LARGE

    Phil Plait:

    What they found, with good statistical significance, is that stars with planets had far less lithium then stars that did not have planets.

    That should be than, not “then”.

  2. John

    Interesting, Are binary star systems also lithium poor?

    It will also be facinating to see the results from Kepler and see how these add to the situation. So far the vast majority of planets found have been Gas giants or at the very least significantly more massive than the Earth. Would a planetary system of only low mass planets have a similarly lithium poor star.

  3. Sarah

    Because there’s more light over here.. It’s an old Sufi one, told historically by Mullah Nasrudin. Who was not an astronomer, and this has nothing to do with Lithium .. just sayin. He had some good ones. See “delivering a Kutbah” for an idea on your next talk…

    http://en.wikipedia.org/wiki/Nasrudin

  4. JohnC

    I have yet to graduate with a degree for armchair astronomy, but could it have something to do with the stars producing heavier and heavier elements as they progress through generations? Or are heavier elements primarily produced from H/HE primarily?

  5. Aaron K

    I hope planet hunters act on this finding and concentrate their efforts and telescopes on Lithium-poor stars… that way econometricians like myself will have a new field to invade.

  6. NewEnglandBob

    Just make sure the scientist looking for planets are not taking lithium as a medication.

  7. does this mean stars with planets are more prone to be psychotic?

  8. DrFlimmer

    Very interesting, indeed! And I agree with your “conclusion” that lithium alone should not be the search criterion.
    But probably it is easier to find targets in the first place. Because if we see a star with low lithium abundance, then it is worth taking a second look at it with other methods.

    In fact, all methods should go side by side. We have just found another one – and it could be potent where the other methods fail.

  9. TheBlackCat

    How do we know it isn’t that the presence of lithium is somehow interfering with planet formation, or that having lithium in the outer cloud is required for planet formation so clouds that have it concentrated in the center (which later forms the star) cannot form planets, or that star systems that are generally rich in lithium have trouble making planets? It seems you are assuming that the presence of planets is the cause and less lithium in the star is the effect, but couldn’t it be the other way around, or that both are the effects of a third issue which is the underlying cause?

  10. My disagreement would be that there’s no use doing extra work on spectroscopy of stars just to find the lithium abundances THEN go planet-hunting. But, if you are just looking to rack up more planet detections quickly with limited resources, it looks like an already established population of lithium-depleted stars isn’t a bad place to try.

    Targeted surveys are helpful, but yes, eventually low-yield unbiased surveys will catch those interesting few you would have otherwise missed.

    Very interesting!!

  11. DeepField

    It is good to have in mind that “correlation does not mean causality”: the two circumstances (no lithium and planets) may occur simultaneously but not necessarily one causes the other. There may be something in those stars’ histories that may explain it better.

  12. A graph with means and standard deviations of the amount of lithium and non-lithium stars would be helpful to see how big the difference really is …

  13. ch

    Could there be a correlation to the type of gas cloud the star was formed from? A cloud derived from former lithium eating stars would be rich in heavy elements. A primeval cloud with the natural big bang lithium abundance would be rare in heavy element. Do the heavy elements play a pivotal role in planet formation?

  14. onshay

    “Somehow, having planets means a star loses its lithium. How the heck does that happen?”

    Why is this true? Taking the above posts about correlation not meaning causality into account, why can’t this info mean that planets tend to form around stars with little lithium present? Is lithium always present and that’s why the presence of planets means that star loses this element?

  15. Somehow, having planets means a star loses its lithium. How the heck does that happen?

    Heh, that sure got me scratching my head!

    as far as we can tell, all the lithium in the Universe was created in the Big Bang.

    Holy cow, I had no idea. That’s crazy. I mean, I assume it could be made in a particle accelerator (at least the nucleus), but not in any appreciable quantity.

    Anyway, how difficult is it to pull out a lack of Li from the spectrum of starlight? If this correlation holds, and the spectroscopy is doable, and the goal is to quickly build up a (non-comprehensive) catalog of planetary systems…. why the heck not?

  16. My (admittedly uneducated) guess while reading the post was that a dust cloud with a relatively high amount of lithium would have relatively low amounts of heavier, planet-forming, elements*. So when a star forms from that dust cloud, it will have high lithium content, but not much chance for planets.

