Professional Documents
Culture Documents
V. S. C. Santos
February 2023
1
d= pc (1)
p”
Where 1pc = 2.062 × 105 AU = 3.0857 × 1016 m
And p = parallax angle.
Thus 1 parsec is the distance from which the radius of Earth’s orbit, 1AU ,
subtends an angle of 1”.
• Luminosity:
”Brightness” of a star. Total amount of light energy of all wavelengths
emitted per unit time. Formula given by SB equation which we will see
later.
1
• Radiant Flux:
”Brightness” of a star measured by observer. Total amount of light energy
of all wavelengths that crosses a unit area oriented perpendicular to the
direction of the light’s travel per unit time. This is the inverse-square law
for light.
L
F = (2)
4πr2
• Absolute Magnitude M :
Apparent magnitude a star would have if it were 10 pc away.
• Relationship between magnitude and radiant flux:
F1
m1 − m2 = −2.5 log10 (3)
F2
• Distance Modulus (m − M ):
Since m depends on the observer’s distance and M ”depends” on a con-
stant distance, the quantity (m − M ) is a measurement of the observer’s
distance to the star.
d
m − M = 5 log1 0(d) − 5 = 5 log10 (4)
10pc
• At first glance, it may seem that astronomers must start with the measur-
able quantities F and m and then use the distance to the star (if known) to
determine the star’s intrinsic properties. However, if the star belongs to an
important class of objects known as pulsating variable stars, its intrinsic
luminosity L and absolute magnitude M can be determined without any
knowledge of its distance. The d equation then gives the distance to the
variable star. These stars act as beacons that illuminate the fundamental
distance scale of the Universe.
2
Figure 1: Blackbody radiation spectrum.
L = AσT 4
3
– V : visual magnitude. Filter centered at 550nm with bandwidth of
89nm.
• Colour index:
Difference between two colour magnitudes. (e.g. U − V , V − B)
4
1.6 Perguntas
1. Porque é logarı́tmica a escala de magnitudes? Há outros tipos de medida
que encontramos na vida quotidiana e que também estejam numa escala
logarı́tmica? Explique.
R: A escala de magnitudes serve para quantificar o brilho das estrelas
como percebido pelo olho humano, que por sua vez detecta o brilho em
escala logarı́tmica. Exemplos de outras coisas em escala logarı́tmica:
• escala de pH
• escala de Richter
• entropia em termodinâmica
2. Que fatores podem ter feito com que a observação astronômica clássica
fosse realizada na região ”visı́vel” do espectro?
R: Primeiro, não se sabia da existência das cores invisı́veis. A atmosfera
absorve a luz advinda de fora, a depender do seu c.d.o., de acordo com o
seu espectro de absorção. A luz visı́vel é, obviamente, transmitida e não
absorvida.
3. Como é que quantificamos o conceito de cor?
R: No contexto de astronomia, utilizamos o ı́ndice de cor.
1.7 Problemas
1. A magnitude aparente do Sol é −26.8. Quão mais brilhante aparenta ser
o Sol que a estrela mais brilhante, que tem m = −1?
R:
f1
= 100(m2 −m1 )/5 ⇔
f2
fSol
⇔ = 100(−1+26.8)/5 ⇔
fEstrela
fSol
⇔ = 2.1 × 1010
fEstrela
2. • Qual é o módulo da distância do Sol?
R:
5
3. • O espectro contı́nuo de uma estrela tem o pico em 400 nm. Qual é a
temperatura da estrela?
R:
λmax T = 2.9 ∗ 10−3 mK ⇔
2.9 × 10−3
T = = 7250K
400 × 10−9
• E se o pico estivesse a 450nm?
R:
T = 6444.4
Ou seja, quando aumentamos o comprimento de onda, diminui-se a
temperatura: c.d.o.’s mais baixos correspondem a fotões mais en-
ergéticos.
4. Se dobrarmos a temperatura de um corpo negro, quanto deverı́amos diminuir
a superfı́cie para manter a luminosidade constante?
R:
1
L = AσT 4 = cAσ(2T )4 = 16cAσT 4 ⇒ c =
16
Isto é, precisamos diminuir a área 16x.
2 Spectral Lines
Do capı́tulo 3 do Kutner, e do capı́tulo 5 de Caroll e Ostlie, ”The Interaction of
Light and Matter”.
6
• If white light only passes through a prism, colours overlap and we don’t
see a clear spectrum. If we let the light pass through a slit first, however,
then the light is spread out (wave characteristic) and we can see a clear
spectrum. If a colour is missing from the white light, it will show as a gap
on the spectrum.
• Every element produces its own pattern of spectral lines and thus may be
identified by its unique spectral line ”fingerprint”.
• Kirchhoff ’s Laws:
– A hot, dense gas or a hot solid produces a continuous spectrum with
no spectral lines. This is the continuous spectrum of blackbody radi-
ation emitted at any temperature above absolute zero and described
by the Planck functions Bλ (T ) and Bν (T ).
– A hot, diffuse gas produces bright spectral lines (emission lines).
Emission lines are produced when an electron makes a downward
transition from a higher to a lower orbit. The energy lost by the
electron is carried away by a single photon.
– A cool, diffuse gas in front of a source of a continuous spectrum
produces dark spectral lines (absorption lines) in the continuous
spectrum. Absorption lines are produced when an electron makes
a transition from a lower orbit to a higher orbit. If an incident
photon in the continuous spectrum has exactly the right amount of
energy, equal to the difference in energy between the electron orbits,
the photon is absorbed by the electron, which makes the upward
transition.
• Rydberg Formula:
Since electrons can only absorve/emit photons in certain quantised wave-
lengths, we relate the transition levels and wavelength using this formula.
!
1 1 1
=R − 2 (9)
λ n2f ni
7
• In a star, we are not talking about the radiation from a single hydrogen
atom, but from a large number of them. However, having a lot of hydrogen
does not mean a star will have strong Hα absorption in the spectrum we
see from it. In order for absorption to take place, a significant number of
atoms must be ready to absorb a photon, with n ≥ 2. If all the hydrogen
is in n = 1 there will be no H absorption.
• Population: number of atoms per unit volume in a given state.
• Excitation Process: process that alter populations, like photon emis-
sion/absorption and collisions between atoms.
• Collisions: when two atoms collide, atom 1 is in state i and atom 2 is
in state j (we’re really referring to the electrons here by the way). In the
process, the kinetic energy of atom 2 changes by Ej − Ei .
8
• Average Kinetic Energy per atom:
3
K= kT (10)
2
9
• Saha Equation:
3/2
ne n(Xr+1 ) 2gr 2πmr kT
= eEi /kT (12)
n(Xr ) gr+1 h2
Where
– ne = electron density
– n(Xr ) and n(Xr+1 ) are the densities of the r and r + 1 ionization
states, respectively, of element X
– Ei = ionization energy
– gr and gr+1 are the statistical weights of the ground electronic states
of Xr and Xr+1 .
• The roman numeral I is used to designate the neutral species, II the singly
ionized species, III the doubly ionized species, and so on. For example,
neutral hydrogen is H(I), ionized hydrogen (H + ) is H(II), doubly ionized
carbon is C(III).
• Strength of spectral lines:
We are now in a position to determine the strengths of the various absorp-
tion lines in stars. We will take Hα as an example to see the combined
effects of excitation and ionisation.
