I read popular physics: Explosions at the edge

This is part of my series where I read physics articles in Scientific American, and rant about barely related tangents in order to provide “context”.

After the November issue, which didn’t really have any physics articles at all, the December issue has two major articles! One is astronomy, the other one is about the fusion reactor, ITER. But, after complaining about how all the physics articles are about astronomy, it looks like I’m still choosing the astronomy article. The ITER article is just a bunch of photos of the engineering, and I don’t have much to say about that.

So, the astronomy article is “Explosions at the Edge” (or that’s how it’s titled in print). It’s about the surprisingly diverse ways that massive stars can go supernova. For example, rather than simply exploding, a star may first shed a layer of gas, and then the subsequent explosion will collide with that gas, producing a prodigious burst of light.

As I hinted in the intro, I’m about to rant about something that is barely related to the article. There’s a saying, “the more you know, the more you know you don’t know.” Very true when it comes to astronomy, and stellar evolution in particular. Popular astronomy does a great job teaching the basics: how stars become red giants, white dwarfs, neutron stars, black holes, and so on. You might even know that red giants form because stars run out of hydrogen fuel for the fusion reactors at their cores. And if you’re not a physicist, you might be satisfied with this explanation.

But as a physicist who never studied astronomy, I understand that it is barely an explanation at all! It’s a summary of unexplained observations. One major question remains: why does running out of fuel lead to expansion? If I may point out, I’ve run out of fusion fuel, and yet that didn’t cause me to become a red giant. In frustration, I wonder, why do astronomers withhold the explanation from their popular writings? Is it because it’s not entirely understood, or because it’s only understood by a few astronomers, or because it’s too complicated for popular audiences, or because they know popular audiences are too incurious to ever ask?

Since this is a public blog, I felt obligated to take a moment to look up the answer. This article seems to be a good summary of the mechanism of red giant formation. Running out of fuel causes the core to collapse, which triggers fusion in a larger shell, which generates even more heat and causes the star to expand. I probably understood that explanation much better than the average person, but I still have so many questions about it.

The SciAm article isn’t even about red giants. It’s about stars in a more massive range, that go supernova instead. All I’m saying is I already knew stellar evolution is a lot more complicated than it’s usually made out to be. So if the thesis is that supernovae are more complicated than we thought, who is “we”?

An infographic within the article explains that under the classic model of supernovae, the kind of supernova depends on the mass of the star. There are six distinct categories, including a “Type-Ibc core-collapse supernova”, a “electron-capture supernova”, and a “Pair-instability supernova”. Now, that may be simple enough…

Wait you’re saying it doesn’t sound simple at all? Exactly, this is what I’m saying!

When this article says supernovae are more complicated than we thought, what it really means is that supernovae are more complicated than astronomers previously thought–basically, the article is about a handful of events that don’t fit into the classic model. But what astronomers previously thought was already way way more complicated than what lay people think. This article suffers from a problem that is common to many popular science articles: it’s dedicated to explaining how cutting edge research modifies the established scientific understanding, but it ignores the much larger distance between established scientific understanding and popular understanding.

And honestly? What can the article do? Explain all six types of supernovae, and exactly how and why they occur? To popular audiences? Not happening.

The angle this article takes instead, is to tell a more personal account of the author’s research experience. Yeah… that’s probably the better way to go.  But to me, not particularly exciting, because I already know what research is like.


So, this is the final issue of Scientific American I’ve subscribed to, and please, no gift subscriptions. Later, I’ll post a wrap-up post.  As for 2021, I might continue this series, if I receive enough reader requests.  I’ll talk about that more in the wrap-up.


  1. says

    Thanks for this, Siggy. You’ve convinced me that my understanding of physics is so limited that I could subscribe to Scientific American and be satisfied with the content 🙂

  2. says

    I think that “popular” accounts don’t have to dumb things down as much as they do. That’s not to say they won’t simplify (and over-simplify) in ways that make professionals in a field want to tear their hair out. I just think that we non-scientists can handle just a little more truth than we’re given.

    To prove this, I would have to show that non-scientists actually understand a little more science than scientists (& science journalists) think that we do.

    To that end here is my non-physics-informed, non-scientific, wild-ass guess of how main sequence stars become red giants:

    Gravity collapses gas into a main sequence star by heating & compressing hydrogen to the point of fusion. The “burning” of hydrogen into helium creates an outward radiative pressure that naturally balances against gravity: if too much hydrogen is burning, the star fluffs out more & there isn’t enough pressure to continue fusing hydrogen at the same rate, so radiative pressure decreases, so gravity collapses the star towards a denser state. The collapse, however, recreates the pressure necessary for a higher rate of fusion, which in turn increases the outward radiative pressure that holds back gravitational collapse. Because of momentum, you probably actually get pressure waves where the star “bounces”. In other words momentum pushes the core density higher than the equilibrium point, which pushes rates of fusion higher than the equilibrium point, which pushes radiative pressures higher than the fusion point, which “bounces” the mass of the star back out to a level of fluffiness that, again, cannot sustain sufficient fusion to prevent gravitational collapse. BUT this time it doesn’t fluff up quite as much as the first, so when the star collapses again, it doesn’t over-collapse quite as far as the first bounce. Eventually the lessening over collapse & lessening fluffy bursts more-or-less balance. There are probably always pressure waves, always small variations in fusion rates, but as a percentage of the overall fusion rates, these are small.

