Introducing the Rosette Nebula / Stellar Nursery
When I first started to image stellar nurseries, I really didn’t know anything about them. I was told that stars are being born there – that is pretty awesome, but I was curious what was it about these light generating molecular clouds (MCs) that made them prolific star builders. Sure, stars are also created in turbulent dark molecular clouds, but stellar nurseries really churn out the stars at a much higher level – often creating whole open clusters of stars. Many of the stellar nurseries get very large and can even be mapped from their Halpha light signal in other galaxies. Ok, so my interest was piqued – I had to figure out why stellar nurseries appear the way they do, and what is it about them that is responsible for so much star production.
Fundamentally, a stellar nursery consists of one of more stars that pushes around a molecular cloud – in this case a central star cluster “pushes” the MC radially away from it. Somehow the stars are imparting a pressure on the surface of the cloud causing it to flow away from the cluster. At the same time, this pressure is exactly what is needed to accelerate the creation of more stars in the MC. So the story is simple and biblical – stars beget pressure that begets more stars.
The answer isn’t quite so simple though, because the pressure often comes with heat. This heat prevents the hydrogen and heavier elements from nucleating to form an embryonic protostar – a condensed body that can gravitationally attract additional hydrogen from the MC to bulk up and used as future fusion fuel. Dust, and its ability to radiate heat away, is employed by the MC to get rid of the heat, while the pressure from the stars is used to create new ones.
The processes of energy and mass transfer fundamentally controls the shape of the molecular clouds we like to image, and understanding this process will help any photographer to interpret what is going on in just about any SN image, and even tell when a nebula is or isn’t a stellar nursery (SN).
While the stars in a SN are shining visible wavelengths that we can see, they are also shining UV light that we cannot see. When UV light hits a hydrogen molecule, it first dissociates the molecule into atoms and then excites the atoms’ sole electron to a higher energy level within the atom. If the UV light is strong enough it will even cause the atom to lose its electron altogether, making it a ion. For astrophysicists, they call it an HII ion, but a more generally acceptable and correct nomenclature would be H+ ion or in this case, simply a proton. For this reason, SNs are often called HII regions even though the HII state, is likely this ionic state is likely the least populated of all the hydrogen states in an SN.
Fairly rapidly, coulombic (charge attraction) forces allow the H+ ion to find a new electron to lose its net positive charge and then begins a process of returning to its original, ground state. The now atomic hydrogen emits light of its own, generally in specific wavelengths (UV or IR) that we cannot see, but sometimes in specific (narrowband) wavelengths that we can. The most dominant of these wavelengths that is in the visible spectrum is a Halpha red. For this reason, most broadband RGB images of SNs are dominated by this red colour. Many astrophotographers don’t shoot SNs with broadband filters for colour, instead opting for narrowband filters that not only pick up the specific wavelength of Ha, but also other specific wavelengths given off by other substances – heavier atoms/ions that may be present.
Hydrogen atoms, if in the higher pressure/density MC, will find another partner to make a molecule and within a molecular cloud an equilibrium will take place where hydrogen as a molecule will vastly outnumber the free atoms. Outside of the molecular cloud (hence the name), where pressures and density are much lower, monatomic form hydrogen either as atoms or ions will dominate.
The dominant narrowband signal in most of our images of MCs is the red/orange/yellow light of excited hydrgoen atoms (Halpha) and it it brightest at the front facing the stars that are bombarding it with their UV light. The other dominant colour in our narrowband images can be blue (or white) associated with oxygen ions, most often seen in the ultra low density interstellar media (or ISM) between the stars and the molecular cloud. The brightness of all the emissions is indicative of the concentration of these elements in their respective regions, but is also modulated by the intensity of UV light hitting them and exciting their electrons.
Also modulating the brightness of the MC front is the presence of dust, that mutes and even blocks light from the hydrogen emissions, and the stars behind it. Unlike gases, dust can adsorb, emit and even reflect light in broadbands, according to its temperature. It is dust’s ability to adsorb UV and visible light that actually makes it visible to us as negative (dark) space against emissions coming to us from behind the dust.