– At low temperatures, essentially all H is neutral and most of it is in
the ground state. Since little H will be in n = 2, there will be few
chances for Hα absorption, and the Hα line will be weak.
– At moderate temperatures, most hydrogen is still neutral, but a rea-
sonable amount will be in n = 2. As the temperature increases, the
Hα line gets stronger.
– At very high temperatures, the hydrogen becomes ionised. Since
there is less neutral hydrogen, there are less electrons making Hα
absorption and the line gets weaker.
We can apply a similar analysis to other elements. The details will differ
because of different energy level structures and different ionization ener-
gies. It should be noted that, after H and He, the abundances of elements
fall off drastically.
10
Figure 3: HR diagram.
• Stars are only in certain places of the diagram. This tells us that arbitrary
combinations of luminosity and temperature are not allowed. Remember,
for a given temperature, the luminosity depends only on the radius of the
star, so the HR diagram tells us that arbitrary combinations of radius and
temperature are not allowed.
• Main Sequence: The narrow band where most stars are found. The
significance of this sequence is that, since most stars of the same tempera-
ture have the same luminosity (the sequence is kinda vertical), and hence
essentially the same size.
• This close relationship between size and temperature must be a result
of the laws of physics as applied to stars. It gives us hope that we can
understand stellar structure by applying the known laws. It also gives us
a crucial test: any theory of stellar structure must predict the existence
of the main sequence.
• Giants: Stars that appear above the main sequence, meaning they are
more luminous, and thus larger, than other stars of the same temperature
(by L = AσT 4 that means they have larger A, i.e. larger in size. So, if
two stars have the same temperature but one appears more luminous, it
must be larger). Divided between subgiants, giants and supergiants.
• Dwarfs: Stars in the main sequence.
• White Dwarfs: Stars below the main sequence. Really tiny. Called white
because most are in the middle spectral types, i.e. appear white.
11
• Luminosity Class: keeps track of the size of a star (L = AσT 4 ).
– Main sequence: V
– Subgiants: IV
– Giants: III
– Between Giant/Supergiant: II
– Supergiant: Ia (really big), Ib (really f****** big)
12
4. Por que não vemos riscas de absorção de hélio em estrelas como o Sol?
R:
2.3 Problemas
1. Calcule o comprimento de onda da transição Hβ.
R:
Hβ: 4 → 2 " #
1 1 1
= RH − 2 ⇔
λ n2i nf
λ = 486.3nm
1
Ef − Ei = 0 − (−13.6eV) = −3.4eV
4
Isto é, é necessário 3.4 eV para arrancar o eletrão do átomo.
3
Ec = kB T ⇔ T = 26280.19K
2
Temos
13
Então
T = 19720K
Para nj /ni = 10/100 = 0.1, teria-se
T = 32030K
14
6. Calcule a razão entre os raios de estrelas M0Ia e M0V.
R:
m − M = 5 log10 d − 5 ⇔ M ≈ 0.505
• Intensidade especı́fica
∂I Eλ dλ
Iλ = = (13)
∂λ dλdtdA cos θdΩ
Constante (não diverge) ao longo do percurso do raio no espaço vazio.
Figure 4: Intensidade Iλ
15
• Intensidade média Em geral, a intensidade varia com a direção. Define-
se portanto uma intensidade média de toda a esfera da estrela:
Z Z 2π Z π
1 1
⟨Iλ ⟩ = Iλ dΩ = Iλ sin θdθdϕ (14)
4π 4π 0 0
É a energia média por comprimento de onda, por segundo e por área que
passa na sua direcção perpendicular (eixo z).
Para um campo de radiação isotrópico, não há transporte de energia
lı́quido (lı́quido no sentido econômico):
Fλ,isotrópico = 0 (16)
– Fonte resolvida ⇒ Iλ
– Fonte não resolvida ⇒ Frad
• Pressão de radiação
– transmissão: Z
1
Prad,λ dλ = Iλ dλ cos2 θdΩ
c sphere
16
– campo de radiação isotrópico:
4π
Iλ dλ
3c
se o campo não for isotrópico o cálculo pode ser mais complicado
– pressão total Z ∞
Prad = Prad,λ dλ
0
Iλ = Iλ,0 e−κλ ρs
dR = l × N
17
• Ganho de intensidade por emissão:
1 dIλ
− = Iλ − Sλ
κλ ρ ds
3.1 Aproximações
• Aproximação de atmosfera plano-paralela:
18
– Profundidade ótica vertical:
Z 0
τλ,v (z) = κλ ρdz
z
Tem-se
τλ,v
τλ = = τλ,v sec θ
cos θ
Reescrevemos a equação de transferência como
dIλ
cos θ = Iλ − Sλ
dτλ,v
• Aproximação da atmosfera cinzenta: elimina dependência em λ
– O valor de τv,z depende do comprimento de onda, porque τλ,v (z) =
R0
κ ρdz
z λ
– Para simplificar a análise, considera-se que a opacidade é indepen-
dente do comprimento de onda. Usamos κ (opacidade média de
Rosseland) ⇒ Podemos escrever τv ao invés de τλ,v
– As outras dependências ao comprimento de onda podem ser elimi-
nadas integrando
R∞ a equação
R ∞ de transferência sobre todos os c.d.o.,
i.e. I = 0 Iλ dλ S = 0 Sλ dλ
– Ficamos com uma equação de transferência para atmosfera cinzenta,
dI
cos θ = =I −S
dτv
– Depois de fazer alguma bruxaria (pag. 294 Caroll Ostlie) temos
dFrad
= 4π(⟨I⟩ − S)
dτv
– Em coordenadas cartesianas, temos
dPrad κρ
= − Frad
dr c
– Em coordenadas esféricas, temos
dPrad 1
= Frad
dr c
The above result can be interpreted as ”the net radiative flux is driven
by differences in radiation pressure, with a photon wind blowing from
high to low pressure”
• No equilı́brio, todo processo de absorção é balanceado por um processo de
emissão e vice-versa ⇒ numa atmosfera plano-paralela temos o fluxo
radiativo com o mesmo valor em todos os nı́veis da atmosfera, incluindo
sua superfı́cie
Frad = constant = Fsurf = σTe4
19
• Depois de mágica (Carroll Ostlie pag. 294) obtemos
3
⟨I⟩ = Frad (τv + 2/3)
4π
20
3.2 Perfil das linhas espectrais
• Profundidade da linha:
(Fc − Fλ )/Fc
1. Alargamento natural:
Princı́pio da incerteza de Heisenberg: ∆E ≥ 4πh ∆t ⇒ menos tempo
de medição = mais incerteza na energia, e um elétron num estado
excitado só fica lá por um breve perı́odo ∆t ⇒ há uma incerteza
natural que alarga a linha espectral.
λ2
1 1
∆λ ≈ +
2πc ∆ti ∆tf
Onde ∆ti e ∆tf são os tempos de vida do eletrão nos estados inicial
e final respectivamente
2. Alargamento de Doppler: q
2kT
Velocidade média das partı́culas dada por M-B: v = m
21
Os c.d.o. da luz absorvida/emitida pelos átomos no gás são doppler-
shifted de acordo com (não-relativista) ∆λ/λ = ±|vr |/c
Logo, r
2λ 2kT
∆λ ≈
c m
Análise mais profunda:
r
2λ 2kT ln 2
(∆λ)1/2 = (asas decrescem exponencialmente)
c m
Considerando turbulência com dist. de velocidades segundo M-B,
s
2λ 2kT 2
(∆λ)1/2 = + vturb ln 2
c m
22
• Densidade de coluna N : suponha que um tubo oco com seção de 1m2
conectasse a superfı́cie e o observador. O tubo conteria, então N átomos
do tipo especificado.
