Astrophysics blog - Cosmology

Teoita

Italy12246 Posts

September 28 2016 16:17 GMT

I have decided i'll make my astrophysics blogs a more or less constant feature for a while, since they have been pretty well received and i have ideas for at least a couple more. This time i'll talk about a particular branch of astrophysics called Cosmology. Unlike last time it's not my field of work, but feel free to ask away regardless!

Cosmology is, by definition, the study of the entire Universe on the largest possible scale. A cosmologist doesn't care about the properties of individual objects, wether they are tiny neutron stars or immense galaxy clusters. Instead, a cosmologist tries to figure out how large and old the Universe is, much stuff it contains, and how this sutff can form the roughly 200 billion galaxies present in the Universe. At these scales, a galaxy cluster made of thousands of galaxies is nothing but a tiny little dot in a box.

Distances

So first off, how do we measure the distribution of objects, any object, in the sky? You can measure their angular distribution in the sky, but certainly not their distance if you just look at them. The only exception are of course the other planets in our solar system: observe them for a long enough time and you'll be able to reconstruct their orbit around the Sun. If you know how they orbit, you also know their distance from Earth's orbit. The problem is that we can only do this with planets, not stars or galaxies, so we need to come up with something else: stellar parallax. Imagine that a star is standing still right above the Earth's orbit around the Sun. As the Earth orbits, the projected position of the star varies; the closer the star, the larger the displacement. If an instrument is accurate enough to measure that displacement, one can derive the distance with basic trigonometry. The current best instruments for it, mounted on the Hubble and Gaia space telescopes, can measure a displacement of the order of a microsecond (one arcsecond is one sixtieth of an arcminute, which is one sixtieth of a degree), which corresponds to distances of about 10000 light years. This covers about one third of the distance to the galactic center...so again, if we want to measure how every galaxy is distributed in the Universe, we need to come up with something else.

That something else is called standard candles, and it's one of the most important concepts in astronomy. A standard candle is, very simply, an object whose intrinsic luminosity (not the the luminosity we observe at a certain distance, ie the flux) is known. If an object is emitting (in every direction) a luminosity L at a distance R from us, the flux F we observe is simply the luminosity, divided by the spherical surface over which it is emitted: F = L/4piR^2. Therefore, if i measure F and know L, calculating R is immediate. Unfortunately, very few objects are standard candles. An average star like the Sun isn't a standard candle, and neither is any galaxy, or most other stars...with one exception: cepheid variables. Cepheids are a particular type of old, very large but fairly cold star (also known as red giants) whose envelope pulsates radially, changing the luminosity over a very regular pattern. In 1908 Henrietta Swan Leavitt discovered that the pulsation period of a cepheid is strongly correlated with the star's intrinsic luminosity, which she measured for nearby cepheids thanks to their parallax. So, if we measure the pulsation period of a cepheid we know its intrinsic luminosity, and since we have been observing it we also have the flux, and therefore the distance.

Hubble's law and the Big Bang

Cepheids were the first and most important standard candle, because thanks to them in 1929 Edwin Hubble made the most important discovery in astrophysics of the 20th century. He looked at several cepheids in galaxies outside of our Milky Way, and compared their distance with the host galaxy's spectra. He found that the host spectra were consistently red-shifted - they appeared redder than they should have been because of the Doppler effect (exactly like a siren sounds more acute, "bluer", when it's approaching, and more grave, "redder", when it's moving away). This meant that every galaxy is moving away from us. More importantly, he knew the galaxy's exact distance thanks to their cepheid stars, and he found that far galaxies move away from us faster than nearby ones, following a very precise law: v = H * d, where H is a constant. If far away objects move faster, that means the Universe is expanding. Today we kind of take it for granted, but at the time it was utterly mind-blowing. Now, Hubble's original data wasn't particularly precise, but i swear, that's actually what's happening:

Hubble's law from his original 1929 paper; distance is on the x axis, velocity is on the y axis...let's face it, it's a pretty optimistic line