Dust, as it turns out, is the Swiss Army knife of star formation. It can both cool and warm surrounding material and tranfer heat by bumping into it. It can transform adsorbed light into heat (kinetic energy), or even emit IR light to deep space to cool things off. While it blocks visible and UV light, IR light passes relatively free from adsorption. It is dominant within the molecular cloud, where its light/temperature abilities are crucial to the formation of stars.
Between the cluster and molecular cloud is an transparent expanding ISM bubble of ions that either appears blue or colourless depending upon the narrowband emissions of oxygen ions. The ions making up the bubble originate either from the stars or the receded MC.
Before I get into the nitty gritty, I need to address why stellar nurseries are such prolific star producers. As I suggested in previous blog, I believe the rate determining for star creation is the nucleation of stars through condensation of hydrogen and dust at low temperatures and relatively high pressure. In dark nebula, the high pressure is delivered at random places according to the vagaries of chaotic, turbulent flow. While in stellar nurseries, high pressures are created somewhat more consistently just behind the cloud displacement fronts, within the pillars/columns/elephant trunks, and even within isolated Bok globules of dust and hydrogen left behind as the main front receded. The raises the odds of star formation within these emission nebula considerably and the result is prolific star production. The key is how this high pressure is delivered to a nucleation site while maintaining the low temperature necessary for material to condense.
Nucleation sites in a stellar nursery occur much more frequently and at less random locations than in dark nebula as a result of stellar energy used to compress the molecular cloud.
The Rosette Nebula or Stellar Nursery (Monoceros) in Narrowband (SHO)
Askar 151phq; AP Mach2 GTO; ASI6200MM, – Chroma 5nm Narrowband Filters
R,G,B: (35,27,28 x 120s, Bin 1, Gain 100) – for stars; H,S,O: (52,51,43 x 600s, Bin 1, Gain 100)
Total integration time = 27.3 hrs (Dec 3 & 11, 2024; Jan 24-26, 2025) Maple Bay, BC
The beautiful Rosette Nebula configuration is that of a molecular cloud of hydrogen surrounding a large cluster of stars originally formed from the cloud itself. The cloud is receding radially from the central cluster, pushed away by the star cluster itself.
The cloud’s brightness and colour was captured using narrowband filters tuned to the wavelengths given of by monatomic hydrogen atoms, O+2 and S+ ions that have been disassociated, excited, and ionized by the UV light emanating from the star cluster. Bright yellow, orange and red indicate the presence of hydrogen (and sulphur) the most dominant element in the nebula. Blue colour indicates the presence of oxygen ions and the very high energy UV light required to ionize them. Also present is heavy elements, collectively called dust, that casts a dark shadow towards the camera.
When examining spheroidal nebula, such as the Rosette, one needs to bear in mind that one is looking into a sphere of semi-transparent gases. The best look one gets of the MC cloud front that faces the stars is to look at those parts of the nebula where the MC is seen edge on, rather than the thin front and back face, where only a faint indication of the narrowband emissions are apparent.
In real life, the nebula is some 130 light years in diameter and is likely near its prime as a star generator. Astronomers have identified nearly 2500 new stars in this nursery, with likely many more to emerge from behind the cloud displacement front, and those dark dusty pillars and globules seemingly left behind by the receding cloud.
The Molecular Cloud Front
The molecular cloud front that we see in SNs is different from those seen in either dark nebula or supernovae remnants. In dark nebula the edge of the molecular cloud is wispy due to the erosion of the cloud by the turbulent winds of the surrounding monatomic interstellar media. In supernovae remnants, the emitting substances are curved and forming sharp tendrils as a result of supersonic shock fronts. While turbulent flow also exists, the movement of the molecular cloud is governed by the compressible version of the Navier-Stokes equation – flow along a pressure gradient. In other words, an elevated pressure is imparted by the “blow” of the stellar cluster upon the molecular cloud and this pressure is highest at the front facing the stars. The natural flow of the cloud, along the pressure gradient is what causes the MC, as a whole, to recede from the stars.
