You are on page 1of 29

Astronomy, Astrophysics, and Cosmology

Luis A. Anchordoqui

Department of Physics and Astronomy


Lehman College, City University of New York

Lesson III
February 23, 2016

arXiv:0706.1988
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 1 / 29
Table of Contents

1 Stellar Evolution
Nucleosynthesis
White dwarfs and the Chandrasekhar limit
Supernovae

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 2 / 29


Stellar Evolution Nucleosynthesis

In previous class we have seen that...

G Star form when gaseous (mostly 1 H) clouds contract due to pull of gravity
G Energy releasy in 1 H fusion reactions
Literature: Longair: §8, 10, 12 & 13 HEA 09/ pag
produces outward pressure to halt inward gravitational contraction
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 3 / 29
Stellar Evolution Nucleosynthesis

As hydrogen fuses to form helium @ star’s core


helium formed is denser and tends to accumulate in central core
As core of helium grows
hydrogen continues to fuse in a shell around it
When much of hydrogen within core has been consumed
production of energy decreases at center and . . .
cannot prevent gravitational force to contract and heat up core
Hydrogen in shell around core fuses more fiercely
as T rises causing outer envelope to expand and cool
Surface T reduces + spectrum peaks at longer wavelength
(reddish)
By this time the star has left the main sequence:
It has become redder
It has grown in size
It has become more luminous
It enters red giant stage
Model explains origin of red giants as step in stellar evolution
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 4 / 29
Evolution of Sun-like
Stellar Evolution NucleosynthesisStars

Example
Sun has been on main sequence for ∼ four and a half billion years
olutionIt will
of Sun-like Stars
probably remain there another 4 or 5 billion years
As becomes red giant expected to grow out to Mercury’s orbit

E. Chaisson, S. McMillan “Astronomy Today”

E. Chaisson, S. McMillan “Astronomy Today”

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 5 / 29


Stellar Evolution Nucleosynthesis

If star is like our Sun or larger + further fusion can occur


As star’s outer envelope expands + core shrinks and heats up
When the temperature reaches about 108 K
helium nuclei reach each other and undergo fusion
Reactions are

Two reactions must occur in quick succession


because 84 Be is very unstable
Net energy release of the triple-α process is 7.273 MeV
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 6 / 29
Stellar Evolution Nucleosynthesis

Further fusion reactions are possible + 42He fusing with 126C to form 168O.
In very massive stars + higher Z elements ( e.g. 20Ne or 24
10 12Mg ) can be made

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 7 / 29


Stellar Evolution Nucleosynthesis

Red supegiants
As massive red supergiants age
stars produce “onion layers” of heavier elements in their interiors

@ T = 5 × 109 K nuclei as heavy as 56 56


26 Fe and 28 Ni can be made

Average binding energy per nucleon


begins to decrease beyond iron group of isotopes

Formation of heavy nuclei by fusion ends at iron group

As a consequence
core of iron builds up in centers of massive supergiants

Process of creating heavier nuclei from lighter ones


or by absorption of neutrons at higher Z
is called nucleosynthesis
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 8 / 29
Stellar Evolution Nucleosynthesis

Red Supergiant

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 9 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

Sirius
Sirius @ 2.6 pc + fifth closest stellar system to Sun

Analyzing motions of Sirius Bessel concluded


it had an unseen companion with an orbital period T ∼ 50 yr

In 1862 + Clark discovered this companion + Sirius B

Following-up observations showed that for Sirius B M ≈ M

Sirius B’s peculiar properties were not established until 1915

Adams noted high temperature of Sirius B + T ' 25, 000 K


which together with its small luminosity + L = 3.84 × 1026 W
requires extremely small radius and thus large density of this star

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 10 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 11 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

From Stefan-Boltzmann law we have


   
R L 1/2 T 2
= ≈ 10−2 (1)
R L T
Mean density of Sirius B + ρ = 3 × 106 g/cm3
Lower limit for central pressure of a star in hydrostatic equilibrium
M2
Pc > = 4 × 1016 bar (2)
8πR4
What would be central temperature Tc needed
if pressure is dominated by ideal gas?
From ideal gas law
Pc
Tc = ∼ 102 Tc, ≈ 109 K (3)
nk
For such high central temperature
dT/dr in Sirius B would be a factor 104 larger than in Sun
This would in turn require larger luminosity L( R)
and larger energy production rate than in main sequence star
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 12 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

Stars like Sirius B are called white dwarfs

They have very long cooling times


because of their small surface luminosity

White dwarfs are numerous + mass density in solar neighborhood


main-sequence stars + 0.04M /pc3
white dwarfs + 0.015M /pc3

Typical mass in range 0.4 − 1M + peaking @ 0.6M

For white dwarfs + no further fusion energy can be obtained

White dwarf continues to lose internal energy by radiation


decreasing in T and becoming dimmer until its light goes out

Star has then become cold dark chunk of ash


L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 13 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

For a classical gas + P = nkT


+ in limit of zero T also P inside star goes to zero

How can star be stabilized after fusion processes


and thus energy production stopped?