The discovery of the expansion of the Universe was debated for a very long time, and it was only widely accepted in the 1960s thanks to another mind-blowing discovery. As always in science, you don't want just a model that draws a pretty line on a graph (or not so pretty in Hubble's case, let's be honest here), you want to produce some prediction that can be tested. Take the expansion of the Universe: if it is expanding right now, it should have been smaller once, and smaller, and smaller as we go back in time, until it was a tiny dot. At some point (the Big Bang), that tiny dot started expanding. If we take the entire, well, everything, electrons, protons, neutrons, neutrinos and photons in the Universe, and put them in a much smaller space, it's definitely going to heat up and a lot...it's the difference between 100 people packing a small pub and in a stadium. At some point, there's going to be so little room to move around that even atoms can't form: as electrons try to attach themselves to protons, they immediately bounce off of any photon passing by. The photon loses some of its energy, and in doing so it pushes the electron away. Meanwhile, the photons themselves can't freely move around either, because every time they move a tiny space, they smash into another electron, and another, and another...at this point, the whole Universe isn't made of nicely ordered galaxies and stars, it's just a boiling plasma. As the Universe expands however the plasma becomes colder and colder, and eventually the photons do not have enough energy to keep atoms from forming. As soon as that happens, 380000 years after the Big Bang, they stop bouncing on electrons and are free to go as they please, which results in a flash of ultraviolet light being emitted at roughly the same time in the whole Universe. These photons where emitted a long distance from us, and as travel towards Earth they are redshifted, so they go from the ultraviolet, to optical, to infrared, to microwave frequencies where it is observed today. The Cosmic Microwave Background, or CMB, is the strongest testable proof we have of the Big Bang; no other physical model can reproduce its characteristics (to be precise, no other solution of the Einstein's field equations can, but i'll spare you the math).

Dark matter and dark energy

So far everything makes sense, somewhat. Now for the really, really wierd stuff. During the 1960s and 1970s several groups were studying how stars orbit in spiral galaxies, and found that the orbits of the ones lying away from the center are actually very, very wierd. As the star gets farther away from the center of the galaxy, where most of the galaxy's mass is, you'd expect its rotation speed to slow down. Instead what we observe is this:

Typical rotation curve of a spiral galaxy; the x axis is distance from the galaxy's core, the y axis is rotation speed

What's happening is that the rotation speed reaches its maximum, as is reasonable, but then becomes perfectly constant. Data to larger radii shows that this happens in every single spiral galaxy up to 10 times the radius shown in that graph. If we assume that the orbits of the stars is ruled by newtonian dynamics (which is a pretty sensible assumption), then that means there must be some amount of matter that doesn't emit any light, but which does interact gravitationally with the stars and forces their orbits to be faster than they should be. That something is called dark matter. There is similar evidence in the orbits of galaxies in galaxy clusters. We also know that without the gravitational pull of dark matter, galaxies as we know them couldn't possibly form. Despite seeing its effects on large scales, on tiny scales we aren't quite sure of what dark matter is. It can't be a normal particle predicted from quantum field theory, because all those particles do interact with light, but it could be some kind of particle beyond quantum mechanics, perhaps arising from its extensions like supersimmetry or string theory. Basically every one of these models predicts the existance of some kind of particle could well behave as dark matter on large scales, but none if them have ever been detected in a particle physics experiment.

So, now we know a) that the Universe is expanding, b) that gravity shapes its evolution, otherwise galaxies, nerds and cannon rushes wouldn't exist. So the question is, what now? What will happen to the Universe in the future? That depends on how much stuff (ie gravity) is in the Universe, vs how quickly it is expanding. On paper there are two possible outcomes: either gravity wins and the Universe crushes back on itself again (called the Big Crunch), or there isn't enough matter to stop the expansion, which continues on forever (the Big Chill). Unfortunately, cepheids aren't quite good enough to help us differentiate between the two, because we can see individual stars only in fairly nearby galaxies. We need a more badass standard candle: type 1a Supernovae.