    * my guess being that a dust cloud with heavier elements can only come from a supernova, which came from a hot star that likely destroyed most of its lithium during its lifetime

    But as I read further, I’m not seeing how the correlation necessarily implies that the lack of lithium is the cause of more planet formation any more than more planets causes a lack of lithium. Surely they are the result of the same cause.

    Of course, I’m just a speculating noob. This is all freshly out of my hindquarters :)

  17. J

    If Li is destroyed in the core of stars, how did it survive the nucleosynthesis process after the BB?

  18. amphiox

    #8: But lithium treats bipolar disorder, not psychosis. Of course having a bunch of planets, especially planets with humans on them, could drive anyone bipolar.

    “as far as we can tell, all the lithium in the Universe was created in the Big Bang.”

    Something to think about the next time you throw out that Lithion ion battery.

  19. Goheels

    The idea that a planet causes a star to lose its lithium is to me unsatisfactory. Admittedly I am not a physicist or an astronomer, and perhaps there is such an explanation, but the idea that they are correlated through some other factor seems better to me. The cliche “correlation does not imply causation” (I am a statistician, so I feel much more comfortable in this realm than the physics of planetary systems) seems to apply here. Could the fact that the sun is a second or third generation star be relevant here? It seems intuitive that second or third generation stars are less likely to contain lithium (or likely to contain less lithium), could it also be that second or third generation stars are more likely to have planets?

    A very interesting finding, regardless of the explanation, I look forward to the explanations offered.

    EDIT: Some other people made the same points I did, but their comments had not yet appeared when I made my post.

  20. Aleksandar Kuktin

    Great post!

    I would just like to comment on the issue of using the correlation in aiding planet-hunting. May it be noted, however, that I have not looked into the actual numbers which define this correlation (no experience interpreting astronomical papers :( ), so my thinking may be a little skewed.

    I believe the discovered correlation can be used, but that it could have a nasty consequence: no exo-Earths.

    Namely, most (if not almost all) of the planets thus far found included in this study (unless the authors corrected for this) are hot Jupiters, or something similar. So, we know there is a correlation between the star housing a hot Jupiter and it having low lithium. And, on the other hand – as far as I know, hot Jupiters are mutually exclusive with exo-Earths (or am I wrong?).

    And, since no exo-Earths have been found so far, plus that mutual exclusiveness.. leads me to conclude that it is the lithium RICH stars which probably house exo-Earths. ;)
    BTW, how does the Sun stand in that study?

    So, if we (meaning the astronomers) look only at the lithium-poor stars, we will probably never, ever find an exo-Earth. Simply because we will be looking in the wrong place.

    However, this can be run in reverse – by looking at lithium rich stars, astronomers may actually significantly up their chances of finding exo-Earths.

  21. Keith Harwood

    Strikes me there is a fairly obvious hypothesis. (OK, half-baked idea.) The stars with planets have low lithium because most of their lithium is locked up in the planets. A bit more baking suggests that if the star forming environment is conducive to planet formation the lithium binds chemically to the dust and the dust and gas preferentially separate, the gas going into the star, the dust into the planetary disk. (OK, quarter-baked.)

  22. ppnl

    I think it just reflects the different composition of the cloud that formed the star. A lithium free cloud has more heavy elements that can form planets.

  23. Chris

    I don’t think having a planet causes the star to lose it’s lithium, perhaps the gas cloud which condensed was already poor in lithium. If you want to form planets you need to have gas that has heavier elements which can make solids. That of course came from a previous star which went nova and used up much of it’s lithium. On the other hand if the star forms from virgin gas that hasn’t been used since the Big Bang, it’ll be higher in lithium and you won’t have any heavier elements to form planets.

  24. Jolly Random Info

    Alpha Centauri A has twice as much lithium than our sun does but then it is a more massive star.

    Alpha Centauri B has no observable lithium but then it is a K1v star, quite a bit smaller than our G2V sun.

  25. James

    We can in fact reduce this and point out that the paper is making an assertion that follows the Fallacy of Affirming the Consequent. They’re claiming that the presence of planets implies low lithium, (P -> Q) and that one should then find Q, and therefore P. This is 101 level stuff, right here.