Solution to this puzzle:


+ main source of P in such compact stars has different origin

Pauli principle forbids fermions to occupy same quantum state

In statistical mechanics
Heisenbergs uncertainty principle + ∆x∆p ≥ }
together with Pauli’s principle
imply that each phase-space volume }−1 dx dp
can be occupied by only one fermionic state
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 14 / 29
make the following as- Stellar Evolution White dwarfs and the Chandrasekhar limit

A (relativistic or non-relativistic) particle in box of volume d3


averagecollides
separation
per timeAssumptions
interval ∆t of
= the
d/vmolecular
x once with yz-side of box
This means that the model of an ideal gas
Thereby it exerts Fx = ∆p x /∆t = p x v x /d
is consistent with the
Pressure produced by N particles
nt-like.
P = F/A = N p x v x /(dA) = np x v x
hole theyFormove ran- distribution with hv2 i = hv2 i + hv2 i + hv2 i = 3hv2 i
isotropic x y z x
in any direction with
olecules move at high 1
P= nvp (4)
3
ng elastic collisions. y
ecules exert no long-

vi

s, all molecules are


d m
vxi
toms, we can assume
ases rather well at low z x
the average, on the d d
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 15 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

If we use ∆x = n−1/3 and ∆p ≈ }/∆x ≈ }n1/3


together with v = p/m (valid for non-relativistic particles)

}2 n5/3
P ≈ nvp ≈ ⇒ P ∝ ρ5/3 (5)
m
For relativistic particles + v = c

P ≈ ncp ≈ c}n4/3 ⇒ P ∝ ρ4/3 (6)


Note the following important points:
Both pressure laws are polytropic equations of state + P = Kργ
Non-relativistic degenerate Fermi gas
has the same adiabatic index as an ideal gas + γ = 5/3
Relativistic degenerate Fermi gas
has the same adiabatic index as radiation + γ = 4/3
In non-relativistic limit + P ∝ 1/m
degeneracy will become important first to electrons
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 16 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

Pressure of degenerate non-relativistic electron gas

To good approximation + e− in Sirius B are non-relativistic


ρ
ne = (7)
µe m p

µe ≡ A/Z + average number of nucleon per free electron


For metal-poor stars + µe = 2
!5/3
h2 n5/3
e (1.05 × 1027 erg s)2 106 g/cm3
P ≈ ≈
me 9.11 × 10−28 g 2 × 1.67 × 10−24 g
≈ 1023 dyn/cm2 = 1017 bar (8)

Consistent with lower limit for central pressure of Sirius B

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 17 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

G Relate mass of star to its radius by combining

GM2
Pc ∼ (9)
R4
and  5/3
5/3 M KM5/3
P = Kρ ∼K = (10)
R3 R5
we have
GM2 KM5/3
= (11)
R4 R5
or
M(10−12)/6 1
R= = (12)
K KM1/3
G If small differences in chemical composition + neglected
there is unique relation between mass and radius
G Radius of white dwarf stars decreases for increasing masses
suggesting that there exists maximal mass
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 18 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

Assume pressure described by non-relativistic degenerate Fermi gas

Total kinetic energy Ukin = N p2 /(2me ) + n ∼ N/R3 and p ∼ }n1/3

}2 n2/3 }2 N (3+2)/3 }2 N 5/3


Ukin ∼ N ∼ = (13)
2me 2me R2 2me R2

Potentail gravitational energy approximated by Upot ∼ GM2 /R

Balance equation

}2 N 5/3 GM2
U ( R) = Ukin + Upot ∼ 2
− (14)
2mR R

For small R + positive term dominates


and so stable minimum Rmin exists for each M

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 19 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

If Fermi gas in star becomes relativistic + Ukin = Ncp

c} N 4/3
Ukin ∼ Nc}n1/3 ∼ (15)
R
and
c} N 4/3 GM2
U ( R) = Ukin + Upot ∼ − (16)
R R

Both terms scale like 1/R

For fixed chemical composition + ratio N/M remains constant

If M is increased + negative term increases faster than first one

If U becomes negative
and can be made arbitrary small by decreasing R
star collapses
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 20 / 29
Stellar Evolution White dwarfs and the Chandrasekhar limit

Critical mass for U = 0 + Chandrasekhar mass MCh


Using M = NN m N + c} Nmax4/3 = GN 2 m2
max N
!3/2  3
c} MPl
Nmax ∼ ∼ ∼ 2 × 1057 . (17)
Gm2p mp