Supernovae are violent stellar explosions; type 1a's in particular are caused by the explosion of a white dwarf. A white dwarf is a particular type of star that doesn't have any nuclear reactions happening inside of it, and therefore it collapses on itself until it becomes about as large as the Earth. In the 1930s, Subrahmanyan Chandrasekhar discovered that white dwarfs have a mass limit (1.42 times the mass of the Sun), beyond which they become unstable and explode. Because they all explode at roughly the same mass, it makes sense that they could be used as standard candles. In reality, they aren't quite all identical, but similarly to cepheids their properties can be used to infer the intrinsic luminosity: longer supernovae are also brighter. Anyway, because in general a star exploding is much, much brigther than a star on its own, they can be obseved up to about 10 billion light years away from us, as opposed to a about 10 million light years for cepheids. When one of these explosions goes off in a nearby galaxy we can look at its cepheid stars, calibrate the distance, and then extend our distance measurements to far away supernovae for which we can't see the host galaxy's cepheids.

In the late 1990s and early 2000s, several groups studying supernovae 1a updated Hubble's diagram to distances never reached before, and found something completely fucking ridicolous yet again: far away galaxies move too fast for Hubble's law. In other words, the Universe isn't only expanding, but the expansion is accelerating. It's as if something was pushing it from the inside. This something is called dark energy and it doesn't make any fucking amounts of sense, at all. There are some models that try to explain it but they are far, far from producing any kind of remotely testable prediction.

Data from Supernovae 1a is accurate enough to tell exactly how much matter and dark energy there are in the Universe. The result is that the Universe is made of about 70% dark energy (so completely YOLO stuff that makes no sense), 25% dark matter (fairly YOLO stuff that makes very little sense), and 5% of the barionic matter which we are all used to. This is called the standard cosmological model (or the lambda-cdm model), and it's widely accepted as a good description of the stuff that's constitutes the Universe, even though it obviously makes no sense. I heard stories that the conference of these groups started with "it's with great regret that i announce that we lost about 95% of the Universe", which is pretty accurate to be honest. Yay for science! But wait...there's MORE shit that makes no sense.

Some of the best data we have on the expansion of the Universe using Supernovae as well as other standard candles. The purple line is the prediction of the lambda-cdm model.

Inflation

The standard cosmological model predicts among other things that the universe MUST have had to go through a period of extremely accelerated expansion in its early stages. The reason for this is the existance of three very basic problems in lambda-cdm models that do not include inflation. These are called the flatness problem, the horizon event problem, and the magnetic monopole problem. In order:

1) Flatness: we know from empirical observation that the structure of space-time is flat. Ideally, general relativity allows it to be any kind of 4-dimensional surface with positive, negative or null curvature (just to picture things, in 3 dimensions a sphere has positive curvature while a plane is flat). Not only is the universe flat, but it can be shown easily that the further back in time you go, the more flat the universe gets. This is seriously bizarre; why the hell would the universe be absolutely perfectly flat at its start, when it can assume any possible value for its curvature while still following the same exact physical rules? It's an amazing coincidence, and it's extremely unlikely to just so happen by chance.

2) Horizon distance: this is even wierder. Basically, most photons existing in the universe are part of what is called CMB, or cosmic microwave background. These photons were emitted together billions of years ago, when the universe was still extremely hot. In fact, it was so hot that photons kept interacting with matter, keeping atoms from forming, bouncing from one nucleus to the other instead of being free to go their own way. As soon as the universe cooled enough, the photons suddenly stopped interacting with matter so strongly, and were free to go their own way. This radiation permeates the entire universe. By studying the CMB you can easily see that it was emitted at the same exact temperature at every point in the sky. On paper this makes sense, but if you look at things more carefully, you realize that light (and therefore, information) at the time of the CMB emission did not have time to travel through the entire sky as we see it. In fact, you can divide the sky in roughly 20000 patches of equal size; each of those would be able to transmit information within itself, but not to its neighbours. If these patches can not transmit any information to each other (ie, they can not reach the same temperature), why the fuck are they at that temperature?

3) Magnetic monopoles: this comes from more complex quantum mechanics stuff so i won't go in detail about it. Essentially, some quantum mechanics models predict that magnetic monopoles should exist, and their presence should be easily detectable. So why do we not see them? Is it possible that the universe evolved in a way that made them disappear?

Inflation theories solve all these problems. Inflation essentially states that there was a period in the early history of the universe, during which the expansion of the universe was INSANELY fast. This solves the 3 problems of traditional lambda cdm models:
1) Even if you start with a very curved universe, if you stretch it immensly it ends up being very very flat.
2) If the universe expands extremely fast for a while, then before it expanded it was extremely tiny, to the point where it could exchange information with every part of it, reaching the same temperature easily.
3) Slightly more complex, but ideally even if magnetic monopoles exist, the expansions "stretches" them so much that they become insanely rare, almost non existant.

