course overview - home - walter scott, jr. college of ...marchese/combustion11/notes08.doc · web...

MECH 558 Combustion Class Notes - Page: 1notes08 0-Dimensional Chemical Reacting Systems Text: Ch.6

Technical Objectives: Derive the governing equations for constant pressure, constant volume, rapid compression

machine, single-zone HCCI, well-stirred and plug flow reactors.

Use CHEMKIN to solve each of the above problems with detailed chemical mechanisms.

1. MotivationNow that we have finished learning about detailed chemical kinetic mechanisms for some practical fuel/oxidizer systems, we can actually begin to solve some chemically reacting systems. We will begin by looking at homogenous (i.e. 0-dimensional) systems.

The major assumption in these models is that the diffusion of species is ignored. This assumption is acceptable if the system is well-stirred (i.e. infinite mixing rate of all species as they are produced) or if the chemical reaction occurs very quickly and simultaneously over a large homogenous volume.

These systems are of interest in studying combustion because it is possible to perform numerical simulations of these systems even with very large detailed chemical kinetic mechanisms (100’s of species, 1000’s of reactions). Thus, they are used to validate and test chemical kinetic mechanisms without having to worry about the difficult fluid mechanics and mass transport present in real flames.

1.1 Examples of 0-Dimensional Chemically Reacting Systems are as follows:

Constant Pressure, Fixed Mass Reactor Constant Volume, Fixed Mass Reactor

Well Stirred (or Longwell) Reactor Plug Flow Reactor


2. Constant Volume, Fixed Mass ReactorConsider the following closed system, filled initially with a reacting gas mixture at a specified temperature and pressure. The volume is fixed:

Lets assume that we have developed a detailed chemical kinetic mechanism for this fuel/oxidizer system that considers N species and L reactions. (Recall the H2 mechanism considered 8 approximately species and 20 reactions). The unknowns in this problem are as follows:

Thus, for N species, there are N + 2 unknowns that needs to be solved as a function of time. Therefore we need to solve N+2 equations. The N+2 equations are:

Net Production Rate for Each Species

The first N equations are the differential equations that are developed by applying the Law of Mass Action for each the N species for all L reactions:

(4.28a)

(4.28b)

Where

is the net production rate of the ith species in the mechanism in mol/cm3-s, [Ci] the concentration of the ijth species in mol/cm3


The (N+1)th equation can be derived by applying the First Law of Thermodynamics for a transient, fixed mass system:

(6.18)

The total internal energy, U, is the equal to:

(6.18a)

Substituting and taking the derivative of the internal energy term in (6.18) yields;

(6.18b)

Recalling that du = cvdT and u = h – Pv, and substituting, yields:

(6.20)

Equations (4.28) and (6.20) can be used to solve for the concentration of each of the N species vs. time and the temperature vs. time.

Since the volume is constant, any change in temperature will be accompanied by a change in pressure, which can be calculated from the ideal gas equation:

(6.21)


Taking the derivative with respect to time yields the following:

(6.24)

Differential equations (4.28), (6.21) and (6.24) can be used to solve a constant volume, homogenous, transient system, with detailed chemistry, such as the hydrogen-oxygen system from your previous homework problem set. In fact, these equations are virtually identical to the differential equations solved by Chemkin for this same type of system.

3. Variable Volume, Fixed Mass Reactor: Rapid Compression MachineA rapid compression machine (RCM) is an instrument designed to simulate the compression stroke of a single engine cycle thereby allowing auto-ignition phenomena to be studied in a much more controllable environment than would be possible in an actual engine. The device functions by rapidly compressing a homogeneous premixed volume of fuel and air. The rapid compression is accompanied by rapid heating of the mixture and the process results in an elevated final temperature and pressure with negligible heat loss. Since only the test gas is heated and not the vessel walls, wall reactions are avoided and the reactions that occur inside the cylinder are nearly homogeneous. The figure below shows the Rapid Compression Machine currently installed at EECL along modeling results of temperature, pressure and NO mole fraction vs. time for the homogeneous compression ignition of methyl butanoate/O2/N2 at a compression ratio of 20 and initial temperature of 315 K.