While irregular and uneven, the front is sharp in the radial direction, creating a radial step change in composition and pressure/density moving from the ion bubble to the molecular cloud itself. Thanks to this lack of transition, many features of the MC front are captured in the image and this gives us a lot of clues as to how the high pressure is imparted to the edge of the cloud by the fairly remote stars.
Unlike a soap bubble that children like to blow, the resulting bubble in the MC formed by the “blow” of the stars is very uneven and far from the truly spherical that surface tension creates in a soap bubble. The frontal interface where the outside of the bubble pushes on the molecular cloud is very irregular, creating a mountainous landscape type of front.
The cause of this rough interface is two-fold – natural heterogeneity in the density and composition of the cloud itself coupled by the viscous fingering that occurs due to the thicker molecular cloud being displaced by the thinner bubble – but this is a huge interesting topic for a future post.
The variations in brightness and colour of the MC front is indicative of both its composition and intensity of the UV light arriving from the stars that energizes the MC visible light emissions. Parts of the front that directly face these stars and the UV light, emit the most visible light, and shine brighter in the image. Parts of the front that are at an angle to the UV starlight are dimmer. Dust, that is opaque to both visible and UV light also blocks emissions, and appear as dark areas against the background of MC emissions. All this requires considerable staring and visual explorations of the images to take in, but is critical to the understanding of what is going on.
At certain locations at the receding MC front, the concentration of dust will be higher than at others. Dust is composed of elements and molecules that are much heavier than hydrogen or helium and not so easily pushed by the expanding bubble. As a result, areas of high dust concentration tend to get left behind as the MC recedes back.
Not is the dust heavy, but it is also opaque and protects the MC behind it from receding. The result is that behind dust nodules, mountains can form behind them where the recession must occur from the sides. Upon further bubble expansion, the protected material can form a “pillar” sticking out from the receding MC front.
One thing worth noting, but parking for now, is the halo glow of bright Halpha emissions most pillars show at the tips facing the UV light from stars causing the bubble expansion. We’ll get to this shortly.
The head-on attack from the stars makes slower progress on any pillars than the bulk of molecular cloud. The more rapid recession of the cloud at large can leave the pillars isolated.
While somewhat dust protected in the shadow of the head of the pillar, the column’s more vulnerable sides also come under attack, causing them to eventually disperse. Ultimately, what was a column will be left as a dusty H2 cloud nodule, known as a “Bok Globule” seemingly isolated from the main receding cloud front.
You may be thinking that I have spent an inordinate amount of time describing the molecular cloud front, and particularly pillars and nodules, but behind the front and within the pillars and globules are where most of the new stars are formed. New stars formation is directly linked to the cloud recession away from the very stars it creates.
There is a lot that isn’t known about the contribution levels of the various mechanisms of molecular cloud recession, so I will describe three of them, and relate them to star production along the way. In the end, hopefully, I leave you with a better idea of how these stellar nurseries work.
Crowded and Cold
Mechanism 1 - Momentum From Stellar Winds Stellar Winds
Stellar winds consist largely of positive ions (and electrons) ejected at speeds from all of the stars at hundreds or even a thousand of km/s driven by the magnetic forces from the stars. They are similar to the solar winds that are thought responsible from driving away and dispersing the planet Mars’ atmosphere, but only a hundred million times stronger when emitted from massive, hot, blue stars like the ones within the Rosette. As they flow, they create a current and magnetic fields and can transfer their momentum to anything vaguely magnetic (containing dipoles) or ionic without a direct collision.
While other ions are undoubtedly present, the stellar winds largely consist of protons and electrons, so the amount of momentum they can impart on any single collision is limited by their mass. It is more the relentless onslaught that can get things moving. This is why we don’t see much in terms of supersonic fronts in these SNs. Rather than the bulldozer analogy used to describe a supernova shockfront, a better analogy for solar winds is trying to get a Volkswagen to move by shooting tennis balls at it. They will eventually get the small car to move, but it takes a relentless number of tennis balls. This is also why stars can pass through molecular clouds without creating multi-light year radius holes in them, and instead we see reflection nebula as stars pass through.