There's another problem: well if everything becomes perfectly uniform, how come the universe we see isn't uniform? Why do we see stars, globular clusters, galaxies, galaxy clusters and superclusters? So far, you'd think that inflation would make everything perfectly homogeneous. What does happen is that when you first start with your tiny, small universe before inflation, it isn't perfectly homogeneous because of quantum mechanics. At a very very tiny scale the universe isn't empty: particles are constantly destroyed and created, according to Heisenberg's uncertainty principle: E*t>h, where E is the energy, t is the time, h is planck's constant. As long as a pair of particles of energy E exist for a time shorter than t, in order to respect the uncertainty principle, they are free to be born out of nothing, and return to nothing shortly after. This phenomenon is called vacuum energy. The way it ties into inflation is simply that these fluctuations are blown to huge sizes when inflation starts, and are what forms the large scales structures like galaxy superclusters that we see today.

I hope that made sense. Even though it really doesn't, let's be honest.

My previous blogs:
1) A basic introduction to astrophysics: http://www.teamliquid.net/blogs/483814-a-basic-introduction-to-astrophysics
2) Black hole things: http://www.teamliquid.net/blogs/511004-astrophysics-blog-black-hole-things

+ Show Spoiler [Bonus: the entropy of the Universe] +

Sometimes, my thesis supervisor comes up with these random paradoxes (which he calls existential doubts) that he'll ask anyone that's near him. He'll literally walk in our office in the middle of the afternoon, start talking about whatever is on his mind, and keep going for hours because it's actually fucking fun. The entropy of the Universe is one of my favorites.
Entropy is a measure of how disordered a given system is. At the start of the Universe, it must have been low, while now it should be higher, because according to the second principle of thermodynamics the entropy of a can not possibly decrease without expending some energy from outside the system. In the case of the entire Universe, there is nothing outside of it that can affect it, so it must increase over time, period.
So why does the Universe now appear more ordered now than it used to be? Neatly dividing matter in superclusters, clusters, galaxies, star clusters, stars and planets is certainly more ordered than a single plasma where atoms can't even form; it's the equivalent of having coffee and sugar outside of a cup, vs melting the sugar in the coffee. Well, that's has to be because of the expansion of the Universe. As the expansion goes on, there are more and more possible configurations for all matter, and therefore the entropy increases even though it appears to be decreasing. So far it makes sense.
However, here's the problem. If there is enough matter for the Universe to collapse in a Big Crunch, the equations of general relativity tell us that the collapse must be exactly symmetrical to the initial expansion...but if during expansion entropy was increasing, this time it must be decreasing, thus violating the second principle of thermodynamics which is kind of a big deal. So, does this mean that the Universe is forced to continue its expansion forever, as indeed appears to be the case? Does dark energy come from thermodynamics? My supervisors answer was literally "and with this, i mean that thermodynamics is fucking annoying".

lichter

1001 YEARS KESPAJAIL22272 Posts

September 28 2016 16:41 GMT

so i heard there are 13 zodiacs how do you feel about this

SonuvBob

Aiur21549 Posts

September 28 2016 18:50 GMT

As much as I want to make a dumb trolly post.. this is really well written and thorough, and serves as a great review of a cosmology course I took recently. These are some pretty tough concepts, and I think you explained them well without dumbing it down much.

My question: Can you explain how redshifting of photons doesn't violate conservation of energy?

(Also, Hubble's data was off because he didn't know there were multiple types of Cepheids, right?)

Thaniri

1264 Posts

September 28 2016 19:25 GMT

I have nothing to contribute but I wanted to post that I actually read it all.

Why was it strange to observe that far away galaxies can exceed the speed of hubbles law? It seems like the law simply wasn't extrapolated enough to deal with (even more) gargantuan distances. If you observed even a limited number of galaxies at varied enough distances you'd be able to determine that they are accelerating.