This leads to + MCh = Nmax m p ∼ 1.5M


Critical size determined by two conditions:
gas becomes relativistic Ukin . Nme c2 and N = Nmax
4/3
 1/2
2 c} Nmax c} c}
Nmax me c & ⇒ m e c2 & (18)
R R Gm2N

This corresponds to radii found for white dwarf stars


 
} c} 1/2
R& ∼ 5 × 108 cm (19)
me c Gm2N

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 21 / 29


Stellar Evolution White dwarfs and the Chandrasekhar limit

Life cycle of the Sun

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 22 / 29


Stellar Evolution Supernovae

Supernovae are massive explosions


Thermonuclear Supernova (Type Ia)
that take place at end of star’s life cycle
They can be triggered by one of two basic mechanisms:
I by sudden re-ignition of nuclear fusion in degenerate star
II by the sudden gravitational collapse of massive star’s core

Core-Collapse Supernova (Type II, Ib, Ic)

E. Chaisson, S. McMillan “Astronomy Today”

E. Chaisson, S. McMillan “Astronomy Today”


L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 23 / 29
Stellar Evolution Supernovae

Supernova explosion

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 24 / 29


es
Stellar Evolution Supernovae

Sudden release of a large amount E into fluid of density ρ1


creates strong explosion characterized by strong shock wave
emanating from the point where the energy was released

22 Sedov andTaylor solvedbbc.co.uk


the problem of point explosion
in context of atomic bomb explosions
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 25 / 29
Stellar Evolution Supernovae

Order of magnitude estimates


Mass of swept up material + order M (t) ∼ ρ1 R3 (t)
Fluid velocity behind shock + order mean radial velocity

ush (t) ∼ R(t)/t (20)

Kinetic energy

R2 R5
Ekin ∼ Mu2sh ∼ ρ1 R3 = ρ 1 2 (21)
t2 t
What about thermal energy in bubble created by explosion?
If P is postshock pressure + thermal energy should be of order

3
Ethem ∼ PV (22)
2

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 26 / 29


Stellar Evolution Supernovae

Jumping conditions across a shock


If shock moves to right with velocity ush (t) + in shock rest-frame
background gas streams with velocity u1 = ush (t) to the left
and comes out of the shock with
higher density ρ2
higher pressure P2
lower velocity u2
For strong explosion + sound-speed of medium c1  1
Rankine-Hugonoit relations for shock tell us

P2 2γ(u1 /c1 )2 γ−1


= − (23)
P1 γ+1 γ+1

As background pressure + P1 = ρ1 c21 /γ ý in strong shock limit

2ρ1 u21
P2 ' (24)
γ+1

L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 27 / 29


where P is the postshock pressure. To find this pressure, we need
Stellar Evolution Supernovae
where the energy was released. Such explosions
conditions across a shock.occur
If thefor
shock moves to the right with
the form of supernova then in the rest-frame of the shock the background gas streams
explosions.
Blast wave + spherically symmetric
the left, and comes out of the shock with a higher density ⇢2 , h
and with a lower velocity v2 .
HEA 12: Supernova Remnants (SNR)
Today:
u2 u1
shockwaves
evolution of SNRs

Next Class:
diffuse shock acceleration in SNRs
The Rankine-Hugonoit relations for the shock tell us
SN1604 (Kepler) SNR-0509-67.5 (LMC) SN1006
⇢1 v2 1 2
ock wave travel and what is left behind? The problem = of = +
⇢2 v1 + 1 ( + 1)M2
known as Sedov-Taylor explosion,
where after the two scientists
v1
alytic (and in part numerical) means in the context of M = c
1
Today, the problem can provide
is theaMachusefulnumbertestoftothevalidate
shock. For a strong explosion, the
cal scheme, because an analytic solution
background mediumfor it can small,
is negligibly be so that the Mach number
be compared to numerical results. Also, the problemRankine-Hugonoit relation is
in this limit. For the pressure, the
to demonstrate the power of dimensional
Chandra/HST/VLA/GBT analysis and
P2 2 M2 1
=
P1 +1 +1
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 28 / 29
2
Stellar Evolution Supernovae

More about order of magnitude estimates


With postshock pressure
we estimate thermal energy in shocked bubble

R5
Etherm ∼ P2 R3 ∼ ρ1 u21 R3 ∼ ρ1 (25)
t2
Total energy
R5
E = Ekin + Etherm ∼ ρ1 (26)
t2
Shock front radius  1/5
Et2
R(t) ∼ (27)
ρ1
Expanding shock wave slows down as it expands
 1/5  1/2
2 E 2 E
ush = = R−3/2 (28)
5 ρ1 t3 5 ρ1
L. A. Anchordoqui (CUNY) Astronomy, Astrophysics, and Cosmology 2-23-2016 29 / 29

You might also like