COMBUSTION FUNDAMENTALS: CHARACTERISTICS OF COMBUSTION PROCESSES
Referring to the reaction rate equation, there are two factors which dictate whether combustion takes place. That is for the system to be transformed from a stable mixture to a rapidly reacting combustion process. The first is that there should be sufficient energy to allow initiation of the reaction. This may be supplied by any of a number of methods: electrical spark, heat addition, radical addition.
The propensity of a system to burn is dependent on many factors: fuel type, calorific value, mixture, pressure, velocity, turbulence, enclosure geometry.
SIT may be characterised in some way by the Spontaneous Ignition temperature. This is defined as the temperature above which combustion, once initiated, will maintain itself and below which active combustion cannot occur. This is tabulated in table I for some common fuels.
If heat is conducted away sufficiently rapidly, the temperature may be reduced below the ignition point and combustion ceases. Even rapidly burning explosive mixtures may be quenched by sufficient cooling.
The ignition temperature of gas varies with its concentration and is reduced by pressure increase. Substitution of oxygen for air has little effect implying the process is more governed by the Arhennius expression than concentration. Figure 1 shows the effect of pressure and temperature on SIT.
Once ignition is achieved, the next requirement is that combustion is sustained. A mixture of fuel and oxidant is not necessarily able to sustain combustion. Combustion may only be sustained if the heat released due to combustion is greater than that absorbed by the surroundings. Requirements for a flame in gas / air mixture:
This range depends on
These also depend on:
The Flammability range will be higher if:
Flammability limits for some common fuels are given in table II.
The following rough formula to estimate the weak limit value to within approx. 20% may also be used as a first estimate.
Weak limit (% fuel) ´ Calorific Value (MJ / kmol) = 4605
The formula below is also useful to estimate the effect of temperature on the lower limit based on the Burgess-Wheeler law (Lefebvre).
LT = Weak limit at temperature, T, (percent fuel by volume)
LCV = lower CV of fuel (MJ / kmol)
FLAMMABILITY OF MIXTURES
We often have the situation where the gas is composed of a number of species, not just one. In this case, the flammability limit of the new mixtures may be calculated using LeChatellier’s formula.
x are the volume percentages of combustibles in the mixture.
l is the limit (upper or lower) of flammability of the individual components in air (%)
L is the limit of flammability (upper or lower as the case may be) of the mixture in air (%)
If one of the components in the mixture is not combustible then a different formula must be used.
For the weak (lower) limit we use:
For the rich (upper) limit we use:
- If the diluent is air
- If the diluent contains no oxygen
If the diluent (other than air) contains oxygen.
Where L1 and L2 are the lower and upper limits of flammability
L1’ and L2’ are the limits of the combustible containing the diluent.
xdil, xcomb and xO2 are the percentages of diluent, combustible and oxygen respectively in the mixture .
When calculating the flammability limits of a mixture with a number of combustibles and diluents then first calculate the flammability of the pure combustible mixture (minus diluent). Then calculate the effect of the diluent afterwards.
Flammability is VERY important in the explosions field. Flash point is an important property of liquid or solid fuels. Defineed as the temperature at which the vapour just above the fuel surface is flammable.
1) Calculate the flammability limits of a mixture of 25%CO, 40% H2 and 35%CH4.
2) Calculate the flammability limits of a mixture of 25%CH4, 45%H2, 10% CO, 10% CO2 and 10.0% N2. What are they if the N2 is replaced with oxygen.
Another vital characteristic of combustion of a fuel is the flame speed. The speed at which a laminar flame propagates through a premixture of fuel and air in theory is completely dominated by combustion chemistry (in particular the activation energy of reaction). In practice this is not entirely true since surfaces have a profound effect acting as heat sinks but also soaking up radicals and terminating the reaction. Davy’s work in the 1800’s.
It is possible to measure laminar flame speed using three methods:
laminar burning velocity, Su, gives a good indication of the combustion charactreristics of a fuel. Itis very sensitive to mixture. The maximum is normally well on the rich side of stoichiometric. fig 2.
Maximum burning velocity is tabulated in table III for a variety of fuels at NTP. H2 has the highest flame speed of all. It is important to realise that the flame speed for most hydrocarbon fuels is very similar (around 0.4m/s). This is because all hydrocarbons are quickly pyrolysed to smaller sized molecules before combustion and hence behave in a similar manner. NB The velocities involved (circa 1m/s) are far smaller than the velocity in any practical device. No easy way of calculating the burning velocity of a mixture because of internal effects.
A dimensionless number, the Weaver flame speed number is used to charadterise flame speed. It is simply calculated as Su / Su (H2). This guarantees a value less than 100%.
Turbulence has a profound effect on flame speed increasing it significantly. Early work by Damkohler postulated that this was due to flame wrinkling increasing the flame surface area and hence flame spread. Much work since has shown this is only true in certain regimes depending on the level of turbulence (or turbulent length scale).
Early work by Davy revealed that combustion cannot occur in tubes below a certain radius. This is due to local cooling of the flame (below SIT) but also due to radical termination. He defined the quenching distance and made great use in coal mining lamps. dq is a complex function of chemistry and geometry. It is related to SIT and for most HC fuels it is around 1 - 4mm. It is strongly dependent on mixture and is minimised close to stoichiometric. See fig. 4.