Teoita

Italy12246 Posts

September 28 2016 20:02 GMT

On September 29 2016 03:50 SonuvBob wrote:
As much as I want to make a dumb trolly post.. this is really well written and thorough, and serves as a great review of a cosmology course I took recently. These are some pretty tough concepts, and I think you explained them well without dumbing it down much.

My question: Can you explain how redshifting of photons doesn't violate conservation of energy?

(Also, Hubble's data was off because he didn't know there were multiple types of Cepheids, right?)

Thanks a lot

if you need any more indepth explanations for your course feel free to send me a PM!

Yes, he there is another kind of variable red giant called RR Lyrae which at the time wasn't known, so he assumed they were regular cepheids. Their variability also correlates with luminosity but the relationship is different, so some of his distance estimates are wrong. RR Lyrae are also standard candles, but they are indipendent of cepheids.

The short answer is that energy and its conservation in general relativity are tricky to define and not nearly as simple as in classical mechanics. Energy conservation only holds defined in a locally inertial reference frame, however in an expanding Universe no such frame can be defined (at least, not covering the entire Universe) and therefore redshifting and blueshifting don't violate energy conservation.

On September 29 2016 04:25 Thaniri wrote:
I have nothing to contribute but I wanted to post that I actually read it all.

Why was it strange to observe that far away galaxies can exceed the speed of hubbles law? It seems like the law simply wasn't extrapolated enough to deal with (even more) gargantuan distances. If you observed even a limited number of galaxies at varied enough distances you'd be able to determine that they are accelerating.

If something is accelerating, it needs to something else pushing it. If it's the entire Universe's expansion that is being pushed, that something that's pushing is effectively behaving like something with "negative" pressure: the more you expand, the more it wants to expand. It's the complete opposite of any other gas, liquid or material that we know of.

It's not that the law wasn't extrapolated with even longer distances (it always was, type 1a supernovae are only observed up to redshifts of about 1, while the farthest objects ever observed up to the 90s i believe went to about 6), it's that at longer distances it doesn't hold at all because there is more in the Universe than just matter and dark matter, which is really really strange because, again, the properties of dark energy are incredibly wierd.

SonuvBob

Aiur21549 Posts

September 28 2016 21:59 GMT

The short answer is that energy and its conservation in general relativity are tricky to define and not nearly as simple as in classical mechanics. Energy conservation only holds defined in a locally inertial reference frame, however in an expanding Universe no such frame can be defined (at least, not covering the entire Universe) and therefore redshifting and blueshifting don't violate energy conservation.

Not gonna lie, that sounds like a total copout. :p

On a more serious note, any advice for trying to wrap your head around GR? I wasn't a physics major, but I try to learn in my spare time, and that's one of the things on my bucket list. Or is it just like QM, where you just do the calculations and see what pops out?

dravernor

Netherlands6175 Posts

September 29 2016 06:09 GMT

I think this might be my favourite of your blogs so far because this is all stuff I'm currently studying.
Q&A to follow

Thanks Teo

zatic

Zurich15306 Posts

September 29 2016 06:44 GMT

That was fantastic, thanks! So much better to read a structured post like this than clicking through the Cosmology section on Wikipedia and getting lost in articles about freakishly fast spinning things and 32 Richter scale quakes

Teoita

Italy12246 Posts

September 29 2016 07:38 GMT

Thanks guys ^_^

On September 29 2016 06:59 SonuvBob wrote:

Show nested quote +

Im pretty biased against QM because the math behind it is just fucking ugly imo, so i think GR is much simpler and prettier, at least if you have studied classical mechanics and special relativity. If you have done that, GR is actually pretty similar to stuff you're familiar with (for example, the field equation is more or less an "updated version" of the Poisson equation). The results that come out of the math can be really really wierd because differential geometry is hard, but at least to me it feels like what you're doing to get there makes sense, as opposed to QM which is just wierd. Feynman himself said that if anyone claimed he understood QM, that meant he didn't understand it at all.

GornWood

Germany121 Posts

September 29 2016 08:59 GMT

#10

Big Bang, one thing lead to another, gravity created cannon-rushers.