• Densidade de coluna de átomos absorventes: Na
3.3 Problemas
1. Calcule até que distância poderia ver através da atmosfera terrestre se
tivesse a opacidade da fotosfera Solar. Use κ5 00 = 0.03m2 kg −1 para a
opacidade Solar e 1.2kgm−3 para a densidade da atmosfera Terrestre.
R:
κ5 00 = 0.03m2 kg −1 , ρ = 1.2kgm−3
1
livre percurso médio de um fotão: l = = 27.7778m
κλ ρ
distância até a qual enxergamos: d = τλ l
vemos fotões a partir de τλ = 2/3
2
d= × 27.7778 = 18.52m
3
2. Use as leis relativistas da conservação de energia e do momento para provar
que um electrão isolado não pode absorver um fotão.
R:
Ei = Ef ⇔ Eγ + me c2 = γme c2 ⇔ Eγ = me c2 (γ − 1)
Eγ
pi = pf ⇔ = γme v ⇔ Eγ = γcme v
c
Juntando os dois resultados,
v γ v2 v
= ⇔ 2 =
c γ−1 c c
Aqui só há duas possibilidades: v = 0 e v = c. Um elétron não pode
ter v = c pois tem massa de repouso, pelo que só podemos ter v = 0.
Obviamente isto é uma contradição, haja vista que o elétron absorve o
fóton.
23
3. A densidade central do Sol é, segundo um modelo, 1.53 × 105 kgm−3 e a
opacidade média de Rosseland no centro é de 0.217m2 kg −1 .
d=l×N
24
R:
dr
dr = ds cos θ ⇔ ds =
cos θ
Logo, a equação assume a forma desejada.
(b) Use esta forma da equação de transferência para deduzir que
dPrad κρ
= − Frad
dr c
R:
RESOLUÇÃO ERRADA. LER CAROLL E OSTLIE
5. Para uma atmosfera plano-paralela, mostre que a aproximação de Edding-
ton leva às seguintes expressões para a intensidade média, fluxo radiativo
e pressão radiativa:
1 4π
⟨I⟩ = (Iout + Iin ); Frad = π(Iout − Iin ); Prad = ⟨I⟩
2 3c
R:
• Encontrando ⟨I⟩: Z
1
⟨Iλ ⟩ = Iλ dΩ
4π
Com a aproximação da atmosfera cinzenta, consideramos que não há
variação com o c.d.o.:
Z Z 2π Z π
1 1
⟨I⟩ = IdΩ = I sin θdθdϕ
4π 4π 0 0
25
Computando a integral acima, ficamos com
1
⟨I⟩ = (Iin + Iout )
2
• Encontrando Frad :
Z
Fλ dλ = Iλ dλcosθdΩ
• Encontrando Prad :
Z
1
Prad,λ dλ = Iλ dλ cos2 θdΩ
c sphere
1 2π π
Z Z
Prad = I sin θ cos2 θdθdϕ ⇔
c 0 0
2π π
Z
⇔ Prad = I sin θ cos2 θdθ
c 0
Aproximação de Eddington:
!
Z π/2 Z π
2π 2 2
Prad = Iin sin θ cos θdθ + Iout sin θ cos θdθ
c 0 π/2
26
Além disso, sabemos que
3σ 4
Frad = σTe4 , ⟨I⟩ = T (τv + 2/3)
4π e
Encontramos então
3σTe4 τv σTe4
3τv
Iin = Iout = +1
4π π 4
27
9. Considere uma camada plano-paralela de gás com espessura L, mantida
a uma temperatura constante T . Assuma que o gás tem profundidade
ótica τλ,0 , com τλ = 0 no topo da superfı́cie da camada. Assuma, ainda,
que mais nenhuma radiação exterior passa pelo gás. Use a forma geral da
equação de transferência
Z 0
Iλ (0) = Iλ,0 e−τλ,0 − Sλ e−τλ dτλ
τλ,0
⇔ Iλ (0) = Sλ (1 − e−τλ,0 )
• Se τλ,0 >> 1:
Iλ (0) = Sλ τλ,0
Iλ (0) = jλ ρL
28
10. Considere a camada do problema anterior, mas em que agora radiação
vinda de fora entra pela base da camada. Use a equação geral da equação
de transferência para mostrar que quando olhamos de cima para a camada,
vemos a radiação do corpo negro se τλ,0 >> 1. Se τλ,0 << 1, mostre que vê
linhas de absorção sobrepostas ao espectro da radiação incidente quando
Iλ,0 > Sλ e linhas emissão sobrepostas ao espectro da radiação incidente
quando Iλ,0 < Sλ . Pode considerar que a source function, Sλ , não varia
com a posição dentro do gás e assumir equilı́brio termodinâmico quando
τλ,0 >> 1.
R:
4 Interiores Estelares
4.1 Hydrostastic Equilibrium
Context: Stellar spectra doesn’t give us much data on a star’s interior, which is
why we’re studying this.
29
4.1.1 Determining the internal structures of stars
• How do we determine the interal structures of stars: put known, truthful
laws of physics on PC, simulate, make conclusions. The truthful data we
already have forces the other information to be truthful as well.
• We know that stars radiate energy, and they don’t have an infinite supply
of energy, so eventually they DIE.
• Stellar evolution is the result of a constant fight against gravity.
2
• F = ma ⇔ Fg + FP,top + FP,bottom = dm ddt2r
30
• However,
– FP,t = −(FP,b + dFP )
– Fg = −G Mrr 2dm , such that Mr = mass inside the sphere of radius r.
The contribution to the gravitational force by spherically symmetric
mass shells located outside of r is zero.
– P = F/A ⇔ F = P A ⇒ dFP = AdP
– dm = ρAdr
2
• Thus we get dm ddt2r = −G Mrr 2dm − AdP
2
• Dividing by the volume V = Adr, we get ρ ddt2r = −G Mrr2 ρ − dP
dr
From Figure 7, assuming the shell is sufficiently thin (dr << r), the vol-
ume of the shell is, approximately, dV = 4πr2 dr. Dividing by dr we get the
31
mass conservation equation, which dictates how the interior mass of a star must
change with distance from the center.
Mass conservation equation:
dMr
= 4πr2 ρ (19)
dr
1 ∞
Z
P = np pvdp (20)
3 0
P = N kT /V (21)
32
• Ideal Gas Law in Terms of Mean Molecular Weight:
ρkT
Pg = (23)
µmH
• The mean molecular weight depends on the composition of the gas as well
as on the state of ionization of each species. The level of ionization enters
because free electrons must be included in the average mass per particle
m. This implies that a detailed analysis of the Saha equation is necessary
to calculate the relative numbers of ionization states. When the gas is
either completely neutral or completely ionized, the calculation simplifies
significantly, however.
