Boiling

Boiling is considered when the temperature of the fluid film is above the boiling temperature. Below the boiling temperature, the same equilibrium conditions as for evaporation are assumed, but the vapor mass rates are determined differently.

There are two distinct choices for modeling fluid film boiling in Simcenter STAR-CCM+, namely the Rohsenow and the Habchi Boiling models.

Rohsenow Boiling

The empirical standard pool boiling correlation [632] is used to model liquid film boiling at a hot wall. The film starts to boil when the wall temperature exceeds the boiling temperature of the liquid. The heat flux passing from the wall to the film increases in proportion to the cubic power of Δ T W = T W - T sat as:

1. EQUATION_DISPLAY
q b oil = μ f Δ H vap Pr f n g ( ρ f - ρ ) σ ( C p , f Δ T W c s f Δ H vap ) 3
(2802)

where:

  • the viscosity μ f , the Prandtl number Pr f n , and the specific heat C p , f are evaluated for the liquid phase.
  • ρ f and ρ denote the densities of the liquid and gas phases.
  • g and σ denote the gravity acceleration and surface tension coefficient respectively.
  • c s f and n characterize the fluid wall interface.

    For example, for water-brass contact, the values are c s f = 0.006 and n = 3 .

  • T sat represents the saturation temperature of the mixture and is computed from the temperature-dependent saturation pressure of its components by applying Dalton’s law.

The heat flux increases as T W increases but does not exceed a critical heat flux that is:

2. EQUATION_DISPLAY
q max = c max ρ 1 / 2 ΔH vap [ g ( ρ f ρ ) σ ] 1 / 4
(2803)

where c max = 0.15 . After q max has been attained, the heat flux decreases to its minimum at the Leidenfrost temperature T L :

3. EQUATION_DISPLAY
q min = c min ρ Δ H vap [ g ( ρ f ρ ) σ ] 1 / 4 ( ρ f ) - 1 / 2
(2804)

where c min = 0.09 .

The wall temperature of maximal heat flux T S = T sat + ΔT S can be obtained from equations Eqn. (2802) and Eqn. (2803), resulting in:

4. EQUATION_DISPLAY
Δ T S = c s f c m a x 1 / 3 Pr f n / 3 σ 1 / 4 ρ 1 / 6 Δ H v a p C p , f μ f 1 / 3 g 1 / 12 ( ρ f ρ ) 1 / 12
(2805)

Based on this correlation, the heat flux from the wall is computed as follows:

5. EQUATION_DISPLAY
q W = { q W = q boil , if ΔT W < Δ T S q W = q max , if Δ T W < c s Δ T S q W = q min + Δ T L - Δ T W Δ T L - c s Δ T S ( q max - q min ) , if ΔT W < ΔT L q W = q min ΔT W Δ T L else
(2806)

Δ T L = T L - T sat and c s is assumed to be 1.2.

The component mass rates of vaporization are then computed as

6. EQUATION_DISPLAY
m ˙ v , i = c e Y i q W Δ H vap
(2807)

where c e is the wall boiling evaporation factor that determines how much of the boiling heat flux generates vapor. A value smaller than one means that a portion of the boiling heat flux does not generate vapor but heats up the film, for example:

  • when heat is supplied near the wall to bubbles which then move away into the liquid film and condense into the film
  • when evaporation takes place at the base of a bubble and condenses at its tip.
Bulk Boiling
The heat transfer to the interface between the phases is used to compute the rate of evaporation.

For the Rohsenow boiling model, Simcenter STAR-CCM+ calculates the mass rate of evaporation from:

7. EQUATION_DISPLAY
m ˙ v =   HTCxArea ( T - T sat ) Δ H vap
(2808)
assuming that:
  • The vapor bubbles are at the saturation temperature.
  • The temperature of the liquid is approximately the temperature of the mixture.
  • All the heat flux from the liquid to the interface is used in mass transfer due to evaporation.

HTCxArea is the product of the heat transfer coefficient between the vapor bubbles and the surrounding liquid, and the specific contact area (contact area per unit volume) between the two. Δ H vap is the latent heat of vaporization of the mixture:

8. EQUATION_DISPLAY
Δ H vap = i Y i ΔH i vap
(2809)

where Y i is the mass fraction and ΔH i vap is the latent heat of pure component i . This latent heat is the difference between the enthalpy of its liquid phase H i , f and its gas phase H i .

9. EQUATION_DISPLAY
ΔH i vap = H i - H i , f
(2810)

Habchi Boiling

The Habchi boiling model calculates the vaporization of liquid films across both nucleate and transition boiling regimes. Additionally, the Habchi boiling model considers the effects of ambient gas pressure and wall roughness when calculating the liquid evaporation rate.

The Habchi boiling approach uses the saturation, Nukiyama, and Leidenfrost temperatures to differentiate between the fluid film boiling regimes.



  • Below the saturation (boiling) temperature, when 𝑇 𝑤 < 𝑇 𝑠 𝑎 𝑡 , the heat flux is low, and evaporation occurs to some extent.
  • When 𝑇 𝑠 𝑎 𝑡 < 𝑇 𝑤 < 𝑇 𝑁 ​, vapor cavities form around nucleation sites on the wall. The maximum heat flux is reached at the Nukiyama temperature 𝑇 𝑁 .
  • For 𝑇 𝑁 < 𝑇 𝑤 < 𝑇 𝐿 ​, above the Nukiyama temperature, vapor pockets coalesce to form a cushion, which reduces the contact of the liquid with the wall, thus slowing conduction.
  • When 𝑇 𝑤 > 𝑇 𝐿 ​, in the Leidenfrost regime, the presence of the vapor cushion further slows conduction.