I enjoyed reading your post even though I don't really have a clue about any of this, but it was fun to at least hear all this weird stuff, which may or may not make any sense.
10/10 (would appreciate more weird stuff in the future)

Uldridge

Belgium4455 Posts

September 29 2016 21:16 GMT

#11

How much of what absolute entropy does, is similar to the mechanics of diffusion?
If every piece of energy was uniformly distributed in space, what would happen? Can you draw similarities with the concept of the perfect crystal?
What is space and how is it able to expand? Everyone always talks about matter, but I find space just as interesting. It's the medium that is able to hold it all. I don't know if this is a weird/dumb question or not tbh..

Teoita

Italy12246 Posts

September 29 2016 22:31 GMT

#12

On September 29 2016 17:59 GornWood wrote:
Big Bang, one thing lead to another, gravity created cannon-rushers.

I enjoyed reading your post even though I don't really have a clue about any of this, but it was fun to at least hear all this weird stuff, which may or may not make any sense.
10/10 (would appreciate more weird stuff in the future)

Thanks! As i said, i already have a few ideas (although they probably aren't as wierd as inflation and whatnot)

On September 30 2016 06:16 Uldridge wrote:
How much of what absolute entropy does, is similar to the mechanics of diffusion?

Not much at all. Entropy is a measure of something (in this case, how ordered or disorded a system is). Diffusion is a process (for example, a photon diffusing through plasma, or sugar melting in coffee).

If every piece of energy was uniformly distributed in space, what would happen? Can you draw similarities with the concept of the perfect crystal?

For large scale studies (such as measuring the expansion rate of the Universe), we actually do assume that matter (and therefore energy) is distributed uniformly. On large scale nothing changes at all. If the distribution of matter was truly, perfectly uniform, with no fluctuations or perturbations at all, then ordered structures like galaxies, clusters or stars wouldn't be able to form, because you need some form of over-density above a mean value to initiate a gravitational collapse. I don't know what you mean with perfect crystal, sorry.

What is space and how is it able to expand? Everyone always talks about matter, but I find space just as interesting. It's the medium that is able to hold it all. I don't know if this is a weird/dumb question or not tbh..

Space is just 3 of the 4 coordinates in General (and Special) Relativity. What space time actually is, is more of a philosophical question than a scientific one, if that makes any sense.
We don't know what first triggered the expansion of the Universe, that's a much deeper question than what science can answer. What we do know is that if it does expand, it must do so in a very precise way governed by General Relativity and by how much stuff is in it.

Uldridge

Belgium4455 Posts

September 30 2016 00:31 GMT

#13

I don't really exactly know how to describe it. But aren't they closely tied together in a sense? If entropy is never decreased and diffusion makes homogenisation happen in a medium, isn't that kind of the same thing?
So if you have always increased entropy in space (matter will start to decay, higher energy will become lower energy through the pull space has on it (I see this as an analogy how liquid water becomes gaseous under lowered pressure)), doesn't it reach a point where all the energy at the lowest it can go (in the order of Planck's constant, idk, let's just use it for a cutoff point), where you can say everything is perfectly homogenous everywhere, not just on how we view things now, but literally everywhere.

And then there comes my point of the perfect crystal, from the third law of thermodynamics, which is classicly described as this:

The entropy of a perfect crystal at absolute zero is exactly equal to zero.

. When all the energy has been completely homogenized, you have maximal energy, but the energy is so low you basically don't have a temperature anymore and so the entropy becomes zero as well.
Then I take some leaps, but for some reason I see this space, with all the energy in it to be perfectly symmetrical at which point the space cannot support itself anymore and collapses back onto itself.
So let's say I'm kind of a big bounce guy. Ofcourse you'll probably (and I encourage you if you see big, huge holes in my story) destroy my weird ass destiny of the universe theory, but I kind of like this romantic thought.

Sorry for my ramblings by the way

PS: why can't the dark energy/matter be the seemingly missing antimatter that's should have been predicted by current models we have now? Couldn't antimatter behave like our matter (cluster together, form crystals and compounds etc), but with other antimaterial properties (e.g. exotic properties we haven't found out about yet)?