• For a neutral gas:
1 X 1
= Xj (24)
µn j
Aj
and
1 1 1
=X+ Y + Z (25)
µn 4 A n
where
– Aj = mj /mH
– Xj = mass fraction of atoms of type j
1
– A n = weighted average of all elements in the gas heavier than
helium. For solar abundances, A1 n ≈ 1/15.5
• For a completely ionised gas: now we must include the free electrons
as well (e.g. each H atom contributes 1 nucleus and 1 free electron)
1 X 1 + zj
= Xj (26)
µi j
Aj
33
• Contribution due to Radiation Pressure: Because photons have mo-
mentum p = E/c, they are capable of delivering an impulse to other par-
ticles during absorption or reflection. Consequently, EM radiation results
in another form of pressure.
1 4
Prad = aT (28)
3
• Total Pressure:
Pt = Pg + Prad
dm = 4πr2 ρdr
Thus
Mr 4πr2 ρdr
dUg = −G
r
– Integrate over all shells:
Z R
Ug = 4πG Mr ρrdr
0
Substitute in:
3 GM 2
Ug ≈ −
5 R
– Apply virial theorem: E = 12 U
3 GM 2
E≈− (29)
10 R
34
– Assuming the Sun was much larger in the past, we have Ri >> RS .
Then the energy radiated away would be
3 GMS2
∆E = −(Ef − Ei ) ≈ −Ef ≈ ≈ 1.1 × 1041 J
10 RS
GM 2
tKH = (30)
2RL
• Mass Number:
A=Z +N (31)
Since protons and neutrons have very similar masses which greatly exceed
that of the electron, A is a good indication of the mass of an isotope.
• Fusion: when lighter particles combine to form heavier particles, the
product has less mass. By conservation of mass-energy we know therefore
that the reaction released energy.
• Nuclear Timescale: assume Sun was originally 100% hydrogen and that
only the inner 10% of the Sun’s mass becomes hot enough to convert
hydrogen to helium. ∆m = 4mH − 1mHe = 0.03u, so ∆m/mi = 0.007,
i.e. only 0.7% of the mass of hydrogen would be converted to energy in
forming a helium nucleus. The amount of nuclear energy available in the
Sun would then be
35
Figure 9: The potential energy curve characteristic of nuclear reactions.
Z1 Z2 e2
Tclassical = ≈ 10K
6πϵ0 kr
Z12 Z22 e4 µm
Tquantum = ≈ 107 K (32)
12π 2 ϵ20 h2 k
4 11 H → 41 He + 2e+ + 2νe + 2γ
36
Figure 10: The three branches of the pp chain, along with the branching ratios
appropriate for conditions in the core of the Sun.
• CNO Cycle: another cycle for the production of helium-4 from hydrogen.
Here, carbon is a catalyst. It is not destroyed in the chain and must
be present in the original material. To work in a stationary state, the
abundances of the isotopes in the slower reactions have to be higher in
order to have the same rate as the faster reactions.
• Low mass stars, which have smaller central temperatures, are dominated
by the pp chains during their ”hydrogen burning” evolution, whereas more
massive stars, with higher central temperatures, convert hydrogen to he-
lium by the CNO cycle.
• Helium combustion: When hydrogen is depleted in the nucleus via pp
37
or CNO chains, gravity causes the nucleus to contract, heating it up and
provoking helium combustion.
• Triple-alpha process: reaction sequence via which helium is converted
into carbon.
4
2He + 4 2He ↔ 8 4Be
8
4Be + 4 2He → 12 6C + γ
why the most abundant nuclear species in the Universe are, in order,
1
1H, 4 2He, 16 8O, 12 6C, 20 10N e, 14 7N, 24 12M g, 28 14Si, 56 26F e
38
These abundances are the result of the dominant nuclear reaction processes
that occur in stars, together with the nuclear configurations that result in
the most stable nuclei.
• The study of stellar nucleosynthesis strongly suggests that all heavier nu-
clei were generated in the interiors of stars.
• Energy Generation Equation AKA Luminosity Gradient Equation: To
determine the luminosity (power output) of a star, we must now consider
all of the energy generated by stellar material. The contribution to the
total luminosity due to an infinitesimal mass dm is simply dL = ϵdm,
where ϵ = ϵnuclear + ϵgravity is the total energy released per kilogram per
second by all nuclear reactions and by gravity. For a spherically symmetric
star, the mass of a thin shell of thickness dr is just dm = dMr = ρdV =
4πr2 ρdr. Substituting and dividing by the shell thickness, we have
dLr
= 4πr2 ρϵ (33)
dr
Where Lr = interior luminosity due to all of the energy generated within
the star’s interior out to the radius r.
39
• Pressure Scale Height HP (Convection energy transfer):
If the temperature gradient dT
dr becomes too steep (too negative), convec-
tion can begin to play an important role in energy transfer. Pressure scale
height = measure of characteristic length scale for convection.
P
HP = (35)
ρg
dU = dQ − dW (36)
such that
– dU = internal energy of mass element dm
– dQ = heat added to mass element
– dW = work done by mass element
Throughout our discussion, we will assume that these energy changes are
measured per unit mass.
• State function: depends only on present conditions of the gas, not on the
history of changes leading to that state. U is a state function, therefore
dU is independent of the actual process involved in the change.
• Inexact differential: depends on the ways in which the processes are
carried out, i.e. they are path-dependent. dQ and dW are inexact differ-
entials.
• Kinetic Energy in Ideal Gas
3
K= kT (ideal gas) (37)
2
• Specific Heats:
∂Q ∂Q
CP = |P CV = |V (40)
∂T ∂T
40
• Work per unit mass: Suppose a cylinder with cross-sectional area A, filled
with gas of mass m and pressure P . The gas exerts a force F = P A on an
end of the cylinder, which is a piston that moves by dr. Then the work
done by the gas is
F PA
dW = dr = dr = P dV (41)
m m
Such that V = 1/ρ specific volume, or volume per unit mass.
• We can rewrite the 1st law as dU = dQ − P dV . At const. volume,
dV = 0 ⇒ dU = ∂Q
∂T |V dT , which is
dU = CV dT (always) (42)
It’s always true because dU is independent of any specific process
• But we saw that dU = 3nR/2 for a monatomic gas. Thus
3
CV = nR (monatomic gas) (43)
2
• Now finding CP for monatomic gas. First note that
∂Q ∂V
dU = |P dT − P |P dT
∂T ∂T
Also let’s consider all possible differential changes in pV = nRT and we
get
P dV + V dP = RT dn + nRdT
Constant P, n ⇒ dP = dn = 0 ⇒ P dV dT = nR. Together with the first
∂Q
equation, dU = CV dT and CP = ∂T |P we get
CP = CV + nR (monatomic gas)
• ”Adiabatic Gamma”:
CP
γ= (44)
CV
• Adiabatic Process: dQ = 0, i.e. no heat flows in/out of the mass
element dm.
• Adiabatic Gas Law:
From the 1st law and dQ = 0 we get dU = −P dV . With the differentiated
1st Law with constant n, we get P dV + V dP = nRdT .
dU
Also, since dU = CV dT , we have dT = C = − PCdV . Combining the two
V V
nR
results, we get P dV + V dP = − C V
P dV .
That may be rewritten to yield γ dV dP
V = − P . With P V = nRT we finally
get
PV γ = K (45)
Where K = constant.
41
• Adiabatic Temperature Gradient dT dr |ad (also convection):
Consider the situation where a hot convective bubble of gas rises and ex-
pands adiabatically, meaning it doesn’t exchange heat with its surround-
ings. After it has traveled some distance, it finally thermalises, giving up
any excess heat and dissolving into the gas.