Here is a close-up of the central core of the Rosette Nebula, where stellar winds are likely strongest. The bright blue oxygen signal right of centre and near the largest stars in the cluster may indicate where stellar winds are colliding and creating interference patterns in the winds from different sources. It is possible that such collisions could raise both the density and temperature greatly, but I am pretty much guessing at this. Another alternative is that the stellar winds are colliding with radiation, perhaps an Xray source and forming a bowfront. This would also explain the curved, sharp bright signal we receive here.
It is difficult to assess how much impact the stellar winds have on pushing the MC away by the time it reaches the MC front 65 light years away. The largest, oldest stars in the Nebula are only 5 million years old, and this would require the stellar winds to be travelling at four times the maximum rate that astrophysicists tell us stellar winds can travel – and that is just to get there. The largest brightest stars, can shed up to 40% of their mass in solar winds over their lifetimes. That is a lot of mass, but miniscule compared to the entire mass of the molecular cloud. It is difficult to believe that such little mass could impart that level of momentum to the cloud.
To be sure, 5 million years ago when this nebula was just getting started, the stellar winds would rapidly blow the MC radially away from the stars. However, the influence of stellar winds at the current radius must be relatively small. If we think of the momentum carried by the stellar winds as a momentum flux, that flux would diminish radially with the square of the radius. In terms of the material filling the ISM bubble itself, even though it is at extremely low density, stills amounts to a lot of material just by it shear volume. I am not sure that the stars have that much mass to give to its stellar winds to continue to displace the molecular cloud.
Nonetheless, some of the observations we have made on the Rosette are consistent with stellar wind bombardment. This includes the apparent viscous instability creating the fingering /mountainous /undulating frontal shape and the formation of dusty pillars and Bok globules.
The mountainous landscape form is possible result of a more mobile (less friction and viscous drag) displacing a less mobile, higher drag/friction fluid – in this case the molecular cloud. When stellar winds impact upon the MC, the winds will make more displacement progress where the MC is “thinnest” or offering up the least resistance. The winds would tend to go around parts of the MC that are thicker, more viscous, and have higher inertia. This kind of unstable displacement, Staffman-Taylor instability – or viscous fingering, will result in the uneven displacement seen at the MC front and potentially describe how both pillars and Bok globules are formed. It is very interesting and I will be making it a focus point of a future posting on stellar nurseries.
In terms of pushing the MC away from the stars, direct stellar winds undoubted played a major role in clearing a volume around the stars to create the ionic bubble, but how much push stellar winds have in Rosette’s current geometry is questionable. Let’s look at some other potential mechanisms.
Mechanism 2 - Photon momentum transfer
Photons of light, while not having mass, do carry momentum. When photons are captured or reflected by a material, this momentum is imparted onto the material itself. The photons exert a “radiation pressure” at the material’s surface, and is worthy of a quick digression.
At right is Crooke’s device for demonstrating (visible) light momentum. It is in essence a light-mill and when light is shone upon it, the pressure on the blades will cause the blades to spin depending upon the intensity of the light shone upon it. It actually works on the principle that twice the momentum will be transferred to the shiny side of the blade, when light bounces off or reflected, than the dark side where the light is simply adsorbed.
You can see that the device is very delicate and sensitive because it needs to be – the pressure exerted by visible light is very small. Of particular note is that the device is encased in glass under a vacuum to remove the viscous drag – a concept that is all important to understand the movement of molecular clouds and even galaxies.
When lasers are used to bolster light intensity and the specificity of adsorption, all hell breaks loose on the imagination of scientists and engineers to invent new devices – everything from levitation devices, to super-cooling devices (my favourite), to light sails to propel spacecraft.