Maenander

Germany4919 Posts

September 30 2016 02:07 GMT

#14

On September 29 2016 16:38 Teoita wrote:
Thanks guys ^_^

Show nested quote +

I get that you don't like QM for philosophical reasons, but the math behind it isn't ugly to me! You can get some fascinating results with relatively simple calculations (e.g. hydrogen atom, Bell's theorem, spin-1/2 particles with the Pauli matrices) and I find the formalism to be at least as beautiful as the one for general relativity.

Also, computations in general relativity can get ugly very fast (even Einstein got confused, just look at one of his later papers in which he tried tell the world gravitational waves didn't exist

the story is here

Teoita

Italy12246 Posts

September 30 2016 08:09 GMT

#15

On September 30 2016 09:31 Uldridge wrote:
I don't really exactly know how to describe it. But aren't they closely tied together in a sense? If entropy is never decreased and diffusion makes homogenisation happen in a medium, isn't that kind of the same thing?

Diffusion is a process which increases entropy, but it's not the only one at all. It's just a convenient way of explaining what entropy is. A black hole evaporating because of Hawking's radiation also increases its entropy, but it's not exactly as intuitive as "like when sugar melts in coffee".

So if you have always increased entropy in space (matter will start to decay, higher energy will become lower energy through the pull space has on it (I see this as an analogy how liquid water becomes gaseous under lowered pressure)), doesn't it reach a point where all the energy at the lowest it can go (in the order of Planck's constant, idk, let's just use it for a cutoff point), where you can say everything is perfectly homogenous everywhere, not just on how we view things now, but literally everywhere.

I don't understand what you mean by this, why would matter start to decay during expansion? What happens is that the matter energy density decreases (ie, energy per unit of volume), because the volume of the Universe is increasing while the amount of matter remains constant, not because something is losing its energy. That also only holds true for matter; the energy density of dark energy is constant (which is why it's also called the cosmological constant).

That said, you are right that if the Universe keeps on expanding forever (as we currently believe), eventually it will be pretty much uniform because everything, even the quarks in a baryons, will be infinitely "stretched out".

And then there comes my point of the perfect crystal, from the third law of thermodynamics, which is classicly described as this:

Show nested quote +

This is pretty much correct for a Universe that keeps on expanding: eventually it's average "temperature" would tend asymptotically to absolute zero. In a Universe that collapses back on to itself (like my supervisor is thinking about when he wonders about the overall entropy of the Universe) that can't happen however, because the collapse needs to be symmetrical to the expansion, at least on large scales.

That said, all of our measurements of what's happening in the Universe rule out any Big Bounce or Big Crunch model very, very strongly. At this point the Big Chill is the only reasonable outcome, unless we figure out that General Relativity is completely wrong (which isn't going to happen any time soon). This is actually only from about 15 years ago, when the first Supernova 1a surveys were completed; before that all 3 destinies were possible. In fact, some papers before the 2000's have slightly fucked up distance computations because the assumed cosmological model didn't include any dark energy in it.

Dark energy and dark matter are completely different entities that have nothing in common except the fact that we dont know wtf they are. Dark energy doesn't look remotely like anything we know, because as i wrote earlier it pushes the expansion rather than hindering it through gravity (which is what antimatter would do). Antimatter does interact with the electromagnetic force (when a particle and its antiparticle collide, they produce photons), and so it can't be dark matter either (which again, by definition does not interact in any way shape or form with radiation). More importantly, if galaxies had all this antimatter in them then it would collide with whatever matter is in the galaxy, annihilate both and produce a substantial amount of light, at the expense of most of the matter in the galaxy. If a galaxy was made of both matter and antimatter, it would pretty much just disappear in a flash of light (which of course isn't the case at all).

Saying that models predict an excess of antimatter over what we have now is slightly incorrect. It is true that if antimatter and matter were perfectly symmetrical in all their properties, then our Universe as we know it wouldn't exist. In a Universe that contains the exact same amounts of matter and antimatter, all that can happen is particles annihilating each other, producing photons, which then decay to produce one particle and one antiparticle, and so on. In order for the Universe to evolve as it has, there needs to be some amount of asymmetry between the two. If there is slightly more matter than antimatter (as in our Universe), then most of the particles will annihilate each other and produce a lot of photons, but a small amount of matter remains and can go on to form ordered structures like galaxies. This is exactly what we observe actually; for every particle in the Universe iirc there are about one billion photons, mostly belonging to the CMB.