(After derivation in book p.412):
dT 1 µmH GMr
|ad = − 1 − (46)
dr γ k r2
It is also convenient to write that as
dT g
|ad = − (47)
dr CP
This describes how the temperature of the gas inside the bubble changes
as the bubble rises and expands adiabatically.
• Superadiabatic Temperature Gradient:
If the star’s actual temperature gradient (designated by subscript ”act”)
is steeper than the adiabatic temperature gradient, i.e.
dT dT
> (48)
dr act dr ad
42
Figure 13: Convection bubble vs. surrounding gas
That is, interior mass (mass inside sphere of radius r inside star) and
luminosity approach zero at the center.
2. At the surface of the star:
T → 0
r → R∗ ⇒ P → 0 (55)
ρ→0
43
• It is not generally possible to solve the system of stellar structure equa-
tions and their associated constitutive relations analytically, so we need
numerical solutions to ”build” stellar models. However, under very special
situations, we can find analytical solutions to a subset of equations.
• Polytrope or Polytropic Model: Star model in which the pressure de-
pends on density in the form
P = Kργ
Where
– Dn = dimensionless function that is used to determine density ρ.
0 < Dn < 1
– ρ(r) = ρc [Dn (r)]n
– n = 1/(γ − 1) = polytropic index
– ξ = dimensionless independent variable that is used for the radius r.
– r = λn ξ
– Dn (ξ1 ) = 0 specifies the surface at ξ = ξ1
h (1−n)/n i1/2
Kρc
– λn = (n + 1) 4πG
Boundary conditions:
– At the center (ξ = 0): D(0) = 1 and dD/dξ = 0
– At the surface (ξ = ξ1 ): D(ξ1 ) = 0
• Analytic Solutions:
ξ2
√
– n = 0 ⇒ D0 (ξ) = 1 − 6 , with ξ1 = 6
44
sin ξ
– n = 1 ⇒ D1 (ξ) = ξ , with ξ1 = π
– n = 5 ⇒ D5 (ξ) = [1 + ϵ2 /3]−1/2 , with ξ1 → ∞
r = L = O at M = 0, ρ = T = 0 at M = MS
45
• Repeating the calculations for each mass and variations of the other pa-
rameters is obviously computationally retarded. We would thus like to
normalise the variables in a way that the solutions are independent of
the total mass, having fixed the chemical composition.
• Homologous model:
4.6 Problemas
1. Mostre que a equação de equilı́brio hidrostático também pode ser escrita
em termos da profundidade óptica como:
dP g
=
dτ κ
R:
Profundidade ótica: dτλ = −κλ ρds. Tomemos o valor médio do compri-
mento de onda: dτ = −κρds
Equilı́brio hidrostático: dP
dr = −ρg.
Nesse contexto, ds = dr = vetor que segue a direção radial para a su-
perfı́cie. Então,
dP g
=
dτ κ
2. Usando o teorema do virial, faça uma estimativa grosseira da temperatura
tı́pica do Sol.
R:
De (29):
3 GM 2
Eg ≈ −
10 R
Da termodinâmica,
3
K = N kT
2
Do Teorema do Virial,
1
K=− U
2
Assim,
3 3 GM 2 GM 2
N kT = ⇔T =
2 10 R 5N kR
Nesse caso, N = número de partı́culas no Sol, ou seja N = MS /mH × 2. 0
×2 é porque, presumindo que todo o hidrogênio está ionizado, cada átomo
de hidrogênio contribui 1 núcleo e 1 elétron livre como partı́culas. Assim,
T = 2.313 × 106 K
46
3. A partir da equação de Lane-Emden e impondo as condições fronteira
necessárias, mostre que a solução do politropo n = 0 é dada por:
ξ2 √
D0 (ξ) = − , ξ1 = 6
6
R:
1 d 2 dDn
ξ = −Dnn
ξ 2 dξ dξ
Se n = 0,
1 d 2 dD0
ξ = −1 ⇔
ξ 2 dξ dξ
d 2 dD0
⇔ ξ = −ξ 2 ⇔
dξ dξ
dD0 1
⇔ ξ2 = − ξ3 + C
dξ 3
dD0
Condição fronteira: dξ = 0 no centro (ξ = 0).
0=0+C ⇔C =0
dD0 1
ξ2 = − ξ3 ⇔
dξ 3
ξ2
⇔ D0 (ξ) = − +C
6
Condição fronteira: Dn (0) = 1
1=0+C ⇔C =1
ξ2
D0 (ξ) = − +1
6
À superfı́cie, D0 (ξ) = 0. Então,
ξ12 √
− + 1 = 0 ⇔ ξ1 = 6
6
47
a) Estime os tempos de vida de combustão de hidrogénio estas estrelas com
massas próximas dos extremos inferior e superior da sequência principal.
Considere que a estrela de 0.072MS é totalmente convectiva de modo que
tem à sua disposição todo o hidrogénio para queimar e não só os 10% mais
interiores.
R:
∆E ∆mc2
tnuclear = =
L L
. Para a estrela menor,
1 × 0.007 × 0.072 × MS c2
t1 = = 4.66 × 1021 s = 1.477 × 1014 yr
10−4.3 LS
Para a estrela maior,
0.1 × 0.007 × 85 × MS c2
t2 = = 8.61798 × 105 yr
10−6.006 LS
b) Calcule os raios de ambas as estrelas. Qual a razão entre os raios?
R:
r
2 4 L
L = 4πR σT ⇔ R =
4πσT 4
Então, s
10−4.3
R1 = = 5.71 × 107
4πσ(103.23 )4
s
106.006
R2 = = 9.11 × 109
4πσ(104.705 )4
Nesse caso, Led << L, pelo que a pressão de radiação não é tão relevante
para estrelas pequenas.
7.
48
5 Evolução Estelar
Ver bibliografia nos slides (muito diversificada).
49
• ZAMS: Zero-age main-sequence stars.
• M < 0.3MSol : convective cores, because high surface opacity drives sur-
face convection zones deep into the interior, making the entire star con-
vective.
Where
– µenv = mean mol. weight of overlying envelope
– µic = mean mol. weight of isothermal core
• The mass of an isothermal core may exceed the Schönberg-Chandrasekhar
Limit if an additional source of pressure can be found to supplement the
ideal gas pressure. This can occur if the electrons in the gas start to
become degenerate.
50
Figure 15: Evolution of M = 1MSol star.
P = K1 ρ4/3 relativista
51
– Mass of the isothermal He core becomes too great ⇒ core is no longer
capable of supporting the material above it (Schönberg-Chandrasekhar
Limit)
– Core collapses on a Kelvin-Helmholtz timescale. This corresponds to
(4) on Figure 14.
– Red Giant Branch (RGB):
– A convection zone develops near the surface, gradually going deeper
into the star.
– First Dredge-up: convection zone reaches the nucleus, mixing its el-
ements with the ones above and bringing them to the surface.
– At the tip of RGB, the temperature and pressure on the nucleus
finally allow He-burning (3α)
– Since the nucleus is electron degenerate, the temperature increase
does not lead to expansion and subsequent cooling, + the 3α effect
is highly temperature-dependent ⇒ explosive power!