  26. Garfield

    “The reason they’d been missed for so long is that no one had thought to look for them in orbits that small! We’d been looking where the light was good (literally) and not where the planets really were.”

    Really?

    Don’t most detection techniques, especially the ones used in the nineties, look for a planet’s effect on a star, and not for a planet itself? So “where” astronomers were looking really isn’t the issue — they were looking directly at a star.

    So far, most exoplanets are very massive and/or close to their primaries because current observing technology selects for those characteristics. Discovery trends indicate that as observational techniques improve and longer observation records can be compiled to identify regular variations on the orders of year, smaller and more distant planets will be identified.

    It would be interesting to know how lithium percentages vary throughout the galaxy. That could be another factor to add to the front of the Drake Equation!

  27. Chad Gardner

    I read the same “joke” in Anthony de Mello’s Song of the Bird. :o) I hope that is where you got it Phil…great book.

  28. Johno

    To everyone that is suggesting that heavy elements are required for planet formation: Jupiter is mostly (>90%) composed of hydrogen and helium and the same applies to the other 3 gas giants.

  29. Smoothie

    I think Keith Harwood is on the right track here. The causality might be due to chemical binding of the Lithium in planetary bodies and/or the differentiation of elements during planet formation via orbital/gravitational mechanics.

    So, the lithium is bound up in the planets, there’s less of it in the star. To me this seems the most likely to make that bastard Occam happy.

    All that said, there are not that many minerals on earth that contain lithium (http://www.mineralszone.com/minerals/lithium.html), so this needs considerable further mulling.

  30. MadScientist

    From a purely philosophical perspective it makes no sense whatsoever to take this attitude of “we’ll only look where we think it is favorable even though we don’t quite understand why an observed condition seems favorable nor can we even at this point claim an unequivocal favorability”. Such an attitude is plain *wrong* because it results in information being discarded without reasonable or legitimate apriori reasons. My favorite example of such cherry-picking of data remains Richard Feynman’s analysis of launch failures vs. launchpad temperatures. The technicians were discarding information which they did not believe was relevant when they had no good reason to believe that the information was irrelevant. The technicians concluded that temperature was not a good predictor of the occurrence of faults. Feynman’s conclusion was that temperature was an excellent predictor of faults and that in fact a temperature could be selected below which faults were virtually certain.

    In science it is extremely important to avoid controllable situations which may lead to bias and also to be perfectly aware of biases which you are introducing (bias is inevitable in many if not the majority of cases). The conclusion made in the paper is sloppy at best because it encourages a bias and for no legitimate reason. If your only objective was to rack up the number of planets you discovered in the short term, that bias would be acceptable. However, racking up your count of planets is not a valid scientific objective so the proposed bias to staring at stars with less lithium is worse than useless – it can give investigators an extremely skewed view if they should use the data for anything but racking up planet counts.

  31. Petrolonfire

    On the other hand if the star forms from virgin gas that hasn’t been used since the Big Bang, it’ll be higher in lithium and you won’t have any heavier elements to form planets.

    Virgin gas? Don’t tell me Richard Branson’s going into the gas business too! ;-)

  32. Spectroscope

    Also on the topic of which stars have planets it looks like the location of stars could be a major factor and that stars on the Galactic outskirts may have more trouble forming planets – see :

    http://kencroswell.com/MWEdgePlanets.html

    Except from there:

    Talk about location, location, location. If the Sun had been born near the edge of the Galaxy, chances are neither the Earth nor life would have arisen. That’s the implication of the first search for planet-forming disks on the Milky Way’s outskirts.

    The stars on the fringes of our Galaxy have little oxygen, silicon, or iron–chief ingredients of Earthlike planets–so astronomers have long doubted that life could exist there. Now they have solid evidence for their pessimism.

    Also, isn’t there more lithium in brown dwarf stars than regular ones because they don’t destroy it quickly? I think that’s actually used as a test to decide if a given faint star is a red dwarf or a brown one.

    So if brown dwarfs are *all* high in lithium is this useful when applied to them or, I’d think more likely, not? We know that at least some brown dwarfs have planets because, for instance, of the one that was imaged before Fomalhaut & HR8799 – 2M 1207 b.