Undoubtedly, radiation pressure is a dominant force in the creation of x-ray bowfronts and in supernova explosions. Note that a suspected x-ray bowfront can be seen the blue-oxygen signal seen in the core of the Rosette. But in these two examples, the light is either or both of extreme intensity and extremely low wavelength. The momentum of a photon is inversely proportional to wavelength and xrays are able to adeptly deflect stellar winds or migrating interstellar medium away from x-ray source.
My reasons for suspecting that radiation pressure is only a bit player in stellar nurseries is two-fold. The first is gut feel. Here on earth, we cannot feel the pressure of UV or visible light from the sun, despite our proximity. It is no harder to walk towards the sun than with it at your back – even though at your back feels nicer. I liken the photons pushing the molecular cloud away from the stars to firing ping pong balls at the back of truck to get it moving.
My second reason is that if you examine the shape of x-ray bow fronts in images, they tend to have a smooth oval shape. In contrast, our image of the Rosette shows and irregular, mountainous shape. While you might expect dusty regions to cast a shadow of protection on the molecular cloud behind it to form pillars, it fails to explain how the frontal tips of the pillars would neck off to form Bok Globules.
You might find this argument somewhat subjective and indeed this is a hill I am not going to die on (at least not yet), but I think that UV light momentum from albeit bright stars, does not significantly contribute to either MC recession or the conception of new stars within the MC.
As our images illustrate, the impact of photons on our
Lighter molecules are the easiest to push away, but any molecules that are direcly hit by the UV are likely to dissociate into atoms, and further these atoms are likely become ionized, and actually can be recruited somewhat into the stellar winds. However, there isn’t a huge amount of inertial momentum in the alpha (Helium nuclei) particles, protons (Hydrogen nuclei), beta particles (electrons), and light photons and it takes a number of hits to get the heavier “metal” atom/ions to get moving. It is as if the star was trying to get a Volkswagen moving by shooting ping pong balls at it.
The stellar winds are much more rarefied (less dense) than the molecular cloud, consisting largely of H2 diatomic molecules it runs into, and also has much more bulk mass. Nevertheless, hydrogen is a light molecule and the cloud regresses from the star. Hydrogen right at the point where it is hit by the stellar radiation or ions is pushed back upon the molecular cloud material behind it and both the pressure and density increase. The added pressure is transmitted back through the remainder of the cloud and the whole cloud begins to move back under this radiation pressure.
At the same time, UV radiation, disassociates the hydrogen molecules into atoms, and can excite the atom’s electron or even strip the electrons off.
Progress is likely initially fast, but will slow down due to the spherical geometry of the winds and radiation outward dilution.
Most open clusters were likely formed in stellar nurseries, where the rate limiting step of star nucleation is accelerated by the application of radiation pressure on the cold molecular cloud causing material to condense.
Mechanism 3 - Photon Stimulated Jet Propulsion
Our images are testament to the energy impact, as opposed to just momentum impact, that stellar radiation of largely UV and visible light has upon the molecular cloud. In terms of our images, UV photons provide the energy for molecular dissociation, and ionization both within the expanding bubble and receding molecular cloud. It provides ions and atoms with the excitations necessary to emit all the light captured by our cameras. It shouldn’t come as a surprise that the this energy is also partly, or even primarily, responsible for the build-up of pressure, and to some degree heat, at the MC front.
UV light that happens to encounter dust either at the front or just behind it will cause the dust to warm as its energy is adsorbed and turned into sensible heat. The dust, when bumping into hydrogen molecules and atoms will pass on this heat causing the temperature at the front to rise. Temperatures as high as 1000K are thought to prevail at the ionizing MC front – toasty warm, but not extreme and definitely higher than the 10 to 20K that exists further behind the front.
Diatomic hydrogen gases are not able to capture these photons as heat directly, but rather dissociate to an atomic form – either directly or as a result of the indirect warming from dust. Hydrogen in atomic form is susceptible to either electron excitement or even ionization by adsorbing one of the photons directly. If ionized, leftover energy from the photon will result the kinetic energy and velocity of the resultant particles. Since there are a lot of electrons in proximity to the now H+ ion, recapture of an electron happens fairly rapidly and the now restored hydrogen atom emits its own light in terms of UV and visible radiation to eventually return to the ground state.