– Helium Flash: energy released from new He-burning
– The He flash only lasts a few seconds ⇒ tough to model ⇒ dotted
line afterwards
– Structure: He-burning nucleus, H-burning shell, non-burning enve-
lope
– Early Assymptotic Giant Branch:
– He-burning leaves CO non-burning nucleus
– A convection zone develops near the surface, gradually going deeper
into the star (again)
– Second Dredge-up: convection zone reaches the nucleus, mixing its
elements with the ones above and bringing them to the surface.
– Thermal Pulse Assymptotic Giant Branch (TP-AGB):
– He-shell flashes: He-burning shell turns on/off, pulsing
– Planetary Nebula Formation (PN): star makes a planetary neb-
ula :)
• Evolution of an Intermediate-Mass (M = 5MSol ) M-S Star:
Same thing as M = 1 but with a few differences:
– He nucleus not degenerate ⇒ allows for gas expansion and cooling
⇒ no He flash.
– There’s a third dredge-up still.
– Horizontal branch.
– SGB more horizontal.
• Evolution of Massive Main-Sequence Stars:
52
– Star reaches ZAMS (enters the MS)
– pp chain: H → He ⇒ core density increases
– When H is depleted, the core contracts
– Contraction heats nucleus, which starts burning He before the nu-
cleus can even become degenerate.
– Contraction heats shell above, which starts burning H.
– When He is depleted, the temperature is really high ⇒ C-burning
and O-burning.
– Structure: CO-burning core, He-burning shell, H-burning shell, non-
burning envelope.
– As heavier elements are built up, the core develops more ”X-burning”
layers.
– Core luminosity increasing ⇒ radius increasing (outer layers) ⇒ Red
Supergiant
– Structure: Fe nucleus, Si-burning, O-burning, Ne-burning, C-burning,
He-burning, H-burning
– If the star can’t burn the Fe, the Fe nucleus simply becomes degen-
erate.
– Mass of the degenerate Fe core becomes too great ⇒ core is no longer
capable of supporting the material above it (Schönberg-Chandrasekhar
Limit)
– Core collapses ⇒ Fe becomes just protons, neutrons and electrons
– Protons combine with electrons ⇒ more neutrons
– Eventually we have a neutron star... on the nucleus
– Outer layers collapse into neutron nucleus: SUPERNOVA type-
II!
53
– Compressing the star causes P and T to increase.
– T increases ⇒ opacity decreases.
– Less opacity means more radiation flows out, reducing the tendency
to overshoot.
– Thus, the opacity quenches the oscillations.
54
• For the Cepheids, there is a relationship between the star’s period and its
mean apparent magnitude.
• Since they’re all at the same distance within their own galaxies, this means
there is a relationship between the star’s period and its mean absolute
magnitude ⇒ we get M from T .
55
5.3 Close Binaries
• ≈ half of all stars are in binary systems
• Close Binary System: If two stars in a binary system are too close to
each other, and thus they can alter the other’s structure with tidal effects.
• The star’s energy comes from its translational and rotational motion.
• As the star rotates, different layers of material rub against each other, in
a fluid friction, making the star dissipate some of the rotational energy.
• Eventually, the star’s will orbit each other with one side facing the other
forever (like the Moon)
• As the star overshoots past the other in its translational motion, it gets
attracted by gravity and comes back. The same happens to the other star.
Eventually, the stars orbits’ will circularise.
• Circular orbits and stars facing each other = lowest energy arrangement.
We say that the spins are synchronised.
• ...
• Now suppose we set the origin of our coord. system to be the CM of the
binary system, and also make our coord. system rotate with the same
period as the binary system.
• This will ”fix” things in place, but in turn add a ”centrifugal force” com-
ponent to each particle... (J 2 /2mr2 )
• Effective Potential: the gravitational potential + this new centrifugal
potential
• We can draw surfaces with contour lines describing the effective potentials
of the particles, as in Figure 18.
• The effective potential of any particle in a given surface is perpendicular
to the surface.
– ... just like in Earth contour maps, the gravitational force in each
particle on the ground is perpendicular to it.
• Lagrange Points: points where the effective potential is zero.
• Any equipotential surface of a fluid must be of constant pressure.
– If they were not, there would be pressure differences forcing fluid
along the surface, and these forces would not be balanced by any
gravitational forces, which must be perpendicular to the surface.
• Any equipotential surface of a fluid must therefore also be of constant
density.
56
Figure 18: Surfaces of constant effective potential.
• Roche lobe: one of the lobes in the ”number 8” orbit that crosses the
equipotential surfaces plane.
• Three classes of close binary systems:
– detached: each star totally contained within its own Roche lobe
– semidetached: the photosphere of one star exactly fills its side of
the Roche lobe.
– contact binaries: both stars are at/over the Roche lobe.
• If the binaries are detached, then there is no mass transfer between them
and they don’t alter each other’s evolution ⇒ we can study their evolution
as isolated stars.
• If the binaries are not detached, then there is mass transfer and their
evolution is different.
• The more massive star evolves off the MS first. When it becomes a red
giant, it may become large enough to fill its Roche lobe. Thus, mass begins
to be transferred from the larger star to the smaller star. The transfer
stops once both masses are equal.
• At some point, the star that was losing mass will collapse and become
one of the three: white dwarf (low-intermediate mass stars), neutron star
(massive stars) and black hole (big boys).
•
57
Figure 19: Surfaces of constant effective potential.
58
Figure 20: Classes of binary star systems.
59
Figure 21: Lines of sight through a planetary nebula. (a) Appearance.The
shaded regions represent the places where the lines of sight pass though the
shell.The line of sight near the edge passes through more material than that
through the center.This is responsible for the ringlike appearance. (b) Doppler
shifts. Material on the near side is moving toward the observer, producing a
blueshift, and material on the far side is moving away, producing a redshift.
5.6 Problemas
:
1.
L
Ṁ = −4 × 10−13 η M⊙
gR
a) Explique qualitativamente porque entram L, g e R na equação da
maneira que entram.
R:
L é uma medida de output de energia, logo é diretamente proporcional à
60
taxa de perda de massa. g é inversamente proporcional a Ṁ porque mais
gravidade = matéria mais presa à estrela. R é inversamente proporcinal a
Ṁ pois se a estrela é maior, há menos fluxo radiativo escapando a estrela
pois tem de atravessar uma distância maior.
Finalmente,
Ṁ = 8.68 × 10−7 M⊙ /ano
g = GM/R2
Logo,
2 2
g GM R⊙ M R⊙
= 2 =
g⊙ R GM⊙ M⊙ R
Logo em unidades solares temos
M
g=
R2
Substituindo na expressão encontra-se o desejado.
61
R:
dM LR
Ṁ = = −4 × 10−13 η
dt M
dM
M = −4 × 10−13 ηLR = C (constante)
dt
Z t Z t
dM
M dt = Cdt
0 dt 0
Z t Z t
M dM = C dt
0 0
2
M t
| = Ct
2 0
M 2 (t) − M 2 (0) = 2Ct
q
M = 2Ct + M02
q
M = −8 × 10−13 ηLR + M02
M 2 (t) − M02
q
M= 2Ct + M02 ⇔ t =
2C
t = 368663.6 anos
GM 2
tKH =
2RL
Tem de pegar Te e L do diagrama (entre os pontos 4 e 5, no meio da
linha). Com L = 4πR2 σTe achamos R, e podemos calcular tKH
3. p′ = 16′ , d = 213 pc.
a) Calcule o diâmetro fı́sico da nebulosa.