    Finally is anyone else having trouble with the links in this post? The first link isn’t working at all for me & the second caused my computer to come up with a warning about security saying I needed a password and username to access it and another link took me to an “ESO is rebuilding its web” page.

  33. Spectroscope

    We know that at least some brown dwarfs have planets because, for instance, of the one that was imaged before Fomalhaut & HR8799 – 2M 1207 b.

    2M 1207b or to give its full designation 2 MASSW J1207334-393254 is a strong candidate for the “first ever exoplanet imaged” honour – although contending claims have been made.
    A brown dwarf sun about 25 Jovian masses, it has another object which is most likely a 5 Jovian-mass exoplanet (or perhaps another brown dwarf) orbiting it at a distance of 55 AU.
    This probable exoplanet was photographed by a European-American team using the Yepun telescope at the Chilean European Southern Observatory on April 27th 2004.

    For more info. on this see :

    http://en.wikipedia.org/wiki/2M1207b (its wikipage)

    http://www.space.com/scienceastronomy/050430_exoplanet_image.html (Candidates Debate – and contenders for first photo’d exoplanet honour)

    &

    http://www.eso.org/public/outreach/press-rel/pr-2005/pr-12-05-p2.html

    (ESO press release.)

  34. Dave Bissig

    You write, “[..] if a lithium nucleus can avoid the core of the star by staying nearer the surface, it can survive.”

    So how about this (feel free to replace the “all”s with “most”s, etc.):
    Suppose all stars studied started off with similar amounts (by % total mass, or whatever) of Li, and had a relatively even distribution of Li in the to-be-star + accretion disk. In each case, the star forms, destroying most of the Li there – most of the Li that would show up in the spectrum. So in every case, a large amount of the each systems’ Li is now in the accretion disk. Planets form, but sometimes in unstable orbits. If the orbits are unstable, the Li “stored” in the planets eventually reaches the surface of the star, increasing its Li, and showing up prominently in the spectrum. If the orbits are stable, this doesn’t happen, and the Li content of the star stays low. In the former case, we’d be less likely to detect a planet (there may be no planets left). In the latter case, we’d be more likely to detect a planet.

    … depending on the size of the effect (how much more Li we talking about? enough that could be supplied by a few jupiter-sized planets? or am I a few orders of magnitude off?) this is my best guess for now. Seems more plausible than a star *losing* Li because of planets, the tugging of planets mixing things up in the star, etc. … but then, I’m not an astronomer.

  35. I gather we’re not quite at the point of having detailed spectroscopic data for every star in the sky.

    I have a sentimental attachment to the star HIP 20740, aka HD 28113. In my fantasies, it’s the star I’ve chosen as my home. Why? G-class star, which is good for science fiction, and a perspective from which the relatively nearby Pleiades line up neatly against the Cygnus Rift, creating a rather pretty spectacle.

    Now I’m wondering how much lithium it’s got.

    Will I ever know?

  36. Robert

    There are really two classes of stars, those we know have planets around them, and those that might have planets around them but we haven’t found them yet. Just because we haven’t found them does not imply these stars have none.

    Now just as our planet finding techniques are predisposed to finding heavy planets in close orbits, might it be that, with current techniques, it might be easier to find planets in Lithium poor stars? Perhaps the Lithium in the atmosphere muddies up the spectrum just a tad so that finding those periodic Doppler shifts becomes a bit harder. In other words, the stars light is better for planet finding.

  37. Nigel Depledge

    @ Robert (36) – agreed.

    It seems to me that the planets that are likely to be the most interesting – smallish rocky ones with long orbital periods (months rather than a mere handful of days) – are also those that are most difficult to detect.

    So, the correlation could be the opposite of that claimed. I.e., we should be looking at lithium-rich stars to find Earth-like planets.

    However, this falls down completely if you also accept the hypothesis that the best chance for life (and therefore interesting planets) is on small planets in systems that also have large planets.

  38. I’m on lithium to control the voices in my head. This may also explain why I don’t have any planets orbiting me.

  39. ppnl

    Its true that the gas giants are 90% gas but that does not eliminate the possibility that they needed a solid core to seed the collection of that gas. That seems far more likely than the other ideas to explain the lack of lithium in planetary systems.