    But then comes the day when the star begins to run out of hydrogen. At that point, it cannot hold back gravitational collapse at the old equilibrium: there simply isn’t enough hydrogen fusion to produce enough energy to maintain the old out-ward radiative pressures.

    And so the star collapses. And one might think that this would lead to a black hole, but there are still atomic nuclei in the center of the sun. They’re just no longer hydrogen nuclei (by & large). But any nuclei lighter than Iron can release energy while undergoing fusion, so as long as those nuclei aren’t iron or heavier elements, the collapse can only proceed so far before those other nuclei reach temperatures & pressures that can sustain sufficient fusion to push back the gravitational collapse.

    Now, these larger atoms are less efficient at fusing than hydrogen, but that means in part that they fuse at a hotter temperature. Hotter temp = more outward radiative pressure. So the collapse presses inward farther than it ever has before while hydrogen fusion could still stop it, which means it has more momentum & can produce more overpressure than any harmonic mini-collapse that came before. So, of course, the resulting “bounce” is much bigger and the star fluffs up quite a lot.

    Now, where I’m stumped (assuming that any significant part of the above is accurate) is how the interior of the star sustains the high temperature & pressure balance necessary to continue fusing those heavier elements while the star has fluffed out so incredibly. (I would say “massively” but that would confuse the issue.)

    I’ve never quite thought about it this far, but if I had to guess, I would speculate that the new Red Giant star is more stratified than the earlier main-sequence star, Somewhere beneath the surface of the red giant, a vastly disproportionate amount of the stellar mass forms a star-within-the-star, balancing the gravitational pressure of this very dense, hot core (much denser & hotter than the previous core which could not collapse so far since hydrogen fuses more easily & at lower temperatures and so pushed back too hard to find equilibrium at this extra density) with the energy produced by higher rates of fusion of those heavier elements. (Rates being higher because equilibrium is not reached until you get those higher rates, since that fusion is less efficient)

    But the star, despite being fluffed out all to hell & gone (or earth-radius & gone if you prefer) never quite shed all the extra mass. That small, hot, fast-burning core can’t “fluff out” the heavier elements, but all that hydrogen & helium that was in the middle/outer regions of the main-sequence star & so wasn’t under sufficient heat/pressure to fuse, that stuff still wraps around this super-dense core like a down comforter.

    At a much higher temperature, the gravitational equilibrium point where the hydrogen/helium plasma captures just enough outwardly-radiating energy to prevent collapse is going to be achieved at a much larger distance from the core.

    So instead of a semi-homogenous main-sequence star with heat convection occasionally carrying new hydrogen & helium into the core & carrying out some hydrogen that has not yet had a chance to fuse, you get a star where the core interacts with the outer hydrogen/helium layer almost not at all. The light elements wrap & insulate the core, but otherwise they exist separately, with a much more distinct core/shell boundary than would existed earlier in the star’s history.

    And this leads to the crucial bit: the equilibrium point for outward radiative pressure vs. inward gravitation is entirely separate from the equilibrium point of the core. Before there would have been a more continuous density gradient from the surface of the star to the hydrogen-fusing core. Now, there is a sharp boundary, with the density for the core being very, very different from the density of the lighter elements, even the layer of lighter elements existing just barely outside that core.

    It seems like the star would collapse more over time as the more fusion-capable mass is used up and the more lighter elements are transformed into heavier ones. And you will go through different stages of the fusion lifecycle, with different pressure/temperature balance points. (I don’t know what those would be, but you do have to have some stage – or probably different stages – where, say, carbon, nitrogen & oxygen are fusion fuels and you also need stages where heavier elements like aluminum are produced.) But from the outside of the star you won’t see collapse, because the progressive collapse required to reach new equilibrium points during new stages of fusion affect only the core of the star where the fusion takes place. The hydrogen & helium wrapping the core in a big fluffy blanket haven’t been fusing during this time. They’re still light elements. But since the core is necessarily burning hotter now, this same mass of hydrogen & helium is absorbing more energy (heat) and so there’s the separate equilibrium point for the H/He blanket balancing the every-increasing heat of the core against gravity, which is not ever increasing since the star is not adding new mass. (In fact, it’s losing mass.)

