1 1.

Virial Theorem

Last semester, we derived the Virial theorem from essentially considering a series of particles which attract each other through gravitation. The result was that d2 I = 2K + dt2

(1)

where the Virial I = r 2 dm, is the gravitational potential energy and K is the total 2I kinetic energy. In the case of equilibrium where d = 0, we nd that: dt2 1 K= 2 (2)

which is very useful when considering stars in hydrostatic equilibrium. Unlike stars, for molecular clouds, there may be a signicant external pressure. The virial theorem can be modied to take into account external pressure. In molecular cloud studies, a common way of writing the virial theorem is: 1 d2 I =3 2 dt2

P dV +
V S

P r dS +

(3)

where 3 V P dV is the integral of pressure over the volume of the cloud and is equal to 2K , P r dS is the integral of the external pressure over the surface of the cloud, and once S again the gravitational potential energy. We dene Ks = replace:
1 2 S

P r dS. If the external pressure is a constant value P0 , then we can

1 Ks = P 0 2 and we can write: 1 d2 I =3 2 dt2 or if we assume a constant pressure

3 r d S = P0 V 2 S

(4)

(P P0 )dV +
V

(5)

1 d2 I = 3(P P0 )Vcl + 2 dt2

(6)

2 2. The Stability of a Cloud (or Clump)

Following Spitzer (1978), lets assume that a cloud is in equilibrium, i.e. d2 I/dt2 = 0. The we rewrite equation 5 so that: 3 P0 V = 3
V

P dV +

(7)

assuming a homogenous sphere with sound speed cs

3 4P0 Rcl

3 c2 s Mcl

2 3 GMcl 5 Rcl

(8)

where cs is the sound speed and the ideal gas law gives P = c2 s . To examine the stability of the cloud to changes in external pressure, we solve the above equation for P0 and then take the derivative with R P0 = and 1 dP0 = dRcl 4
2 9 c2 12 GMcl s Mcl + 4 5 Rcl 5 Rcl

1 4

2 3 c2 3 GMcl s Mcl 3 4 Rcl 5 Rcl

(9)

(10)

Now lets consider what happens if we increase pressure. If the right side of the equation is negative, i.e. R > Rcrit = 4 GMcl 15 c2 s (11)

then the increase in pressure will cause the cloud to shrink, but the value of P0 required for equilibrium goes up as well. Eventually, the cloud comes into equilibrium, unless R Rcrit . At this point, dP0 /dRcl = 0 and P0 is at a maximum. Increasing P0 will imply that there is no stable equilibrium and the cloud will collapse. For a value of M = 104 M , TK = 20 K (and consequently cs = 0.25 km s1 ), Rcrit 200 pc. This is larger than most clouds, molecular clouds cannot be supported by thermal pressure. We also note that = |/K | = 6, meaning that the cloud would be not in virial equilibrium.

3 3. The Eect of the Magnetic Field

Magnetic elds may have a stabilizing eect on clouds. Although the clouds are neutral, ionization by cosmic rays, and on the outer layers, UV radiation, may create a small number of electrons and ions. These would be coupled to the magnetic eld, and in fact, freeze the magnetic eld lines to the gas. Thus, the magnetic eld pressure could exert a pressure. 1 d2 I =3 2 dt2 where 1 MB = 8 where B0 is the external eld. 1 MB = 8 1 2 3 1 2 (B )dV B Rcl = 2 6 6 R
2 2 (B 2 B0 )dV

(P P0 )dV + + MB
V

(12)

(13)

(14)

is the mean magnetic eld. where b is order unity and B = 2rBdr R2 B (15)

An important value is the ratio of Mcl / where the graviational potential energy is equal to the magentic energy term, = MB . For a uniform sphere, this occurs when:
2 3 GMcl 1 2 = 5 R 6R

(16)

or when Mcl = 5 18
1/2

1 1 = 1 / 2 G 2G1/2

(17)

The ratio Mcl / is known as the mass to ux ratio. If 1 Mcl > 2G1/2 (18)

4 then gravity dominates over magnetic elds and the cloud is supercritical. Note, that since is conserved, the ratio does not change as the cloud collapse. If the ratio is < 1, then the cloud is sub-critical. Note that since N (H 2 ) = implies that Mcl N (H 2 ) = B Measurements show that against collapse.
N (H2 ) B

Mcl , B= 2 2 Rcl Rcl

(19)

(20)

1, suggesting that magnetic elds do not support clouds

4.

Turbulent Pressure

We now consider the addition of turbulent energy to the kinetic energy term. If we just consider the thermal motions of the gas, the kinetic energy is given by: 3 K = Mcl c2 s 2 where cs is the sound speed. Remember that the total 3-D velocity dispersion is
2 2 2 2 = x + y + z = 3 c2 s

(21)

(22)

We can then dene a pressure term: 3 2 P = Mcl (c2 s + turb ) 2 (23)

where turb is the 1-D turbulent dispersion. We can take this from the measured linewidth 2 2 of the cloud, which is tot = c2 s + turb . 3 2 Ptot = Pth + Pturb = Mcl (tot ) 2 Then the equation becomes: (24)

5 For a value of M = 104 M , TK = 20 K (and consequently cs = 0.25 km s1 ), Rcrit 12 pc. This is similar to the size of most molecular clouds. Plus we know that = |/K | is around 1, indicating that it is close to virial equilibrium.