R:
1
d= lAU ⇔ lAU = dp′′
p′′
62
p′′ = 16 × 60′′ = 960′′
lAU = 204480AU = 0.911pc
• That is, E >> N kT overall and Ei >> kT on average, which means elec-
trons are in higher levels Ei than we would guess by just considering the
thermal energy kT available for them. They have much greater energies
than they would in an ordinary gas. These high energy electrons also have
high momenta. They can therefore exert a pressure considerably in excess
of the pressure exerted by an ideal gas at the same temperature.
• The higher pressure is called degeneracy pressure. It is responsible for the
hardness of metals.
• We can also describe this pressure in terms of the uncertainty principle:
∆p∆x ≥ h/4π
• When the density becomes very high, we are confining them to a small
∆x, which implies a large ∆p. Higher uncertainty in momentum means
large momenta are possible, and the electrons with these momenta exert
the degeneracy pressure.
63
• Fermi energy: energy for which T = 0K (zero absoluto) ⇒ total degen-
eracy
h̄2
ϵF = (3π 2 n)2/3
2m
• Vimos antes que:
– Estrelas de massa baixa e intermédia acabariam gerando uma
nebulosa planetária e deixando para trás uma anã branca.
– Estrelas de massa elevada acabariam como supernovas, deixando
para trás uma estrela de neutrões (ou um buraco negro).
– Tanto as anãs brancas como as estrelas de neutrões são degener-
adas.
64
• Para o Sol T
ρ2/3
= 5500Km2 kg −2/3 > D
L = CTc7/2
4
Em que C é uma constante. Note-se que L também é proporcional a Tef f
65
• A energia disponı́vel para radiação é
MW D 3
U= kTc
AmH 2
Onde A = Z + N do elemento, mH = massa do núcleo de H. Como o
elemento é carbono, AmH = mC = 12u = massa do núcleo de carbono.
Logo M mC é o número total de núcleos na anã branca. Multiplicando
wd
pela energia cinética média 23 kTc , temos a energia total disponı́vel para
radiação.
• Tempo de arrefecimento da anã branca (estimativa furreba):
U 3 Mwd k
τcool = =
Lwd 2 AmH CTc5/2
66
• A figura 6.2 mostra a evolução da estrutura da estrela com o aumento da
densidade, e a figura 6.2 mostra um modelo de uma estrela de neutrons
tı́pica (M = MCh ≈ 1.4M⊙ ).
• Although the details are sensitive to the equation of state used, this model
displays some typical features.
1. Outer crust: consists of heavy nuclei, in the form of either a fluid
”ocean” or a solid lattice, and relativistic degenerate electrons.
2. Inner crust: consists of a lattice of nuclei, a superfluid of free neu-
trons, and relativistic degenerate electrons. The bottom of the inner
crust occurs where ρ ≈ ρnuc ⇒ and the nuclei dissolve.
3. Interior: consists primarily of superfluid neutrons, with a smaller
number of superfluid, superconducting protons and relativistic de-
generate electrons.
4. Core: ???? ninguém sabe
• Superfluid = flows with no viscosity; superconducting = transfers energy
with no resistance.
• Rotation of neutron stars: when a core collapses to become a neutron
star, the size shrinks (r down), which means that for angular momentum
L = mvr to be conserved, it has to spin faster (v up).
• Strong magnetic field (complicated reason)
• Extremamente quentes. T ≈ 1011 K no inı́cio, chegando aos T ≈ 106 K
que terá durante umas dezenas de milhares de anos. Para esta T ≈ 106
temos luminosidade comparável à do Sol.
67
• O comprimento de onda máximo no espectro de corpo negro é
λmax ≈ 2.9nm (Raios-X)
6.3 Pulsares
• Pulsars are neutron stars spinning. On the surface of the neutron star,
there is some emission point (or ”hot spot”) producing a beam of radiation,
like a lighthouse beam. We can only see the radiation when the beam is
pointing towards us. This creates a cycle in which we detect a ”pulse” of
EM radiation every pulse cycle P .
68
2.
3.
4. O que é uma binária de contacto?
R:
É um sistema de duas estrelas que orbitam uma a outra, sendo que uma
delas preenche seu lobos de Roche a ponto de transferir massa para a
outra.
5.
6.
6.5 Problemas
1. Assumindo (incorretamente) que as oscilações de δ Cephei são sinusoidais,
com uma amplitude de velocidade de ≈ 20km/s e um perı́odo de 5d8h48m,
determine o aumento máximo que o seu raio pode ter (com referência ao
raio de equilı́brio).
R:
2π
R(t) = R0 + A sin t
Π
dR 2π 2π
V (t) = = A cos t
dt Π Π
2πA
Ou seja, a amplitude da velocidade é Π , que é 20km/s. Então
Π(s)
A = 20(km/s) = 1.476 × 106 km
2π
R = R0 + 1.476 × 106 km
∆K = −∆U
1 2 GmM⊙
0 − mvesc = 2
2 Rwd
r
2GM⊙
vesc =
Rwd
Uma anã branca de 1M⊙ tem raio aproximado de 7 × 106 m, logo
69
3. Qual é a energia térmica armazenada por uma anã branca de 1M⊙ e
T = 107K? Assumindo que radia como um corpo negro de T = 104K,
qual seria a sua escala do seu tempo de vida enquanto objecto luminoso?
R:
Energia disponı́vel para radiação:
Mwd 3 1.989 × 1030 3
U= kT = 1.381 × 10−23 × 107K = 2.068 × 1040 J
mC 2 1.993 × 10−26 2
Taxa de emissão dessa energia:
2
L = 4πRwd σT 4 = 4π(7 × 106 )2 σ(104)4 = 3.49 × 1023 W
Logo
τ = U/L = 1.8764 × 109 yr
70
7 Formação Estelar
7.1 Cenário
• As estrelas se formam a partir do colapso gravitacional de nuvens de gás
interestellar
• A Via Láctea tem uma grande quantidade de material para formar estrelas
71
• The Jeans mass derivation above neglected the important fact that there
must exist an external pressure on the cloud due to the surrounding in-
terstellar medium.
• Bonnor-Ebert mass: critical mass required for gravitational collapse in
the presence of an external gas pressure of P0 .
cBE vT4
MBE = 1/2
P0 G3/2
p
Where vT = kT /µmH = isothermal sound speed (γ = 1), and
dimensionless constant cBE ≈ 1.18.
• Making the simplifying (and probably unrealistic) assumption that any
existing pressure gradients are too small to influence the motion appre-
ciably ( dP
dr small), then the cloud is essentially in free-fall (and thus T is
constant) during the first part of its evolution.
d2 r Mc ρ0 dP d2 r Mc
ρ0 2
= −G 2 − ⇒ 2 = −G 2
dt r dr dt r
72
• At the isothermal extreme, if the energy that is released during a gravi-
tational collapse is radiated away efficiently, the temperature can remain
nearly constant. At the adiabatic extreme, if the energy cannot be trans-
ported out of the cloud at all, then the temperature must rise.
• Adiabatic collapse: from thermo,
T = K ′′ ργ−1
MJ ∝ ρ1/2
I.e. the Jeans mass (minimum mass for collapse) increases with increasing
density for a perfectly adiabatic collapse of a cloud. The result is a mini-
mum mass for the fragments produced. We can make a crude estimate:
– Initially, the early stages of the free-fall collapse are nearly isothermal
because light near the center of the collapse can travel significant
distances before being absorved by any dust.