    So the formation of a red giant is a combination of the initial “bounce” when the core first collapses enough for heavier elements to fuse, and then slow growth as the core keeps slowly collapsing while the outer stellar “fluff” keeps puffing out farther with the increasing heat.

    But wait! You say. If the core is increasing in heat, then shouldn’t the star turn blue, not red?

    Well, no. Because the photons bouncing around the interior of the star are not what we actually see in our telescopes (or eyes). Those photons are emitted and absorbed many, many times from their initial release during fusion to the moment they escape the star. We see the light given off by the surface plasma, which escapes because there’s no additional layer to absorb it again.

    Here it’s important to remember that gravity diminishes with distance. So although holding a ball of hydrogen & helium of the same mass out away from the core at a greater distance does require more energy production, the actual atoms in the plasma need less energy to remain in equilibrium at 100 million miles from the core than they did at 70 million miles from the core. So the internal temperature goes up, but the temperature of the outermost layers goes down – and continues to go down more as the radius of the star increases.

    In fact, if you knew the mass of the star and the radius of the star, I bet you could calculate its surface temperature very precisely.

    And this is Crip Dyke’s wild ass guess at how & why a main sequence star becomes a red giant, to the very limits of her understanding.

    The question now is this: is any of this shit actually true?

    Who knows? No True Scientist! IANAAorP! TAKE THIS IN FUN OR LEAVE IT ALONE, PEOPLE.

    Still, if our dear host Siggy wants to chime in, I’d be curious as to how much of this mess actually bears some semblance to reality.

    So, if you have time, what say you Siggy? Are we non scientists just sophisticated enough that we don’t need the astronomy and physics dumbed down quite as much as is traditional?

    Or am I merely proving that we non-scientists can’t handle the scientific truth?

  3. says

    @Crip Dyke,
    To be clear, I am not qualified to judge your narrative of red giant formation. But the “bounce” metaphor sounds very wrong to me. The process occurs slowly enough that it’s basically in quasi-equilibrium the whole time–“momentum” and pressure waves are not relevant in the long run.

    If I were to attempt an explanation, I would begin by establishing the physics of an ordinary star. Stars produce heat from fusion at the core, and eventually this energy is lost by radiation. Counterintuitively, the more heat that a star’s core produces, the colder it becomes, because that heat pushes against the force of gravity, and ultimately more energy is stored as gravitational potential energy than was supplied by the heat. The temperature of the core depends on distance from the center, and fusion only really triggers above a certain temperature, so a hotter core allows fusion to occur faster, within a larger region of the core; while a colder core restricts the region of fusion and slows it down. So more heat -> lower temperature -> less fusion -> less heat. It’s a negative feedback loop, which leads to a particular equilibrium.

    In a red giant, much of the hydrogen has been fused into helium. You still have the same equilibrium process, but something tilts that equilibrium in the direction of higher temperature, more fusion, and more heat (at least at the star’s core–to say nothing of the star’s surface). My major questions are:
    1. Am I mistaken in any of the basics?
    2. What is the mechanism that tilts the equilibrium of a red giant?
    3. Is this a smooth (but accelerating) process, or is it more like a phase transition?
    4. In my brief research, I saw claims that the whole process occurs in multiple cycles. If true, it sounds like the kind of thing that needs a whole dissertation to explain. What is the nature of the multiple cycles?

    Note that my questions are likely very different from the ones that curious lay people might ask. For example, lay people might be more confused by the “higher heat -> lower temperature” process, but I have a pretty good understanding of that. And I ask about phase transitions because I have a condensed matter background, but most lay people probably don’t even understand the distinction.

    I think this is all stuff that popular audiences could potentially understand, but there are two major constraints. The first is that it would only be accessible to a small dedicated audience, and maybe you’re better off targeting a larger audience with less explanation. The second constraint is that at some point you need to dive into the math, and audiences are not ready for that.

  4. Rob Grigjanis says

    Siggy @3:

    The process occurs slowly enough that it’s basically in quasi-equilibrium the whole time

    Except for the helium flash for a certain mass range, caused by the degenerate helium core.

  5. says

    @Rob Grigjanis,
    Thanks for pointing to that one. I’ve never heard of a helium flash. It’s obviously very complicated, and perhaps that’s why it doesn’t make it into popular accounts.

  6. Rob Grigjanis says

    Siggy @6: I hadn’t heard of it either. Your post got me looking at wikipedia, especially looking for things that were a bit out of the ordinary.

    The totally unsurprising Big Picture takeaway: even if you understand the physics underlying every detail of some phenomenon, it can be complicated enough that you couldn’t possibly put it all together without specialist training, and it can still surprise you. Which is cool, but also problematic when it comes to communicating science to the public.

Leave a Reply

Your email address will not be published. Required fields are marked *