Thats a lot of process, but how does this impact pressure at the front. Firstly, when a hydrogen molecule dissociates, it creates two particles (atoms) from one (molecule) stoichiometrically, as we chemists like to say. Depending upon how isothermal this process is, this can cause the pressure to increase up to double – simply by having twice as many particles. Any rise in temperature associated with photon capture by particles, including dust, will also serve to increase pressure.
Hydrogen atoms thrown back behind the front will likely reassociate with another atom and through viscous drag, share its receding momentum with other molecules in the MC, making the whole front recede from the UV photons origin star. But we are used to thermal and stoichiometric expansion, not momentum gain in any one direction.
This net momentum is actually gained by considering hydrogen atoms that travel in the opposite direction – back towards the originating star – and counter to any photon pressure and stellar winds. It is as if the MC has jet engines – powered by the stars’ photons that power the MC’s recession. The molecular cloud has rocket engines that power its recession through momentum gained opposite to the jets. For the same reason, the molecular cloud front is a position of relatively high pressure – with the jets moving toward the star cluster while the MC redes from it.
The jetting hydrogen often appears as a very bright halo of Halpha signal (sometimes [SII] as well), almost outlining both the front and even around pillars and Bok globules. When processing images of stellar nurseries, I try to be careful not to blow out these halos, since they are likely to be the brightest Halpha signal in the image.
Occasionally our telescopes are big enough and the SN is close enough that we can actually make out individual jetstreams – as in the image left – of the hydrogen atoms leaving the front.
The jets will likely become part of the expanding bubble, only to be reionized by incoming photons. This helps explain where the mass of the bubble comes from. Indeed, it is likely that the most of the blue colour in the bubble comes from oxygen ions also jetted out by the molecular cloud
High Pressure, but Cold is what a Nucleating Star Needs
In addition to creating pressure at the MC front, needed to catalyze star production, the photons arriving at front also bring heat and an elevated temperature that can prevent new protostar condensation. Like goldilocks, the porridge of dust, helium, and hydrogen must be just right – pressure high enough and temperature cold enough.
Dust plays an important role in this pressure/heat balance. As far as UV and visible light goes, it is a black body emitter and adsorber. This allows dust to create a protective shell against heating behind it, adsorbing energy from incoming UV and re-emitting it back with only partial energy shared with surrouding material. Meanwhile, while dust readily emits ifrared (thermal) light, it does not block its escape. Dust within the protective shell can keep the contents cool through emissions to cold space outside of the nebula.
Another property of dust, is that its molecular weight and density are much higher than hydrogen giving it much higher inertia. Despite the hydrogen jets that increase the pressure on areas of high dust concentrations, these areas tend to lag behind the recession in the general MC. In such areas of high dust concentration, hydrogen lost to jetting further increases the dust concentration and this outer layer of very high dust ends up protecting all of the material in its shadow – creating a pillar. Conditions within the pillar, and particularly near the head end up being cold and at relatively high pressure – ideal for the creation of new stars.
Wind erosion coupled with different stars within the cluster shining at different angles can neck off the heads of pillars leaving the head detached from the main MC completely and create a Bok Globule. However, these Bok Globules also represent spots of high new star formation.
Stars also form within the molecular cloud itself. At left is an Chandra image I shamelessly copied from Wikipedia (I did donate to them). As the image indicates, area where dust concentrations are highest and the image is darkest, yet no too far from bright fronts where the flow within the MC is promoted by the high pressures induced by arriving photons.
Actual star nucleation / conception within the MC is further promoted further by turbulence caused not only by viscous flow, but electromagnetic forces that are present whenever ions and dust are involved. This process is similar to that seen when stars are created in dark nebula. Furthermore, star nucleation is likely promoted by the thermodynamic influence of dust chemicals helping the material to condense.
In any event, there is a lot more things to explore in future postings I will make regarding stellar nurseries. If you enjoyed this posting (a little long, I know), please leave me some feedback from the home page.