– Owing to an initial slight increase in density toward the center of
the cloud, the free-fall timescale is shorter near the center and the
density increases more rapidly there (inside-out).
– When the density of the material near the center of the collapse
region reaches ρ ∼ 10−10 , the region becomes optically thick and the
collapse becomes more adiabatic.
73
– The increased pressure of heat not being able to escape when the
collapse becomes adiabatic substantially slows the rate of collapse
near the core. At this point the central region is nearly in hydrostatic
equilibrium (r ∼ 5AU ). This central object is the protostar.
– When the infalling material meets the nearly hydrostatic core, a
shock wave develops where the speed of the material exceeds the
local sound speed. The infalling material loses a lot of its K to give
T to the core, increasing its L.
– When T ∼ 1000K, the dust within the developing protostar begins
to vaporise and the opacity drops. This means that the radius R
where τ = 2/3 is reduced, approaching the surface of the core. Since
L = 4πR2 σTe4 and L remains high in this phase, we know Te must
increase to compensate for the loss of R.
– When T ∼ 2000K, molecular hydrogen begins to dissociate into in-
dividual atoms. To separate H2 , energy is absorbed which would
otherwise provide pressure to maintain hydrostatic equilibrium. As
a result, the core becomes unstable and a second collapse occurs.
At this point, the core mass is still much less than its final value,
implying accretion is still ongoing.
– After the second core collapse, a second shock wave happens as the
envelope continues to accrete falling material.
– We reach the flat part of Figure 7.2, where the temperature is high
enough such that deuterium (21 H) begins to burn, producing up to
60% of the luminosity of the 1M⊙ protostar.
– When the deuterium is depleted, the nuclear reactions stop and the
luminosity decreases, leading to the sharp turn downwards in Figure
7.2.
– Now we change timescales from free-fall to nuclear reactions: tf f →
tKH .
– Extra electrons coming from the ionisation of heavier elements in the
gas go to H ⇒ the opacity of the outer layers becomes dominated
by the H − ion. This large opacity causes the envelope to become
convective.
– Hayashi Track: track in HR diagram that happens because of the
constraints that convection puts on the structure of a star. To the
right of the track, no stable stars can exist; To the left of the track,
stable stars exist.
∗ With the high H − opacity on the surface, the star is completely
convective in the beginning. Deuterium burning stars on the
black square of Figure 7.2.
∗ As the central temperature continues to rise (star still contract-
ing), increasing levels of ionisation decrease the opacity in that
74
region and a radiative core develops, progressively encompassing
more of the star.
∗ At the point of minimum luminosity following the descent in the
track, the radiative core allows energy to escape into the con-
vective envelope more readily, causing the luminosity to increase
again. Since the star is still shrinking, T is still rising.
∗ The temperature is high enough for the first two steps of the
PP-I chain (H → He) and CNO cycle reaction C → N to begin.
∗ At the point of maximum luminosity after the minimum, the
rate of nuclear energy production has become so great that the
central core is forced to expand.
∗ When the 126 C is finally exhausted, the core reaches a sufficiently
high temperature for the rest of the PP-I chain to become im-
portant.
• Role of angular momentum in disc formation: figure 22
75
Figure 22: Fragmentation of a collapsing interstellar cloud. a) The cloud is ini-
tially rotating as shown. As it collapses, the angular momentum J is conserved.
b) As the cloud becomes smaller, its angular speed ω must increase to keep the
angular momentum fixed. The rotation inhibits collapse perpendicular to the
axis of rotation, and the cloud flattens (massa de pizza). c) Unable to collapse
any further, the cloud breaks up, with the angular momentum being divided
between the spin and orbital angular momenta of the individual fragments.
R:
Se forem mais densas, é mais provável para a força gravitacional causar
colapso, o que não pode ser dito para uma nuvem muito dispersa. Além
disso, se a nuvem estiver mais fria, as partı́culas se movem com menos
velocidade, permitindo novamente que a força gravitacional faça o seu
trabalho.
2. À medida que uma nuvem colapsa, o que acontece à massa de Jeans dos
fragmentos?
R:
Como MJ ∝ ρ−1/2 , a medida que a nuvem colapsa a densidade aumenta,
levando a uma diminuição da massa de Jeans. Durante o colapso, qual-
quer inhomogeneidade inicial na densidade causará partes diferentes da
nuvem a satisfazerem Mc > MJ separadamente, causando colapso indi-
vidual destas partes (fragmentação).
76
3. Como pode a formação de vários sistemas estelares facilitar o processo de
formação estelar?
R:
Estrelas já existentes podem ajudar a fornecer matéria/energia para a
formação de novas estrelas, por meio de ventos solares e encontros próximos
entre estrelas e nuvens. Ambos os processos podem iniciar colapsos grav-
itacionais em nuvens, causando formação estelar.
4. Como pode uma nuvem gravitacionalmente ligada dar origem a uma as-
sociação de estrelas não ligadas gravitacionalmente?
R:
Por meio da fragmentação, a nuvem pode gerar múltiplas estrelas, cujo
movimento pode levar a um subsequente afastamento tal que as estrelas
não ficam mais gravitacionalmente ligadas.
5. Recordando os resultados para o equilı́brio hidrostárico, (...)
R:
7.4 Problemas
1. Mostre que a massa de Jeans também se pode escrever como:
cJ vT4
MJ = 1/2
P0 G3/2
77
expressão para o tempo de queda livre. Mostre que difere por um termo
da ordem da unidade relativamente a expressão dada na aula:
1/2
3π 1
tf f =
32 Gρ0
R:
A aceleração é constante durante todo o colapso, então consideramos o
raio e massa no momento do colapso:
GMJ
a=
RJ2
Seja tq o tempo de queda:
Z tq Z tq
dr dr
= at ⇔ dt = a tdt
dt 0 dt 0
1 GMJ 2
r(tq ) = RJ = t
2 RJ2 q
s
2RJ3
tq =
GMJ
Assumindo uma nuvem esférica,
MJ 3MJ 3MJ
ρ0 = = ⇔ RJ3 =
V 4πRJ3 4πρ0
Substituindo na expressão para tq ,
r
3
tq =
2πGρ0
Comparando com a expressão da aula,
tq 4
= ≈ 1.27
tf f π
3. (...)
a) (...)
R:
d2 r dvr GMr
= vr = − 2 + rω 2
dt2 dr r
L0 = Lf ⇔ I0 ω0 = Iω
Assumindo que a nuvem é uma esfera, I = 52 M r2 , i.e.
r 2
0
r02 ω0 = r2 ω ⇔ ω = ω0
r
78
Substituindo,
Z f Z f
GMr
vr dvr = − + rω 2 dr
i i r2
1 2 GMr ω2 r4
(vf − v02 ) = − 0 20
2 rf 2rf
Mas v0 = 0 pelo enunciado, e vf = 0 pois é quando o colapso acaba. Então
GMr ω2 r4
= 0 20
rf 2rf
Isolando rf ,
ω 2 r04
rf =
2GMr
b) (...)
R:
c (...)
R:
v0 = r0 ω0 = 4.1m/s
d (...)
R:
2 1
Li = Lf ⇔ M r02 ω0 = M r2 ω 2
5 2
ω = 2.253 × 10−10 rad/s
e) (...)
R:
2π
∆t = = 883 anos
ω
r
4π 2 a3
∆tKep = = 1000 anos
GM
79