    Really, shouldn’t all second generation stars be short on lithium and rich in heavier elements? I just don’t see much of a mystery here.

  40. Bill Roberts

    Maybe astronomers like mysteries, but I don’t like mysteries. They give me a bellyache, and I’ve got a beauty right now.

  41. Yousuf Khan

    The lithium is used to create tritium, aka hydrogen-3, through a fission process. The tritium then reacts with deuterium, aka hydrogen-2, in a fusion process to make helium. This is known as a D-T (deuterium-tritium) nuclear fusion reaction. It’s one of several nuclear fusion processes that can possibly take place in the Sun.

    Fusion power – Wikipedia, the free encyclopedia
    “Consequently, the deuterium-tritium fuel cycle requires the breeding of tritium from lithium using one of the following reactions”
    http://en.wikipedia.org/wiki/Fusion_power#D-T_fuel_cycle

  42. Is there a similar correlation with beryllium and boron? They are also light, less common elements.

  43. gss_000

    The authors themselves have a theory on why lithium is not as prevalent in sun-like (and only sun-like) stars. The believe the planets or protoplanetary disks of gas somehow encourage he star to better mix lithium into the interior. By doing so, more of it is burned instead of remaining on the surface where the temperatures are not as hot.

    @31. MadScientist

    While in general I agree with your point, I disagree with your application here because the goal is not to rack up as many exoplanets as possible, but to find an Earth-like planet. If you want an Earth-like you need to make long observations of sun-like stars (see the Kepler telescope). In an ideal world, you could look at every sun-like star in the sky. Instead, if we now have a method (assuming it continues to be true as we find more systems) that allows us to better select our initial targets, then it really is great.

    I think we like to sometimes idealized science, forgetting that sometimes a theory can be true even if it’s not good.

  44. df1299

    @ 22. Keith Harwood & 30. Smoothie

    Even if all the planets in our solar system were made completely of Lithium, it would amount less than 0.15% of the Sun’s mass.

  45. beeDUB75

    Perhaps lithium is more specificly related to Stars with gas giant planets. Seeing as the vast majority (if not all) exoplanets found are gas giants – even the “super earths” discovered could just be Neptunes, then this could be some correlation. The holy grail of a confirmed terrestrial Earth mass planet finding could still be found going round any of the stars up there.

  46. Jess Tauber

    From what I’ve read beryllium and boron are also two elements most (all?) of which was/were created in the Big Bang, and not from stellar nucleosynthesis. None of the known fusion processes listed seem to make them. If so then that ruby in the ring, or that boric acid you’re using, have atoms dating back to the beginning of atoms. Put these together with your antidepressant and that would really be special. Diamonds forever? Hah!

  47. Flying sardines

    @ 45. df1299 Says:

    @ 22. Keith Harwood & 30. Smoothie

    Even if all the planets in our solar system were made completely of Lithium, it would amount less than 0.15% of the Sun’s mass.

    Now there’s a freaky thought! What would an all lithium Earth be like I wonder? ;-)

    Or, for that matter, an all lithium Jupiter, Pluto or Mercury? ;-)

    An all Boron world,OTOH, I imagine would be pretty Boron-ing in its lack of diversity. ;-)

  48. Nigel Depledge

    @ Flying Sardines (48) –

    But without boron, we couldn’t have sodium borohydride, which is anything but boring!

  49. Peter

    @ch (#14)

    Well, if it is indeed so that stars’ life cycle consumes lithium and produces heavier elements that does sound like an explanation that makes William of Ockam happy.

  50. Tarquin

    The caveat here should be restated again and again: it’s not definite. There seems to be correlation with low-lithium stars and high mass planetary companions. It cannot yet be taken as an indicator of planetary companions. Theories of third generation sun-like stars having planets have been put forward before, but the reason for the lack of lithium is not known. We need more data, and lots of it, especially on near Earth-mass exoplanets.

    As to the surprise of finding hot jupiters, that was a real surprise at first, but in hindsight it is obvious: they’re the ones easiest to detect with current methods, so no wonder such planets were among the first to be detected.