At high wall temperatures, the film rapidly heats up and boils away. As a result, when the film thickness approaches zero, Simcenter STAR-CCM+ limits the maximum film temperature to the wall temperature to ensure that a defined zero thickness film behaves as if there was no film. If any film remains and boiling occurs, the film temperature remains below 𝑇 𝑠 𝑎 𝑡 .

The heat flux from the wall into the liquid film phase is defined as:
10. EQUATION_DISPLAY
q w l = P w l Δ T w
(2811)
and:
11. EQUATION_DISPLAY
P w l = k l , s a t δ t h
(2812)
where k l , s a t is the thermal conductivity of the liquid film at saturation temperature and δ t h is the thermal boundary layer height near the wall. The default value for δ t h is based on iso-octane fuel ([622]).

The thermal boundary height can be calculated by using the maximum evaporation rate m C H F . (at critical heat flux or Nukiyama temperature) and the latent heat at saturation temperature L s a t , obtained as:

12. EQUATION_DISPLAY
δ t h = β 1 ( 1 α d r y ) k l , s a t ( T N T s a t ) L s a t m C H F .
(2813)
The heat flux per unit area from the wall through the vapor cushion to the liquid phase is defines as:
13. EQUATION_DISPLAY
q w v l = P w v l ( T W T L )
(2814)

where:

  • P w v l = k v δ v
  • k v is the vapor thermal conductivity.
  • δ v is the thickness of the vapor cushion.
The thickness of the vapor cushion is calculated as:
14. EQUATION_DISPLAY
δ v = δ p ( p 0 p ) 2
(2815)
where p 0 is the reference (atmospheric) pressure and δ p is the thickness of the vapor cushion at p = p 0 .
The total conductive heat flux is the combination of the heat flux from the wall into the liquid phase and the heat flux per unit area between the wall through the vapor cushion to the liquid:
15. EQUATION_DISPLAY
q b o i l = q w l + q w v l
(2816)
Simcenter STAR-CCM+ uses a heat flux blending function to determine the allocation of heat flux between heating the film and contributing to vaporization. The amount of heat flux that contributes to boiling is then:
16. EQUATION_DISPLAY
q b o i l = q t o t q b l e n d
(2817)
The mass evaporation rate is calculated from:
17. EQUATION_DISPLAY
m ˙ v =   β 1 ( 1 - α d r y ) q w l + β 2 α d r y q w v l Δ H vap
(2818)
where:
  • α d r y is the dry fraction of liquid film area due to boiling, and is calculated as:
    18. EQUATION_DISPLAY
    α d r y ( T W ) = α d r y L T * 1 4
    (2819)
  • α d r y L is the dry fraction of liquid film value at the Leidenfrost point. This value is 0.98 .
  • β 2 is the dry area fraction coefficient which sets the strength of the boiling rate through the vapor cushion. The default value of 0.055 is based on iso-octane.
  • T * is computed from:
    19. EQUATION_DISPLAY
    T * = Δ T W Δ T L
    (2820)

and:

20. EQUATION_DISPLAY
β 1 = c l l d max h f sin θ k c l l d 2
(2821)
where:
  • h f is the film thickness.
  • θ is the contact angle.
  • c l l d max is the maximum length of the contact lines of the liquid film.
  • k c l l d is a dimensionless length density of contact lines, defined as:
    k c l l d ( T W ) = ( 1 k c l l d min T * * + k c l l d min )
    (2822)
    and T * * = T L T W Δ T L .
  • k c l l d min is the minimum dimensionless length density of contact lines.
The Habchi method computes k c l l d min as a function of the surface roughness of the wall:
k c l l d min = k R u 1 α d r y L R u k R u 2
(2823)
where:
  • k R u 1 and k R u 2 are adjustable constants of the model. The default values are 1.0 and 0.2, respectively.
  • R u is the average roughness of the wall (height is assumed).

Nukiyama and Leidenfrost Temperature

The Habchi method in Simcenter STAR-CCM+ calculates the Leidenfrost and Nukiyuma Temperatures as:
21. EQUATION_DISPLAY
T c r = T s a t + Δ T
(2824)
T c r is either the Leidenfrost T L or the Nukiyuma T N temperatures, calculated as:
22. EQUATION_DISPLAY
T c r = { T c r | 1 b a r T b , if p 1 b a r ( T c r | 1 b a r T b ) A T c T b ( T c T s a t ) , if p > 1 b a r
(2825)

where:

  • T c is the critical temperature. For multi-component liquid this is the highest of the critical temperatures of the liquid components.
  • P is pressure.
  • A is A = max ( 1 , T c r | 1 b a r T c ) .
  • T b is the normal boiling temperature at 1 atm. For a multi-component liquid, choose T b as a function of composition.

If there are no experimental data available, the Nukiyama temperature is estimated it to be halfway between the film saturation temperature and the Leidenfrost temperature:

23. EQUATION_DISPLAY
T N = T s a t + T L 2
(2826)