The figure below predicted temperature vs. time history for n-heptane/iso-octane mixtures (PRF0 to PRF100) in a RCM in HCCI mode at = 0.4 and compression ratio of 16. As the figure shows, the fuel-air mixtures are initially compressed adiabatically over a period of 5 ms. After a subsequent induction period, which depends on the low temperature reactivity of the mixture, hot ignition occurs and the temperature rapidly rises. During the induction period, the moderately reactive mixtures (PRF 75 and 90) exhibit low temperature heat release, as signified by the temperature rise during this period.

time (s)

0.000 0.005 0.010 0.015 0.020N

O (p

pm)

0

20

40

60

80

100

NO (ppm)

Tem

pera

ture

(K) a

nd P

ress

ure

(psi

a)

0

500

1000

1500

2000

2500

Temperature (K)Pressure (PSIA)


To simulate an RCM using CHEKMIN, it is possible to use the same “fixed volume” reactor model described above, wherein a given Volume vs. time curve is supplied as an input to CHEMKIN as shown below:

4. Plug Flow ReactorA plug flow reactor is a system that can be used to study chemical kinetics, since it enables the experimentalist to sample at various distances along the reactor, thereby allowing the measurement of species as a function of distance in a chemical reaction. By using dilute mixtures of fuel and oxidizer, it is possible to slow down the chemical reaction, which stretches out the chemical reaction zone over a distance of approximately 1 meter (vs. approx. 1 mm in a flame).

Electric ResistanceHeater

Evaporator

Fuel Inlet

Slide Table

Oxidizer Injector

Optical Access Ports

Sample ProbeWall Heaters


The schematic diagram above shows the Princeton Variable Pressure Flow Reactor. This system can be solved theoretically by making the following assumptions:

1.

2.

3.

4.

5.

Note that the only derivatives in this case are d/dx. If one were to “ride along with the plug flow” it would appear that all of the species were varying with respect to time (d/dt), so this problem is really very similar to the homogeneous problem discussed in the previous section.

Consider the flow tube in the flow reactor above:

In this case, the unknowns are (x), v(x), T(x), P(x) and Yi(x) for i = 1, N. These N+4 equations can be solved using the conservation of mass, momentum, energy, equation of state and the N species equations.

Assuming that heat transfer to the wall Q”(x) is zero and the cross sectional area A(x) = A = constant, the following are the governing equations:

Equation of State

(6.47)

In differential form:

(6.48)

Conservation of Mass

(6.39)

Which can be rewritten as:

(6.43)


Conservation of x-momentum

(6.40)

Conservation of Energy

(6.41)

Which can be rewritten as:

(6.44)

But, dh/dx can be evaluated based on the mass fractions:

(6.46)

Resulting in the following energy equation:

(6.44a)

Species Equations

The net production rate of the ith species can be converted from d/dt to d/dx and the concentrations can be converted to mass fractions as follows:

(6.42)


Equations (6.48), (6.43), (6.40), (6.41), (6.44a) and (6.42) can be combined into the following three differential equations in x, which can now be solved:

(6.51)

(6.52)

(6.53)

5. Single Zone, Homogeneous Charge Compression Ignition Engine SimulationIt is also possible to simulate the closed compression-ignition process of an internal combustion engine as a 0-dimensional, transient system, in which volume varies with time based on the geometry of the piston-cylinder assembly. These simulations are useful for studying autoignition in IC engines and also for studying a new class of engines called Homogeneous Charge Compression Ignition (HCCI) engines.

Consider the following engine cylinder:

Where D is the cylinder diameter, Lc is the length of the connecting rod, LA the crank arm radius, Vc the clearance volume and Vs the swept volume.

The maximum swept volume is given by:

(n7.1)


The engine compression ratio is defined as follows:

(n7.2)

Given these geometrical parameters of an engine-cylinder it is possible to derive the following relationship for volume vs. time within the engine combustion chamber:

(n7.3)

Where R is the ratio of the connecting rod to crank arm radius (Lc/LA) and C is the compression ratio.

The time rate of change of volume is given by the following:

(n7.4)

Where is the angular velocity of the crank arm (d/dt).

Equation (n7.4) can be solved along with the equation of state, species equations (4.28a) and conservation of energy (with variable volume and pressure) as follows:

(n7.5)

It is also possible to model the heat transfer to the walls of the cylinder using the following model:

(n7.6)

Where Twall is the wall temperature and h the wall heat transfer coefficient.

Combining Equations (n7.5) and (n7.6) results in the following energy equation:

(n7.5a)


The wall heat transfer coefficient can be approximated using the following Nusselt number correlation:

(n7.7)

Where Reynolds number is approximated as follows:

(n7.8)

Where Sp is the mean piston speed.

course overview - home - walter scott, jr. college of ...marchese/combustion11/notes08.doc · web...

Documents