  51. Cool result, but if you look at the paper (http://www.eso.org/public/outreach/press-rel/pr-2009/Lithium_israelian.pdf), there are still a lot of low Li stars without planets (Figure 1). I’d like to see some plot of Li abundance vs. something more than just planet/no planet. Help me here, BA. Is there more associated with the discovery of extrasolar planets than just a planet/no planet determination?

  52. jkru

    So this comment is totally late, but I think saying that all Li was created in the Big Bang is false. It *is* possible to create lithium in a star through the Cameron-Fowler beryllium transport mechanism where the normal flow of the proton-proton chain is disrupted by convection. Beryllium is created in both PP-II and PP-III. In order for it to progress to the next stage in the chain, it either gets an electron or a proton. If it gets a proton, it turns into boron and if it gets an electron it turns into lithium. in the CFBT, the newly synthesized Be is convected to a cooler region of the star where it only has the energy to capture an electron and thus become lithium. Further convection brings it to the surface where it is seen in the spectra of some asymptotic giant branch stars.

  53. Graham Dungworth

    Keith Harwood and Smoothie have realised that it’s not just a physical process that depletes Li isotopes.
    The key point to note , is that meteorite Li isotope abundances in the solar sysyem are ca. 10^4 th fold that of cosmic abundance. The point is that Li doesn’t need to be destroyed but is fractionated instead . Li abundance is conserved throughout the stellar plus planetary system. Of course the long term fate of Li in the star may be destroyed by nucleosynthesis in hotter stars.

    Hence, in a protostellar cloud, prior to stellar ignition, a chemical process operates whereby the molten chondrules of silicates and aluminates etc (mm size) form like rain droplets in terrestrial clouds, chelating the alkali metal ions that include the Li isotopes. Infact, fractionation of 6Li is less than 7Li in carbonaceous chondrites that display a mobile water phase. On Earth although Li is the rarest element in the universe, as an alkali metal it is fractionated into salt deposits, a water phase; the rain and oceans have leached it from the from the parent planetary mass . The rarest element, at little cost powers our laptops and mobiles. So , although it’s rare Bolivia has enough intermontane salt, largely sodium chloride, but with enhanced Li, to last decades, hopefully with recycling.

    The first stars would lack the heavier elements, prior to later supernova debris which gradually enriches the H/He clouds. Clouds that lack refractories would display no lithium depletion. Clouds rich in supernova debris, from which the Solar System condensed, would fractionate Li into the dust that formed chondrules and then on to meteorite parent bodies and planets. The Earth is 36% Fe by mass; rocky planets are rich in heavy metals, silactes and aluminates. Lithium doesn’t promote planet formation. Planet formation ie. rocky planet generation requires supernoval debris and during the incandescense of planetary formation within the inner disk ALL metals chelate with undifferentiated materials ie. chondrites.

    How much rocky planetary material resides within Gas Giants? I don’t know. By comparison of Earth mass with the Jovian planet masses ca. 1000/1 and densities it’s likely that the total rocky mass of the solar system isn’t many fold greater than the terrestrial planetary masses. It’s an interesting speculation that major Li depletion in stars could reflect the presence of rocky terrestrial type planets rather than generic planets that include Gas Giant planets. The process of depletion is more likely to be chemical in the proto nebula rather than physical destruction by a purely physical and tortuous process of enhanced convection.

  54. Graham Dungworth

    Keith Harwood and Smoothie have realised that it’s not just a physical process that depletes Li isotopes.

    The key point to note , is that meteorite Li isotope abundances in the solar sysyem are ca. 10^4 th fold that of cosmic abundance. The point is that Li doesn’t need to be destroyed but is fractionated instead . Li abundance is conserved throughout the stellar plus planetary system. Of course the long term fate of Li in the star may be destroyed by nucleosynthesis in hotter stars.