On September 30 2016 11:07 Maenander wrote:

Show nested quote +

the story is here

I should be more clear i guess: it's not quantum mechanics i have a problem with (well, maybe a little bit hehe), but quantum field theory and in particular the renormalization group. Dividing by infinity and pretending it didn't happen just feels fucking dirty to me. In general, anything that goes "YOLO scalar /vectorial/tensorial fields i can do whatever i want weeee" just annoys me. Even though im slightly more on the theoretical rather than observational side myself, i want shit to have testable predictions and useful constraints, which some of these models don't produce at all because they just have so many degrees of freedom. I dislike modified gravity models that try to explain cosmology without dark energy and dark matter for the same reasons.

Einstein, like everyone else, made a ton of mistakes in his career actually. He didn't believe in the probabilistic nature of QM and tried his hardest to disprove it for example, on the basis that he personally disliked abandoning the classical physics concept of predictive models over probabilistic ones.

Cascade

Australia5405 Posts

September 30 2016 09:16 GMT

#16

Quite the blog, good job!

But the hate for QFT?? :o
You say that you like models that predict data, and then you say that you don't like QFT?

How about lamb-shift loop corrections giving the hydrogen energy levels to some 8 digits accuracy? How about predicting a third class of particle outside matter particles and force-carrying particles? How about essentially everything in any collider experiment last 60 years? I think QFT has a more solid empirical footing than cosmology by some margin.

And I'd argue that the sum over histories weighted by the Lagrangian is a damn beautiful formulation of QFT. Linear algebra at it's finest.

It's true that the maths gets really messy when you start doing perturbation calculations, which you essentially almost have to do. But what comes out fits data extremely well.

And wasn't tensors popularised by GR? They are just really convenient notation for 4D calculations... Why do you hate them? :o

I mean... GR is pretty as well, and a more stripped down model with fewer parameters. Sure. But... It's nothing yolo with QFT. Common...

mantequilla

Turkey775 Posts

September 30 2016 09:30 GMT

#17

did you study astronomy teoita or is it a hobby?

Teoita

Italy12246 Posts

September 30 2016 10:03 GMT

#18

On September 30 2016 18:16 Cascade wrote:
Quite the blog, good job!

But the hate for QFT?? :o
You say that you like models that predict data, and then you say that you don't like QFT?

As i said, my main problem with QFT is the renormalization group which just feels wrong to my simpleton astronomer brain. It's a pretty known fact that GR and QFT arent compatible, i just so happen to be rooting for GR to be the least wrong of the two

my problem with scalar/vector/tensor fields (not tensors in general) comes from modified gravity models like MOND and TeVeS

On September 30 2016 18:30 mantequilla wrote:
did you study astronomy teoita or is it a hobby?

I have a master's in astrophysics and am currently starting a PhD at the University of Amsterdam.

Cascade

Australia5405 Posts

September 30 2016 10:33 GMT

#19

On September 30 2016 19:03 Teoita wrote:

Show nested quote +

my problem with scalar/vector/tensor fields (not tensors in general) comes from modified gravity models like MOND and TeVeS

Show nested quote +

I have a master's in astrophysics and am currently starting a PhD at the University of Amsterdam.

Both qft and gr are extremely successful theories. No matter how they are unified, that will not change.

From my particle physics perspective it feels more natural to add gravity as the fourth force to the standard model than somehow getting QM from some modified GR.

Teoita

Italy12246 Posts

September 30 2016 10:46 GMT

#20

Yeah im not debating that the Standard Model is crap, it isn't, it's just that from my astrohysics perspective between dark matter, dark energy (which, if associated with vacuum energy, QFT gets horribly wrong), matter/anti matter asymmetries, neutrino oscillations etc QFT as it is right now seems either lacking (the standard model) or a clusterfuck (supersymmetry and other extensions of the standard model).

Plus you can't just add gravity to the standard model because (or so i hear from my theoretical phsyicist firiends) it can't be renormalized like the other three forces can, you need to go beyond into string theory, loop quantum gravity or whatever, and those sure aren't predictive at all yet.

1 2 3 Next All

Please or register to reply.

Astrophysics blog - Cosmology

Completed

Ongoing

Upcoming