    Hence, in a protostellar cloud, prior to stellar ignition, a chemical process operates whereby the molten chondrules of silicates and aluminates etc (mm size) form like rain droplets in terrestrial clouds, chelating the alkali metal ions that include the Li isotopes. Infact, fractionation of 6Li is less than 7Li in carbonaceous chondrites that display a mobile water phase. On Earth although Li is the rarest element in the universe, as an alkali metal it is fractionated into salt deposits, a water phase; the rain and oceans have leached it from the from the parent planetary mass . The rarest element, at little cost powers our laptops and mobiles. So , although it’s rare Bolivia has enough intermontane salt, largely sodium chloride, but with enhanced Li, to last decades, hopefully with recycling.

    The first stars would lack the heavier elements, prior to later supernova debris which gradually enriches the H/He clouds. Clouds that lack refractories would display no lithium depletion. Clouds rich in supernova debris, from which the Solar System condensed, would fractionate Li into the dust that formed chondrules and then on to meteorite parent bodies and planets. The Earth is 36% Fe by mass; rocky planets are rich in heavy metals, silactes and aluminates. Lithium doesn’t promote planet formation. Planet formation ie. rocky planet generation requires supernoval debris and during the incandescense of planetary formation within the inner disk ALL metals chelate with undifferentiated materials ie. chondrites.

    How much rocky planetary material resides within Gas Giants? I don’t know. By comparison of Earth mass with the Jovian planet masses ca. 1000/1 and densities it’s likely that the total rocky mass of the solar system isn’t many fold greater than the terrestrial planetary masses. It’s an interesting speculation that major Li depletion in stars could reflect the presence of rocky terrestrial type planets rather than generic planets that include Gas Giant planets. The process of depletion is more likely to be chemical in the proto nebula rather than physical destruction by a purely physical and tortuous process of enhanced convection.

  55. Craig

    “Somehow, having planets means a star loses its lithium. How the heck does that happen?”

    Isn’t it more accurate to say:

    “Somehow, having planets [of the limited characteristics we can currently detect] means a star loses its lithium. How the heck does that happen?”

    I don’t know if anybody is suggesting the lithium-rich systems don’t have planets. Rather it’s that the planets we’ve found have not been in lithium-rich systems. It’s important to remember that we’re only able to detect a fraction of the possible varieties of planets (we can’t detect planets like many of the ones in our solar system yet).

  56. Graham Dungworth

    It might happen by Li fractionation into chondrites ie. the meteorite or planetesimal pecursors to planetary systems.

    Garik fails to cite , for whatever reason it’s the same journal, the Chaussidon and Robert Nature paper 1999

    http://www.nature.com/nature/journal/v402/n6759/abs/402270a0.html

    At that time the existence of exo planets was conjectural. These authors presumed that the 140 fold enrichment of Li relative to a cosmic abundance of Li (unity or 1 part in 10^10 relative to H) in the Sun was due to a physical process of its destruction in the Sun. A chemical fractionation now appears highly relevant.

    A good test would be to look for Li abundance in stellar systems that are binaries; the closest being Epsilon Ori that has a twin brown dwarf system amongst others.

  57. Alex

    Just a couple of points to note in this:

    Firstly, the vast majority of objects (around 80%) of stars in the Solar neighbourhood are these low-mass M dwarfs. These are objects very frequently found in the so called lithium-chasm, where no/very little lithium is detected – primarily due the large convection zones in these stars. A larger convective zone reaches deeper down towards the core, and therefore can more efficiently deplete the lithium at the surface.
    Exoplanets isn’t my field – so maybe someone can help me on this. But surely there’s some statistical bias at play here. How many higher-mass stars have been detected? Higher mass stars should retain more of their initial lithium supply. Also a strong contaminant in all this lithium observation business is from binary stars, as they are rapidly rotating about their common centre of gravity, the rotate quickly, which can account somewhat for their relatively large lithium abundances.

    Personally, I’m not sure we should jump to the conclusion that planetary systems contain host stars with low amounts of lithium – as mentioned above; this appears to be a bit of the ‘correlation does not cause causality’ at work (as mentioned by DeepField(12)). Obviously more extensive observations are required on larger mass stars. Let’s hope future missions will provide further insights into such an exciting idea.

NEW ON DISCOVER
OPEN
CITIZEN SCIENCE
ADVERTISEMENT

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

ADVERTISEMENT

See More

ADVERTISEMENT
Collapse bottom bar
+

Login to your Account

X
E-mail address:
Password:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »