incineration systems for liquid and gaseous wastesdl.mozh.org/up/dk1860_10.pdfviscosity=density) of...

10

Incineration Systems for Liquid andGaseous Wastes

I. LIQUID WASTE INCINERATORS

In many industrial processes, waste liquids are produced for which disposal by incineration

is effective, economical, and environmentally sound. Such waste liquids include waste-

water contaminated with combustible toxic chemicals, solvents or oils for which purifica-

tion costs are excessive, and heavy pitches and tars. Not all of these wastes fall into the

category of ‘‘hazardous wastes,’’ but it is noteworthy that the largest fraction of the U.S.

hazardous waste incineration facilities involve liquid waste incinerators.

In many instances, waste liquids are burned as fuels in larger furnaces designed

primarily for solids disposal. However, several incinerator designs have been developed for

liquid firing alone. These furnaces are comprised of cylindrical chambers with atomized

liquid and combustion air introduced axially or tangentially. Also, the fluid bed incin-

erators described in Chapter 9 may be used for liquid waste incineration.

A. Liquid Storage

Often liquid waste incinerators are operated only a fraction of the work week or day. Thus,

tanks are required to hold the waste. The design of such containers and associated piping

should include careful consideration of the following.

1. Corrosive attack. The liquids to be stored may range widely in chemical

composition. The availability of an incineration system in a plant will often result in its

use (and misuse) for waste streams perhaps not envisioned during design. The materials of

construction of the storage tank, piping, valves, pumps, etc., should be selected with this in

mind.

2. Chemical reactions. Particularly for incineration facilities serving complex or

multiproduct chemical plants or for commercial incineration facilities, a wide variety of

wastes may be sent to the incinerator system. In many cases, the exact composition is

SOFTbank E-Book Center Tehran, Phone: 66403879,66493070 For Educational Use.

unknown and the full range of possible reactions cannot be investigated. Reactions of

concern include

Exothermic reactions that liberate enough heat to cause boiling, accelerate corrosion,

etc.

Polymerization reactions that could solidify or turn the tank contents into an

unpumpable gel

Gas-forming reactions that could cause foaming or otherwise force liquid out of the

tank

Precipitation reactions that could produce unacceptably large quantities of solid

sludge in the tank

Pyrophoric reactions that could result in spontaneous ignition of the tank contents.

3. Phase separation. Often various waste streams are immiscible or contain

settlable solids. Upon setting, phase separation or settling can occur such that the

incinerator will experience major changes in feed composition over a short time, an

undesirable condition. Decanting systems, recirculating pumps, and=or agitation may

resolve these problems.

4. Abrasion. The presence of solid phase in the wastes can cause rapid abrasion in

pumps, valves, and piping.

5. Freezing and=or viscosity increase. Many wastes freeze or become viscous at

ambient temperatures. In such cases heating of the containers and steam or electrical

tracing of piping may be necessary to avoid freezeups or pumping problems.

6. Vaporization. Introduction of hot wastes into the storage tank may volatilize

low-boiling compounds, sometimes explosively. Proper care in waste dumping and the

installation of tank vents (with flame arresters) are appropriate countermeasures.

B. Atomization

Atomization is the process of physically breaking up a liquid into particles. Liquid wastes

should be atomized so that the combustion air can quickly surround the surface of the

droplets to produce a combustible mixture. Also, fine atomization speeds the rate of

vaporization of the waste, a prerequisite for ignition and combustion.

A number of methods are available to effect the atomization of liquid wastes. They

vary in their relative capital and operating cost, their maximum capacity (l=min), the

proportion of combustion air to be supplied as secondary air, the range of operating rates

(turndown) required, and the desired flame shape.

The minimum energy input requirements for atomization are determined by the

viscosity of the waste at the atomization point. The kinematic viscosity (absolute

viscosity=density) of the liquid is often used to characterize atomizer requirements. In

the CGS system, the units of kinematic viscosity are cm2=sec, or ‘‘stokes.’’ For oils andother viscous liquids, the centistoke gives numerical values in the 0 to 100 range. The

more common unit is based upon a determination of the kinematic viscosity by

measurement of the time t (in seconds) of efflux of a fixed volume of fluid through a


short standard capillary tube. Commercial viscometers are designed to obey the empirical

relationship

v ¼ a1 �a2

tcm2=sec ðstokesÞ ð1aÞ

t ¼ v

2a1þ v

2a1

� �2

þa2" #1=2

ð1bÞ

For a common viscometer (4):

Viscometer a1 a2

Saybolt Universal (SSU) 0.0022 1.8

Redwood No. 1 0.0026 1.72

Redwood No. 2 (Admirality) 0.027 20

Engler 0.00147 3.74

It is also useful to recognize that, for many liquids, a plot of absolute viscosity versus

absolute temperature on log-log paper is, essentially, a straight line.

EXAMPLE 1. Measurements of the kinematic viscosity of a waste oil indicate 200

SSU at 99�C and 400 SSU at 83�C. The specific gravity of the oil in this temperature range

is 1.02. To what temperature should the oil be heated to have a kinematic viscosity of 85

SSU, the viscosity recommended for atomization?

From Eq. (1a), the kinematic and absolute viscosities corresponding to the Saybolt

Universal determinations are:

SSU Centistokes Centipoises

at 99�C (372�K) 200 43.1 44.0

at 83�C (356�K) 400 87.6 89.3

The objective is to find the temperature where the oil has a viscosity of 85 SSU,

which corresponds to v ¼ 16:58 or m ¼ 16:91. Plotting m versus absolute temperature on

log-log paper gives the result 122�C.The droplet size formed through pneumatic atomization is important as it strongly

affects the flame length and heat release patterns in the incinerator. Consequently, several

investigators have analyzed the atomization process. Calvert considered the atomization

process as one where the bulk flow shattered at a critical value of the Weber number he

suggested to be about 12. The Weber number is a dimensionless ratio of momentum to

surface tension forces given by

NWe ¼rgV

2r d

2sð2Þ

where

rg ¼ density of the gas

Vr ¼ velocity of the gas stream relative to the droplet

d¼ droplet diameter

s¼ surface tension


A discussion by Murty (348) noted the utility of the following upper-limit function

(351) to describe the distribution (symmetrical about the axis y ¼ 0) of droplet volume in a

stream sprayed from a pressure nozzle:

dV

dy¼ dffiffiffi

pp e�d

2y2 ð3Þ

where

V ¼ volume of the fraction of droplets with a diameter x

y¼ ln½ax=ðxm � xÞ� with ‘‘a’’ as found by experiment (352)¼ 0.8803

xm¼maximum drop diameter (m)

d¼measure of deviation and found by experiment (352)¼ 0.8739

Given the symmetry about y ¼ 0 and the fact that the median-volume diameter

(xmvd¼ the diameter corresponding to 50% of the total droplet volume) corresponds to the

value of x at y ¼ 0, we thus have

xmvd ¼xm

1þ a¼ xm

1:8803ð4Þ

To find xmvd, we substitute the maximum drop diameter from the following expression

(353), giving

xm

D¼ 57

rvDm

� ��0:48 mvs

h i�0:18ð5Þ

where

r¼ density of dispersed liquid (kg=m3)

m¼ absolute viscosity of dispersed liquid (kg=m sec)

s¼ surface tension of dispersed liquid (N=m)

v¼ drop velocity at nozzle exit (m=sec)D¼ nozzle orifice diameter (m)

From this correlation, we can estimate the Sauter mean diameter (volume-surface mean

diameter) as 85% of xmvd.

1. Low-Pressure Air Atomization

These burners require air at 0.035 to 0.35 atm, usually supplied by blowers. The minimum

air pressure requirements (energy cost) are determined by the viscosity of the liquid to be

atomized. A heavy pitch with a viscosity (heated) of 80 to 90 SSU requires air over

0.1 atm, whereas aqueous wastes can be atomized at 0.035 atm. Usually, the waste liquid is

pumped to a pressure of 0.3 to 1.2 atm. Turndown for these burners range from 3 : 1 up to

6 : 1. The air used for atomization ranges between 2.8 and 7.4m3=l, with less air required

as the air pressure increases. The resultant flame is comparatively short as, even for a pure

fuel oil, about 40% of the stoichiometric air is intimately mixed with the fuel spray as the

mixture enters the furnace.

2. High-Pressure Air or Steam Atomization

These burners require air or steam at pressures in excess of 2 atm and often to 10 atm.

Atomizing air consumption ranges from 0.6 to 1.6m3 air=l of waste as the supply pressure

varies over this range. Steam requirements range from 0.25 to 0.5 kg=l with careful


operation (a wasteful operator may use up to 1 kg= l). Waste heating to reduce viscosity is

required only to the extent needed for pumpability.

In general, high-pressure atomizing burners show poor turndown (3 : 1 or 4 : 1) and

consume considerable compressor energy or steam (typically, for boilers, about 2% of the

steam output). Since only a small fraction of the stoichiometric air or a inert gas (when

steam is used) is mixed with the emerging fuel spray, flames from these burners are

relatively long.

In burning high-carbon pitches and tars, the addition of steam has been shown to

reduce the tendency for soot formation. This apparently results from the enhanced

concentration of hydroxyl radicals, which act as vigorous oxidants with the unsaturated

carbon radicals, which are precursors to soot.

3. Mechanical Atomization

These burners atomize by forcing the liquid, under high pressure (5 to 20 atm), through a

fixed orifice. The result is a conical spray into which combustion air is drawn. In its

simplest embodiment, the waste is fed directly to a nozzle. With such an arrangement,

turndown is limited to 2.5 to 3.5 : 1, since, for example, a 75% reduction in pressure

(atomization energy) is required to reduce the flow rate by 50%. Thus, atomization

effectiveness (droplet fineness) drops rapidly as the burner moves off the design flow rate.

The second type of mechanical atomizing nozzle incorporates a return flow such that

a much smaller change in atomization pressure is needed to effect a change in flow rate.

For these atomizers, turndown can be as great as 10 : 1. The viscosity of fluids atomized

mechanically need not be as low as that for low-pressure air atomization: 150 SSU is a

typical design value.

The flame from mechanical atomizing burners is usually short, bushy, and of low

velocity. The half-angle of the flame can be altered somewhat by changing the atomizing

nozzle but, because all the air is provided by secondary means, combustion is not as rapid

as with other types of burners, and a larger combustion space is usually required.

Mechanical atomizing burners are usually applied where large peak capacities (40–

4000 l=hr) are required and where large turndown ratios are desirable. Its capital cost is

higher than for other designs, but subsequent operating costs are low. In waste-burning

applications, consideration must be given to the problems of erosion and plugging of small

orifices due to solid matter in the waste stream.

4. Rotary Cup Atomization

These burners atomize by the action of a high-speed rotating conical metal cup from the

outer edge of which the waste liquid is thrown into a stream of low-pressure air entering

the incinerator around the cup. The rotary cup is usually attached to an extension of the

low-pressure centrifugal blower shaft, and the waste liquid is delivered to the cup at low

pressure through the shaft or at the side of the cup at its inner edges. Hinze and Milborn

have extensively studied the atomization process in rotary cup devices (405) and offer

several design and process correlations.

Since the rotary cup system has little requirement for fluid pressurization, it is ideal

for waste-burning applications where the solids content of the waste is high. Also, the

viscosity of the waste need only be reduced to 150 to 330 SSU. Turn-down is about 5 : 1,

and burners with capacities from 4 to 1000 l=hr are available.


The flame shape from rotary cup burners is similar to that from mechanical

atomizing burners, but with a somewhat increased combustion rate since a portion of

the combustion air is supplied with the waste stream.

5. Secondary Atomization

Secondary atomization involves the introduction of a volatile component into a waste

stream. When droplets of the mixture are heated as they enter the combustion space, the

volatile component flashes. The sudden release of vapor shatters the droplet, thereby

improving atomization quality. The enhanced atomization can be especially useful when

the waste is unusually viscous or contains solids or when erosion or plugging has

somewhat degraded nozzle performance such that atomization has become the limiting

factor in achieving or maintaining the target DRE. Note that secondary atomization may

already be inherent in a given system due to the normal range of volatilities found in the

waste.

This technique was demonstrated in atomization and combustion tests to be effective

(322) in improving atomization effectiveness and achieving an enhanced DRE of a test

material (No. 2 fuel oil). The dopants used and their boiling points were dichloromethane

(39�C), acrylonitrile (79�C), benzene (80�C), isopropanol (82�C), and benzal chloride

(205�C). These represented a wide range of volatility relative to the No. 2 oil (210–260�C).It was found that the onset of secondary atomization was not simply related to boiling

point in that isopropanol was, by far, the dopant that gave the greatest enhancement to

atomization at the lowest concentration in the base ‘‘waste.’’ The intensity of secondary

atomization was dependent on the dopant concentration. For most dopants, secondary

atomization was active only at a concentration above 2%. Isopropanol showed activity at

only 0.5%.

Combustor tests were made (322) with isopropanol (high-effect dopant) and benzal

chloride (low-effect dopant) under conditions where atomization effectiveness was limiting

on DRE. The results showed the following:

Fraction of waste undestroyed

Weight fraction in mixture Isopropanol Benzal chloride

0.5% 1.8� 10�3 2.5� 10�2

2.0% 2.5� 10�5 1.2� 10�2

10.0% Nondetect 1.7� 10�3

C. Ignition Tiles

In order to ensure rapid ignition of the waste-air spray, a refractory block or ‘‘ignition tile’’

is used. The tile usually consists of a conical depression with an opening at the small end

of the cone which mates to the atomizing burner. Its objective is to facilitate lighting,

maintain ignition under all normal conditions, and confine the air introduced by the burner

so that it will be properly mixed with the vaporizing waste. Its design affects the shape of

the flame and the quantity of air that can be induced by the burner.

When the heat content of the waste is low or when the combustion chamber

temperatures are too low to secure complete combustion, a refractory tunnel extension is


often added to the ignition tile, increasing the intensity of radiation (for waste vaporiza-

tion). When wastes contain a substantial fusible ash content, special care must be given to

the type and design of these tiles to avoid rapid fluxing losses and slag buildup.

D. Combustion Space

The combustor heat release volume requirement (kcal=m3) depends on the combustibility

of the waste and the mean furnace temperature. For difficult-to-burn wastes or low furnace

temperatures, more volume is needed than for the reverse. Typical ranges are as follows:

Temperature Combustion volume

(�C ) (kcal hr�1 m�3)

300–800 30,000–130,000

800–1100 130,000–350,000

1100–1400 350,000–500,000

1400–1650þ 500,000–900,000

Flue areas should be chosen to balance the desire to minimize the infiltration of

tramp air (i.e., keep furnace pressure elevated) and yet avoid pressurization of the furnace,

which will inhibit the flow of needed combustion air. For systems operating such that 25%

to 50% of the air is to be induced by natural draft, the total air supply approximates 20%

excess, and furnace gas temperatures are about 1000�C; approximately 0.25m2 per million

kcal=hr heat release is a typical design point. If all of the air is supplied by forced draft,

one-half this flue area is typical. However, it should be recognized that a large number of

variables are involved in such determinations, and a careful analysis of furnace flow

dynamics is appropriate prior to setting flue dimensions.

The penetration of the liquid jet into the combustion chamber has been studied

extensively. Ingebo (404) showed that the penetration depended on the Weber number [Eq.

(2)] and the Reynold’s number and on the liquid-to-gas velocity ratio. The maximum

penetration distance xmax is related to the maximum droplet diameter dmax by

xmax

dmax

¼ 0:08NRe N�0:41We

Vg

VL

� �0:29

ð7Þ

It should be noted, however, that this relationship does not take into account the

combustion process (including the transition, above a critical gas velocity, from a diffusion

flame all around the droplet to a ‘‘wake flame’’). In the wake flame scenario, the flame

envelope is stripped from the droplet and combustion occurs in the wake behind the

droplet (405).

On the other hand, for volatile fuels, most droplets do not burn individually but

rapidly evaporate and thereafter burn in a jet, much as a gas diffusion flame. This was

proven in experiments conducted with light distillate (406) and heavy (407) fuel oil.

The combustion time (tb in sec) for droplets of hydrocarbon liquid of a molecular

weight MWi, a minimum size of 30 mm, and a velocity equal to that of the gas may be

computed as follows (153):

tb ¼29;800

PO2

� �MWiT

�1:75 d20 ð8Þ


where

PO2¼ partial pressure of oxygen in the ambient (atm)

T ¼ furnace temperature (K)

d0¼ the original droplet diameter (cm)

E. Incinerator Types

Incinerators for liquids are typically comprised of simple cylindrical, refractory-lined

chambers.

1. Axial or Side-Fired Nonswirling Type

In these units, the burner is mounted either on the axis or in the sidewall, firing along a

radius. Such units are simple to design and construct although they are relatively inefficient

in the use of combustion volume. For these systems, the lower combustion volume heat

release rate parameters are appropriate. In essence, these units simply provide a hot

refractory enclosure in which to burn wastes and collect the flue gases for pollution

control.

In the design, special care should be given to realizing good turbulence levels to

ensure that a large fraction of the combustion volume is utilized. To this end, high-pressure

secondary air jets are appropriate. Care should also be given to evaluate the probable flame

length (to avoid flame impingement).

2. Vortex Type

To increase the efficiency of utilization of combustion space, swirl burners or tangential

entry designs are commonly used. In these systems, one of two designs is commonly used:

the axial swirl burner or tangential inlet cyclonic designs. The design concepts for these

systems are described in Chapter 6, Section I.B. In designs where there is concern

regarding flame stability, a small, side-mounted burner is added and operated as the fuel-

supervised burner, leaving the main burner as the mixing burner. This concept provides

continuous ignition and flame stability.

II. INCINERATORS FOR GASES (AFTERBURNERS)

The incineration of gaseous streams differs from all other incineration processes in that the

processing rate is almost always driven by the generation rate of the waste stream since the

convenience and economy of storage available for solid, liquids, and sludge are not

available. Furthermore, the characteristics of the gas streams also cannot be readily

blended as can the condensed phase wastes. Thus, the incineration system burning

gases must adapt to what may be very significant changes in flow rate and heat content

(oxygen demand).

The principal performance parameter is the destruction-removal efficiency or

‘‘DRE’’: the percentage of the inlet feed rate of significant waste(s) that persists in the

exhaust stream following the combustion system and any associated air pollution control

equipment. From a practical standpoint, a second important performance parameter is the

energy consumption required to achieve the target DRE. In general, the achievement of the

DRE is in fairly straightforward relationship to the working temperature and oxygen

concentration within the combustion system (although, of course, mixing efficiency and


other factors play a role). Therefore, it is the energy-efficiency factor that shapes the

physical characteristics of the afterburner and, of course, its operating cost and, thus, its

practicality as a means to control the offending compound(s).

A. Energy Conservation Impacts on Afterburner Design

Afterburner systems break into two broad categories depending on the heat content of the

gas stream to be controlled: fuel-rich fumes and dilute fumes. In processing dilute fumes,

where the energy content of the hydrocarbon portion is small, careful attention to energy

consumption is critical to maintain economic acceptability. Political interest in energy

conservation has coupled with significant increases in the cost of purchased fossil fuels in

recent years to emphasize the need to attend to this aspect of system design. Three

concepts illustrate this focus: the conventional direct thermal oxidizer; the recuperative

thermal oxidizer; and the regenerative thermal oxidizer (Fig. 1). A fourth approach to

minimize fuel use involves use of an oxidation catalyst to reduce the gas temperature

required to achieve the destruction target. This approach is discussed in Section B.2 below.

The conventional thermal oxidizer is a simple combustion chamber into which

atmospheric temperature air and the waste gas are introduced. To the extent that the

resulting flame temperature is lower than is desired or forced by regulation or permit, fuel

is added.

In the second alternative, the recuperative thermal oxidizer, the hot exhaust from the

combustion chamber is passed countercurrent through a heat exchanger to preheat the

incoming waste gas. In most cases, the heat exchanger is fabricated from stainless steel and

both corrosion (especially when the waste gas contains chlorine compounds) and materials

constraints limit the heat recovery to about 60% of the heat originally in the exhaust gases.

The recuperative concept may also be combined with a catalytic oxidation system, thus

combining the energy-conserving feature of recuperation with the reduced temperature

requirement of catalytically enhanced oxidation. The pressure drop experienced across the

catalyst and the heat exchanger is related to the level of energy recovery:

Equipment type Heat recovery (%) Pressure drop (kPa)

Fixed bed catalyst 0 1.5

Heat exchanger 35 1.0



In the third alternative, the regenerative thermal oxidizer (an RTO), hot exhaust gases from

the combustion chamber are passed through a bed of ‘‘cold,’’ refractory material. Over

time, the refractory approaches combustion chamber temperature. Meanwhile, before

entering the combustion chamber, the incoming waste gas is preheated by passing through

a bed of ‘‘hot’’ refractory. Over time, this refractory cools and approaches ambient air

temperatures. At an appropriate time, large gas valves are actuated to switch the gas flow

so the now-cold, originally hot material is reheated and the now-heated, originally cold

material becomes the preheating medium. The energy recovery for this type of system can

exceed 90% of the heat originally in the exhaust gases.

The heat sinks in the RTO can be refractory saddles or other types of random

packing or they can be structured refractory packing blocks. From time to time, the


Figure 1 Design strategies for afterburner energy conservation. (a) Conventional thermal oxidizer; (b) recuperative thermal oxidizer; and (c) regenerative

thermal oxidizer.


packing can be held at high temperatures for an extended time to burn out (‘‘bake out’’)

accumulated organic particulate matter. The effectiveness of the seal on the gas valves and

the disposition of the gases in the internal gas passages in the RTO are important

parameters when the DRE requirements for the RTO exceed 95%. Leaking valves

obviously lead to the bypassing of unremediated gases from the inlet plenum to the

outlet duct. The redirection of the incoming gas in the ducting to the outlet duct within the

RTO during switchover also could result in the avoidance of the combustion zone. To

avoid this, manufacturers purge the unremediated gas before opening the valve to the

outlet duct.

B. Current Afterburner Engineering Technology

A large body of knowledge, both theoretical and empirical, is available to support and

guide the design and evaluation of afterburner systems. This knowledge, which concerns

the heat transfer, fluid mechanics, and kinematics of combustion phenomena, can be used

to determine the design characteristics for optimum performance and thus to develop

criteria against which to evaluate existing afterburner systems.

1. Direct Flame Afterburner Technology

Considerable experience exists in the use of direct flame afterburners for the combustion of

gaseous and gas-borne combustible pollutants (152,153). Many of these systems consist of

little more than a burner in a cylindrical, refractory-lined chamber and are constructed on-

site by facilities engineers. Indeed, the designs of many direct flame afterburners now on

the market are derived from such ‘‘homemade’’ devices.

a. Combustible Gaseous Pollutant Control. Most applications for afterburners concern

the destruction of combustible gaseous emissions and, particularly, volatile organic

compounds (VOCs). The concentration of these pollutants is usually too low to permit

self-sustaining combustion because they are often intentionally diluted below the lower

flammability limit for safety reasons. As a result, external energy must usually be added.

Most insurance carriers insist that the gases entering the incinerator be below 25% of the

lower explosive limit (LEL) [which can be estimated, for gas mixtures (298) using Eq. (9)].

For the LEL of pure substances, see Appendix D.

LELmix ¼1Pn

j

xj

ðPni xiÞ � LELj

ð9Þ

where

LELmix¼ lower explosive limit of a mixture of n components

xi; xj ¼ volume fraction of combustible components i; jLEL¼ lower explosive limit of component j

The energy content equivalent to the lower limit of flammability of most gas–air

mixtures is 0.46 kcal=Nm3; therefore, fuel must be introduced into the gas stream to

increase the energy potential and permit subsequent ignition and oxidation of the mixture.

Because preheat (sensible energy) in the pollutant stream is equivalent to chemical energy,

the flammability energy limit (kcal=Nm3) decreases as the inlet temperature increases until

it is zero at the ignition temperature.


However, operating a system at the limit of flammability (either through sensible or

chemical heat input) increases the probability that residual partially burned combustible

materials will appear in the effluent. The presence of these materials (carbon monoxide,

formaldehyde, methanol, etc.) generally indicates either inadequate mixing or quenching

(by dilution air or cooled surfaces). The combustion characteristics of such homogeneous

systems are described below.

COMBUSTION KINETICS. In general, the combustion of hydrocarbon vapors is

controlled by the mixing processes within the system, rather than by combustion kinetics

(154). The classical results of Longwell and Weiss (155), Hottel et al. (11), and Mayer

(156) on combustion in well-stirred reactors have clearly demonstrated the extremely high

combustion intensities possible (107 to 109 kcal hr�1 m�3) when mixing processes are

eliminated as rate-controlling steps. These workers also showed that the reactions of

hydrocarbons or ketones are very fast relative to that of the carbon monoxide intermediate

formed in the course of the oxidation reactions. It is also clear that all combustion reactions

proceed by a free-radical mechanism and thus are susceptible to wall quenching and the

action of radical stabilizing species such as NO2 and branch chain hydrocarbons.

Nerheim and Schneider studied the burning rate of carbon monoxide and propane

premixed with oxygen, hydrogen, and water vapor in various proportions over ranges of

equivalence ratios and pressures (11). Burning rates were determined from metered flow

rates and analysis of reactor products. The final relationship presented by these authors is

shown in Chapter 2, Eq. (57).

The kinetic mechanism proposed by Nerheim and Schneider which fits the data for

CO called for the rate-limiting step shown in Eq. (10a), equilibrium for Eq. (10b), (10c),

and (10d), and a three-body chain terminating step.

COþ OH ¼ CO2 þ H ð10aÞOHþ H2 ¼ H2Oþ H ð10bÞHþ O2 ¼ OH þ O ð10cÞOþ H2 ¼ OH þ H ð10dÞ

The mechanism proposed for propane combustion involved the addition of a very fast

reaction of propane to CO and H2O at the expense of OH, O, and H. The small difference

between the two correlations suggests the commanding role of CO combustion in

hydrocarbon oxidation reactions.

Combustion kinetics for more complex compounds such as ethers involve more

complex steps at lower temperatures. For example, the slow oxidation of diisopropyl ether

at temperatures between 360� and 460�C apparently consists of the production and

combustion of methyl radicals, the process being facilitated by aldehydes, particularly

acetaldehyde (157). As the temperature increases over 450�C, the thermal pyrolysis of the

ether becomes of great importance in facilitating the production of radicals.

FLUID DYNAMICS. Through the action of free radicals in combustion processes,

intense recirculation patterns near the flame front can be extremely important in

augmenting the combustion rate. By such means, free radicals can be returned to the

ignition areas. Thus, mixing in hydrocarbon afterburners not only promotes intimate

contact of fuel and air but also returns activated species for rapid initiation of the ignition

and combustion reactions. Recirculation effects can be viewed by evaluating homogeneous


combustion systems in terms of plug-flow and well-stirred regions. A number of excellent

studies by Hottel, Essenhigh, and others illustrate such techniques.

An example of a system that uses recirculation to enhance the burning rate is given

by the swirling jet (13). In addition to increasing the combustion intensity, the axial

recirculation vortex causes burning gases to travel back toward the burner, thereby piloting

the flame and increasing its stability. The appearance of an optimum swirl number for

combustion systems suggests the ability of the designer to control the ratio of well-mixed

and plug-flow zones to maximize combustion intensity and flame stability (158).

b. Combustible Particulate Pollutant Control. Many industrial processes emit particulate

matter. The composition and particle-size distribution of these pollutants vary widely, from

the inorganic dusts of the mineral pyroprocessing industry to the high-moisture and

volatile matter, low-ash dusts of the grain milling industry and the dry low-ash furnace

dusts of the channel black industry. In many cases, the recovery of product values or the

relatively large dimensions of the particles suggest the application of conventional

particulate control systems (cyclones, precipitators, filters, scrubbers, etc.) rather than

destruction. As the particle size drops below 50 mm, however, and as the value of the

material to be recovered diminishes (e.g., the aerosols from a drying oven or the soot from

a solid waste incinerator), the applicability and desirability of direct flame incineration

increase.

The sections below discuss the important combustion parameters that apply to the

burning of such particulate materials.

RETENTION TIME. One of the primary considerations in the design of afterburners is

the retention time. In principle, retention time is a derived quantity, calculable as the sum

of the time for preheat of the particulate and the gas stream and the appropriate combustion

time for the particulate or gaseous species under consideration, as influenced by the

dynamics of the combustion system. In practice, however, the retention time is often

considered to be constant, and consideration is not given to the possibility of reducing it

(and thus system cost) through manipulation of the controlling parameters (159). This

arises, in part, from the assumption that the system designer cannot greatly alter the times

required for the various process steps and from the relative complexity of the analysis and

computations required. In fact, the designer does have a measure of control over the

combustion time of particles (through temperature and air control), the preheat time for the

gas stream, and, to some extent, gas phase combustion rates. (The latter can be modified by

utilization of mixing and recirculation principles; recirculation is important for incineration

of hydrocarbon vapors, as discussed below.)

The time required to heat the entering gas to the furnace temperature is dependent on

the combustion intensity within the system and the incremental temperature rise required.

This time (in seconds) may be computed as follows:

t ¼ C�p;avDT_qqv

ð11Þ

where

C�p;av¼ gas heat capacity (kcal m�3 �C�1) over range of DTDT ¼ required temperature rise (�C)_qq¼ combustion intensity (kcal m�3 sec�1)


In low-pressure gas jet mixers, the combustion intensity can be as low as 0.009 kcal

sec�1 m�3, but in premixed mechanical burners it ranges above 4.45 kcal sec�1 m�3 (153).Typical values for premixed high-pressure gas jet, multiple-port burners range from 1.0 to

1.4 kcal sec�1 m�3.Often the difficulty in obtaining high-combustion intensities has been attributed to

the limitations of homogeneous combustion reaction kinetics. Evidence to the contrary is

provided by studies on well-stirred reactors (11,155,156) in which air and fuel are

premixed and fed into the reactors through small holes. The resulting high-velocity

(often sonic) jets promote an intense mixing of the fuel=air mixture with the products

of combustion. These experiments have clearly shown that the kinetics of oxidation for

a large number of fuels are so fast that volumetric heat release rates of 90 to 9000 kcal

sec�1 m�3 are achievable over the 1100� to 1700�C temperature range.

PARTICLE HEATING. The initial step in the combustion of particulate matter is to

raise the surface temperature of the particle to levels where oxidation reactions can occur at

significant rates. In general, the time required to heat particles of the size range of interest

for afterburners to combustion temperatures is small compared to the actual particle

burning time. This is shown in the Nusselt number correlation for zero relative gas velocity

(i.e., when the particulate is moving at the same velocity as the gas):

hc ¼2ld

ð12Þ

where

hc¼ heat transfer coefficient in kcal m�2 hr�1 �C�1

l¼ gas thermal conductivity in kcal hr�1 m�2 (�C=m)�1

d¼ particle diameter (m)

Evaluation of this equation at 700�C for a 50-mm particle, for example, gives an overall

heat transfer coefficient of almost 2440 kcal hr�1 m�3 �C�1. Under such high-flux

conditions, the particle temperature rises quickly to that of the ambient gas, and

combustion ensues under mixed chemical reaction- and diffusion-rate control.

PARTICLE COMBUSTION. The design requirements for the burnout of carbonaceous

particles can be determined from a consideration of the rates of oxygen diffusion to the

particle surface and the chemical kinetics of carbon burning. The time for burnout for the

particles can be shown to be inversely proportional to the oxygen partial pressure, to

increase with particle size, and to decrease with increasing temperature (13). The

approximate burning time in seconds is given by Eq. (63) in Chapter 2.

The two terms within the large brackets of Eq. (63) in Chapter 2 represent the

resistances due to chemical kinetics and diffusion, respectively. At the temperatures found

in incinerators, the burning rate is usually limited by chemical kinetics. Some uncertainty

exists concerning the kinetics and mechanism of carbon and soot combustion, but this does

not influence the general conclusions derived from application of Eq. (63) in Chapter 2.

2. Catalytic Afterburner Technology

The principle underlying the catalytic afterburning of organic gases or vapors derives from

the fact that, from the viewpoint of thermodynamics, these substances are unstable in the

presence of oxygen, and their equilibrium concentrations are extremely small. With a

catalytic afterburner, the combustible materials may be present in any concentration below

the flammability limit. The factors that influence the combustion are temperature, pressure,


oxygen concentration, catalyst selected for use, nature of the materials to be burned, and

the contact of the material with the catalytic surface.

a. Catalyst Systems. An oxidizing catalyst is used. Metals (platinum, etc.), metal oxides,

semiconductors (vanadium pentoxide, etc.), and complex semiconductors (spinels such as

copper chromites, manganese cobaltites, cobalt manganite, etc.) are all known to promote

oxidation of hydrocarbons. Since the surface catalytic reaction consists of a number of

elementary acts such as the breaking and formation of bonds in the reactant molecules and

electron transfer between the latter and the solid catalyst, the electronic properties of

catalyst surfaces are important. Consequently, the catalytic activities of metals and

semiconductors would be expected to differ due to their different electronic properties.

However, under conditions of oxidation catalysis, many metals become coated with a layer

of oxide, and this might be the reason why the mechanisms of hydrocarbon oxidation on

metals and on semiconductors are so similar (160).

Attempts have been made to correlate the activity patterns with the electronic

structure of the catalysts (161–163), the d-electron configuration of cations (164), and the

heat of formation of oxides (165). Only moderate success has been achieved, and at present

a general theory of oxidation catalysis is not available. Consequently, the choice of an

active oxidation catalyst is still based upon extensive empirical information or rather

coarse approximations.

In catalytic afterburner practice, platinum with alloying metals is prevalent because

of its high activity and the lower temperature needed to induce catalytic oxidation when

compared with the other catalysts. It is either deposited on nickel alloy ribbons and formed

into filterlike mats, or deposited on small, thin ceramic rods for the fabrication of small

blocks or bricks. Other possible catalysts include copper chromite and the oxides of

vanadium, copper, chromium, manganese, nickel, and cobalt.

Particularly relevant is a great deal of work done in connection with automobile

emission control. Since automobile engines have various modes of operation (cold starts,

idling, high speed, acceleration, and deceleration), the catalyst system must be effective

over a wide range of exhaust temperatures, gas flow rates, and gas compositions. In

contrast, the conditions encountered in stationary afterburner systems are much more

steady; hence, any catalytic system suitable for automobile emission control should be

effective in afterburners.

b. Catalytic Oxidation Kinetics. Basic data on catalytic oxidation of hydrocarbons have

been available for many years. Anderson et al. (166) showed that even a fairly refractory

hydrocarbon gas like methane can be oxidized completely on a precious metal=alumina

catalyst at 400�C or less. In general, the higher-molecular-weight hydrocarbons are more

easily oxidized than the lower, and hydrocarbons of a given carbon number increase in

reactivity according to the following series:

aromatics < branched paraffins < normal paraffins < olefinics < acetylenics

Some kinetic data are also available in the literature. Work by Caretto and Nobe (167) pays

particular attention to the catalytic afterburning of some substances at low concentrations

in air. They determined the burning rate of saturated and unsaturated hydrocarbons,

aliphatics, aromatics, and carbon monoxide on copper oxide–alumina catalysts. The rate

equations were found to be not of integral orders, and the activation energies were in the

region of 15 to 27 kcal=mol of combustible substance.


In general, the course of a catalytic reaction can be conveniently considered in five

steps as follows:

1. The reactants diffuse from the body of the gas onto the surface of the catalyst.

2. The reactants are adsorbed into the surface.

3. The adsorbed species interact in the surface.

4. The oxidation products are desorbed after the chemical reaction.

5. The desorbed products diffuse into the body of the gas.

Any one of these steps could be the slowest step, whose rate would determine that of the

overall catalytic reaction. When the rates of several steps are comparable, they will jointly

determine the rate of the overall reaction.

The catalytic afterburner usually operates in the region where diffusion rate is

important. Vollheim (168), for instance, found that at temperatures up to 300�C with

copper chromoxide as a catalyst and up to 270�C with palladium, the burning rate of

propane is determined by the reaction rate on the catalyst surface. At higher temperatures,

the effect of diffusion became increasingly noticeable.

In comparison, we may cite that the most commonly encountered hydrocarbons and

combustible organic vapors require catalyst surface temperatures in the range of 245� to400�C to initiate catalytic oxidation. Some alcohols, paint solvents, and light unsaturates

may oxidize at substantially lower catalyst surface temperatures, while aromatics from the

tar melting processes may require higher initiation temperatures. Hydrogen, on the other

hand, will undergo catalytic oxidation at ambient temperatures. Most of the catalytic

burners operate between 345� and 540�C, where diffusion rate is important. Theoretical

relations between mass diffusion and chemical reaction on the catalyst surface are well

developed (169,170).

Laboratory data must be used with caution, because most supporting studies have

used granular catalyst support beds and have paid little or no attention to the fluid pressure

drop. Industrial catalyst systems for fume abatement, however, have had to design with

relatively open structures to minimize the fluid resistance. Since mass transfer is important,

it should be expected that the geometry of these open catalyst support structures should

greatly influence fluid flow, and hence mass transfer behavior.

The influence of pressure on the reaction rate of catalytic combustion processes has

received little attention. This oversight should be remedied because, if reaction rates

increase with pressure, capital cost might be reduced by operating the afterburner

combustion under moderate pressure.

C. Afterburner Systems

Process exhaust gases containing combustible contaminants released at concentrations

within or below the flammable range can, in most cases, be destroyed effectively by either

furnace disposal or catalytic combustion. Properly designed, applied, operated, and

serviced, either system can produce oxidation and odor reduction efficiencies exceeding

98% on hydrocarbons and organic vapors. The choice of one over the other will usually be

based on initial, operating, and service costs and safety rather than on efficiency. With the

thermal disposal technique, the residence temperature may vary from 510�C for naphtha

vapor to 870�C for methane and somewhat higher for some aromatic hydrocarbons. The

operating temperature of the catalytic afterburners is usually about 340� to 540�C. Hein(171), summarizing the economic significance of low operating temperature, showed that


costs for direct flame incineration range from 150% to 600% of the cost of catalytic

oxidation, depending on various factors. Furthermore, a catalytic afterburner minimizes

the problem of NOx generation during disposal. Catalytic systems are, in fact, used for the

chemical reduction of these oxides (172).

1. Direct Flame Afterburner Systems

a. Types. Furnaces for removing undesirable gases or particulate matter from the exhaust

of a chemical or manufacturing process by direct flame incineration are, in general, either

similar or identical to those used for generating heat. There are many design variations in

conventional furnaces; so are there a variety of incinerator systems. The principal

differences between the two types of furnaces are obviously related to the introduction

of the effluent feed stream and the construction materials necessary to withstand special

erosive or corrosive effects.

Most direct flame afterburners utilize natural gas. Since in many cases gas is

employed as the primary fuel in the process to which the afterburner is applied, installation

may be relatively simple, and it produces a cleaner exhaust than does burning fuel oil. Gas

burner designs are conventional and may be of either the premixed or diffusion type: that

is, either the fuel and air are mixed prior to entering the furnace or they are introduced

separately. In some cases the (primary) air is premixed with the fuel and additional

(secondary) air is introduced into the combustion chamber. The method of fuel–air

injection generally defines the type and performance of the afterburner. Typical assemblies

include ring, pipe, torch, immersion, tunnel, radiant flame, and static pressure burners.

Atmospheric, or low-pressure, burners generally are of relatively simple design,

while high-pressure burner systems require either a source of high-pressure gas and air or

special equipment in the burner. Low-pressure systems tend to produce a lower combus-

tion intensity, resulting in larger combustion chambers and, in some cases, lower

efficiencies; however, their initial cost may be less when the plant does not have high-

pressure gas and air supplies.

The method of introducing the effluent feed stream into the furnace depends on its

composition and the type of gas burner; however, since it generally contains a high

percentage of air, it is usually fed into the system in essentially the same manner that air is

introduced into conventional burners.

Other significant components of the furnace system relate to methods of (1)

enhancing the mixing of fuel, air, and waste gas, (2) holding the flame in the desired

position within the chamber, (3) preventing the flame from flashing back through the waste

gas feed stream to its source, and (4) removing undesirable gases and particulate matter

either before entering or after exiting from the furnace.

The furnaces are constructed of high-temperature alloys or lined with refractory

materials. Alloys must be selected on the basis of design stress at maximum temperature

and on the known corrosive effects of the feed stream and combustion products.

Refractories offer the advantages of (1) providing insulation for reducing heat loss, (2)

radiating heat back into the chamber gases and particulate, and (3) resisting erosion and

corrosion. The application of refractories in furnaces is a well-developed art; thermal

shock, shrinkage, spalling, and deformation characteristics have been established for a

variety of refractory materials.

When economically justified, heat is recovered from the products of combustion of

afterburners, either by the addition of heat exchangers in the exhaust stream or by using a

convection (boiler) furnace as the afterburner. The heat exchangers used to recover heat


from the afterburner exhaust are of conventional design and include the many variations of

both recuperative and regenerative systems. Among the furnaces modified to be used both

as afterburners and as a source of heat for some other purpose are boilers, kilns, and

chemical reactors.

b. Design and Performance Characteristics. The effectiveness of afterburners in the

removal of the pollutant from the waste gas depends primarily on the temperatures

achieved within the afterburner, the mixing, and the residence time. For a given rate of

throughput, higher temperatures and longer residence times provide higher levels of

removal of the combustible pollutants, but require more fuel and larger combustion

chambers, thus leading to higher operating and equipment costs.

In most current systems, operating temperatures are relatively high, and so the time

for chemical reaction to take place is short compared to mixing times. Particles larger than

50 to 100 mm may require a relatively long residence time; however, they are generally

removed by techniques other than combustion.

Since mixing of the fuel and air and of burned products with the unburned materials

is important to most furnaces as well as to afterburners, considerable effort has been

expended in increasing combustion intensity and turbulence. Among the methods that are

being employed or investigated are insertion of the gases tangentially within the

combustor, high-velocity injection of the gases, multiple-ported injection of fuel and air,

baffles, recirculation of hot products into the unburned zone, and injectors that introduce

swirl. In general, the design of afterburners has not received that attention applied to

conventional furnaces, since the economic impact derived by improved design is low

compared to that attainable with improvements in, for example, boiler furnace efficiency.

In addition, the afterburner cost can be quite small compared to that for the overall

equipment and operating costs of the associated chemical or manufacturing process.

Control systems for fume incinerators are generally based on temperature. The

temperature is allowed to fall to a preset level as the heat content of the waste gas decreases

and, at the set point, auxiliary fuel burning is started. Temperature alone may be an

unsatisfactory control variable since there are two combustion situations (on either side of

stoichiometric) that result in the same temperature. Since the substoichiometric condition

is dangerous, it is prudent to incorporate sensing of flue gas oxygen content into the

instrumentation and control system. The system should automatically override the

temperature controller when an oxygen deficit is detected.

As with all other processes, the demand to decrease the cost of operating after-

burners will continue; however, we also expect that the need to reduce pollution from all

sources will require that afterburners be more effective and their use more widespread.

These factors will demand that more attention be applied to afterburner design. Obviously,

except for when waste heat recovery is employed, the primary performance criteria of

direct flame afterburners are considerably different from those for conventional furnaces:

that is, the percentage of pollutant removed per unit of fuel is one of the principal

objectives for afterburners, while for conventional furnaces it is the usable thermal energy

per unit of fuel. These differences should be carefully examined and exploited in the search

for improved afterburner systems.

2. Flares

The flare is a special case among direct afterburner systems. Flares are common in

refineries and petrochemical plants, in landfill gas disposal applications, and in process


plants where small quantities of combustible gases are generated in an irregular fashion

such that a more conventional afterburner cannot cope with the surging flow pattern. Flares

are low in cost relative to conventional afterburners and perform satisfactorily for the

disposal of nontoxic gases.

A flare is, in essence, an open pipe discharging a burning combustible gas directly

into the atmosphere. The unit can be mounted at ground level or elevated. Flares are

elevated, most particularly, where the flame, as a potentially dangerous ignition source,

must be physically isolated from a process unit. Also, elevation enhances dispersion of the

products of combustion and reduces noise, heat, smoke, thermal radiation, and objection-

able odors in the working area.

In the 1920s, the primary function of a flare was pressure relief for oil wells and

refineries where excess gas was not usable. These early flares were simple lengths of pipe

ignited by hoisting a burning rag to the tip. In most modern flares, combustion involves a

highly crafted, turbulent diffusion flame. The flare includes proprietary tip designs

incorporating features enhancing flame stability, ignition reliability, and noise suppression.

The flame retention devices can support a stable flame over a flare gas exit velocity range

from 0.5 to 175m=sec. Capacity is usually limited by the pressure available at the flare gas

source. Flare diameter is normally sized such that at maximum firing rate, the gas velocity

approximates about 50% of the sonic velocity of the gas.

a. Ignition. All flares include a continuous pilot designed for stability. The preferred pilot

ignition system uses a remote, grade-level ignition panel that mixes air and fuel,

electrically ignites the gas using high-energy capacitor-discharge igniters and directs the

resulting flame to the pilot through a 2- to 3-cm diameter pipe (491). Because the pilot

ignition system’s integrity is critical to reliable operation, the gas supply must include

condensate traps and drain valves, and the service piping between the pilots and the

ignition panels must be properly sloped.

The pilot burners are positioned around the outer perimeter of the flare tip. Pilots are

commonly monitored using a simple thermocouple with a thermowell and shroud,

although other flame detection methods are also used. Ultraviolet or infrared sensors are

problematic since they cannot readily differentiate between the flare flame and the pilot

flame. The number of pilot burners (typically burning about 2m3 of natural gas per hour)

depends on the flare tip diameter. Good practice (296) suggests the following:

Flare tip diameter (cm) No. of burners

2 to 25 1

30 to 60 2

75 to 150 3

>150 4

Flare pilots must ensure reliable ignition of waste gases flowing at high speed (as much as

270m=sec. To prevent flame-outs, flare pilots must withstand hurricane-force cross-winds

and must be easily reignited if flame failure occurs. Thus, simple retention-type nozzles are

unacceptable as their design is vulnerable to cross-winds.

b. Control of Smoking. Steam is often injected into flares to improve mixing and as an

oxidant to prevent smoking if higher-molecular-weight hydrocarbons are to be burned.

Experience shows that if the molecular weight of the hydrocarbons being burned is above


20, smoke is likely (491). Steam-to-gas ratios are increased as the molecular weight of the

gas and its degree of unsaturation increase. Typical, the mass ratio of steam to gas is in the

range 0.1 to 0.6 with a minimum flow of 150 to 200 kg=hr. For effectiveness, steam at a

pressure over 0.7 atm is necessary to ensure that sufficient momentum and mixing energy

are introduced. In refinery applications, the average steam-to-gas ratio is 0.25 kg=kg.Chemical plants flaring large quantities of unsaturated hydrocarbons may use a ratio of

0.5 kg=kg.Although steam injection improves combustion and helps to control soot formation,

it comes at a cost in energy and water treatment and increased noise (adding a high-

frequency ‘‘jet’’ noise to the crackle and rumble of the flame). If too much steam is used,

the effectiveness of steam injection decreases due to excessive cooling. Also, excessive

steam can lead to instability with associated flame pulsation and low-frequency noise

generation.

Other means to control smoke include the use of air blowers; high-velocity multi-tips

that use the kinetic energy of the waste stream for smokeless flaring; and water sprays.

With good atomization, water injection flares use a water-to-hydrocarbon mass ratio of

1 : 1. With poor atomization, the quantity of water to control smoke increases to as much as

4 : 1 or even 7 : 1 (491). This contrasts with the use of steam at ratios of 1 : 4.

c. Design Parameters

FLARES FOR INDUSTRIAL APPLICATIONS. U.S. EPA requirements for steam-assisted,

elevated flares requires

An exit velocity at the flare tip of less than 18.3m=sec for 75 kcal=m3 gas streams

and less than 120m=sec for >250 kcal=m3 gas streams. Between these extremes,

the maximum permitted velocity (Vmax in m=sec) is given by

log10ðVmaxÞ ¼Bv þ 6;890

7;580ð13Þ

where Bv is the net heating value in kcal=m3.

No visible emissions (a 5-min exception is permitted in any two consecutive hours).

Flame present whenever gases are vented. The pilot flame must be continuously

monitored.

Net heating value of the flare gas must exceed 75 kcal=m3 (either the inherent

heating value of the gas or the result of blending natural gas or other fuel gas with

the vent stream).

A windshield is usually installed around the entire flare tip to prevent flames from

licking down the stack or liftoff or blowout of the flame. The windshield for the pilot

should be at least one-third the total length of the pilot flame. Care must be given to

prevent air penetration within the flare, which could cause an explosion or lead to

burnback inside the flare tip and stack.

FLARES FOR LANDFILL GAS AND DIGESTER GAS. Mixtures of methane, carbon

dioxide, and nitrogen with small amounts of oxygen (from in-leakage), hydrogen sulfide,

and traces of other hydrocarbons are generated by anaerobic biological activity in landfills

and in the digestion of wastewater treatment plant sludge. When the quantity of generated

gas is small or if an economically attractive energy market is unavailable, a flare is often

used to safely effect destruction of the methane and the accompanying odorous species. In

this application, three principal types of flares are used: candlestick flares, enclosed flame


flares, and thermal oxidizers. The advantages and disadvantages of the three types (493)

are shown in Table 1.

The candlestick flare is distinguished by its exposed flame. Its simple design (no

combustion chamber or ignition tube) keeps capital investment low but at the price of poor

performance. Indeed, the high-emission profile often makes this design inadequate to meet

air pollution regulations. Measured destruction efficiencies range from only 60% to 75%.

Enclosed flame flares consist of a vertical combustion chamber with forced air

addition. Ceramic blanket insulation maintains high temperatures in the combustion

chamber and lowers the skin temperature. Well-designed units can achieve hydrocarbon

destruction rates of up to 99.9% and low (<200 ppm) CO and NOx (493).

d. Radiation from Flares. Flares are often elevated in order to minimize the intensity of

heat radiation in the nearby area. A generally accepted thermal intensity for continuous

work is about 1350 kcal=hr per m2. Equation (14), after Hajek and Ludwig, can be used to

establish the minimum distance ‘‘L’’ (m) from the center of the flare flame to the point

where the target thermal intensity is to be maintained (297). Estimates of heat radiation

from flames are presented in Table 2.

L ¼ffiffiffiffiffiffiffiffiffitf R4pK

rð14Þ

where

t¼ fraction of heat intensity transmitted (assume¼ 1)

f ¼ fraction of heat radiated (from Table 2 or a typical value of 0.2)

R¼ net heat release (kcal=hr)K ¼ allowable thermal intensity (1350 kcal=hr per m2)

In assessing the importance of radiant heat emission, one must address the issue of

deciding what maximum heat intensity is acceptable. Equipment is often allowed to

experience a higher intensity than people, and plant workers are often allowed to

experience a higher intensity than the general public. Also, the use of protective clothing

permits exposure to higher intensities; as much as 4.73 kW=m2 for several minutes (494,

Table 1 Advantages and Disadvantages of Flare Designs

Flare design type Advantages Disadvantages

Candlestick flare Low initial cost High emissions

Little control, turndown

Minimal safety controls

Restricted to high heat content fumes

Enclosed flame flare High destruction rates High initial cost

Low CO and NOx emissions Cannot handle heavier fume streams

Safety controls

Thermal oxidizer High destruction rates Higher initial cost

Low CO and NOx emissions Equipment weight

Safety controls

Wide range of fume heat content


495). This is sufficient for many situations where short-term action is needed within an

industrial area or when the duration of the flaring activity is limited.

The establishment of a design maximum radiant intensity should also recognize that

solar radiation adds to the radiative exposure. A correction of 0.47 to 1.10 kW=m2 to

account for solar radiation is, therefore, appropriate. Finally, the general heat stress load

carried by the industrial worker should be taken into account. If the general working area is

hot and humid, the energy rejection system of the worker may already be heavily burdened

and, thus, a reduction in the permissible additional load from radiation should be

considered.

e. Flare Destruction Efficiency. Flare destruction efficiency (typically > 98%) is affected

by several factors relating to the gas being burned and to the environment. Good

performance is favored by wide flammability limits (the range of stoichiometric composi-

tions that will produce a stable flame without a continuous pilot). High heating value gas

(high flame temperature with consequently enhanced combustion kinetics and buoyant

mixing) is also preferable.

The environmental problems commonly associated with flares include noise

(rumbling, steam jet ‘‘scream,’’ or crackling noises) and adverse reactions to the

luminosity.

3. Catalytic Afterburner Systems

a. Types. In most cases, the chief aim in the design and operation of the catalytic

afterburner is to keep the input of additional energy as low as possible. In Fig. 2(a), the

sole purpose of the catalytic unit is to purify the polluted air by passing it, if necessary after

preheating, over the catalyst into the atmosphere. In Fig. 2(b), the heat content of the

Table 2 Heat Radiation from Various Gaseous

Diffusion Flames

Gas

Flare tip

diameter (cm)

Fraction of heat

radiated ( f )

Hydrogen <2.5 0.10

4.0 0.11

8.4 0.16

20.3 0.15

40.5 0.17

Butane <2.5 0.29

4.0 0.29

8.4 0.29

20.3 0.28

40.5 0.30

Methane <2.5 0.16

4.0 0.16

8.4 0.15

Natural gas 20.3 0.19

40.5 0.23

Source: From (297).


purified air is used to preheat the waste gas not yet purified by means of a heat exchanger.

In Fig. 2(c), the heat content still remaining after this stage is utilized as heating energy for

the original process emitting the waste gas. This scheme is applicable to the processes

where the waste gas carries a large fraction of the total energy consumed. In the cases

where the concentration of burnable material is very high, the two-stage afterburner

scheme, as shown in Fig. 2(d), will allow the heat exchange to be operated at lower

temperatures. Further recovery of the heat content of the waste gases is easily carried out in

a subsequent waste-heat boiler. When the concentrations of organic materials in the waste

gases are still higher, imbedding cooling surfaces in the catalyst bed is preferable since

most of the heat of reaction can be removed close to the point of generation, thus keeping

the maximum temperature in the reactor low and avoiding material selection problems.

b. Catalysts. Many substances exhibit catalytic properties, but metals in the platinum

family are conventionally used because of their ability to produce the lowest ignition

Figure 2 Types of catalytic afterburners: (a) straightforward catalytic afterburning of waste gas;

(b) catalytic afterburner coupled with heat exchanger; (c) catalytic afterburner coupled with heat

exchanger and hot-air recycling; (d) two-stage afterburners with heat exchanger; (e) catalytic

afterburner with cooling surfaces imbedded in the catalyst, coupled with heat exchanger.


temperatures. Others are copper chromite and the oxides of copper, chromium, manganese,

nickel, and cobalt.

In industrial use, the active component of the catalyst is usually deposited on a

carrier material, as in the following combinations, for example:

Active metal=metallic carrier (e.g., platinum deposited on stainless steel or nickel

alloy ribbon)

Active oxide=oxide carrier (e.g., copper=manganese oxide deposited on a-alumina

or cobalt oxide on t-alumina)

Active metal=oxide carrier (e.g., platinum and g-alumina)

The following are some of the important characteristics of catalysts:

Catalyst composition

Catalyst total surface area

Surface areas of the active components

Pore volume

Pore size distribution

Initial catalytic activity toward selected reaction

Compression strength

Surface area decrease upon use

Change in pore volume=pore size distribution upon use

Change in catalytic activity upon use

Change in catalyst composition upon use

The catalyst will gradually lose activity through fouling and erosion of the catalyst surface,

so that occasional cleaning is required, and eventually it must be replaced. Common

fouling agents include alumina and silica dusts, iron oxides, and silicones.

If excessively high temperatures are experienced by the catalyst bed, sintering of the

catalyst reduces the concentration of catalyst at the surface and=or the available surface

area. The temperature range where such degradation occurs is catalyst-specific but often

begins above 800� to 980�C.The susceptibility of catalysts to poisons varies, but the following occurrences are

common, especially for platinum-family catalysts:

Metallic vapors such as mercury and zinc deactivate the catalyst.

Phosphorus oxidizes to phosphoric acids, which deposit, on cooling, on the catalyst

surface.

Sulfur-containing compounds form sulfate and coat the catalyst surface.

For chlorine-containing compounds, which present no problem under normal

operating conditions, temporary high temperatures cause the chlorides formed

with the active material to sublime from the carrier.

Compounds containing lead, bismuth, arsenic, antimony, and other heavy metals

pass through as aerosols and cover up the effective surface area of the catalyst.

For some catalysts, ‘‘reversible poisoning’’ or inhibition occurs through deactivation

after absorption of SO2 or HCl. This problem can be removed by treatment or, if

anticipated, through special catalyst design.

c. Performance Potential. Catalytic oxidation could economically dispose of all combus-

tible fumes that are free of appreciable amounts of unburnable solids, which tend to foul


the catalyst bed. However, the catalytic afterburners are most suited for the removal of the

components that will vaporize, because of their greater mass and lack of significant

Brownian movement, liquid droplets are not likely to contact the catalytic surface to any

appreciable degree. For the catalytic burner to operate effectively, the waste stream must

not contain materials that will poison the selected catalytic system.

To achieve destruction efficiencies of 90% to 95%, about 1.5 to 2 liters of catalyst is

required per standard cubic meter per minute (exhaust stream plus supplementary fuel

combustion products). Since the catalytic reaction rate is much greater than the thermal

processes, the mean residence time required to achieve a target degree of volatile organic

compound (VOC) destruction is correspondingly smaller. For example, when burning

hexane with a 95% destruction target, the calculated catalytic system residence time

required at 480�C is only about 25% of that for a thermal oxidizer at 650�F.Catalytic unit destruction potential is often estimated using the space velocity. Space

velocity is defined as the volumetric flow rate of the combined gas stream (emission stream

plus supplemental fuel plus combustion air) entering the catalyst bed divided by the

volume of the bed. As such, space velocity depends on the type of catalyst used. For many

organic pollutants, destruction efficiencies of about 95% can be achieved with precious

metal catalysts at space velocities of 30,000 to 40,000 hr�1. Base metal catalysts achieve

similar performance at 10,000 to 15,000 hr�1 (934). Greater catalyst volumes and=orhigher temperatures are required to achieve higher destruction efficiencies.

Catalytic units can be particularly useful for small air flows. Catalytic units for flows

as small as 3m=min may be only a fraction of the cost associated with installing and

maintaining ductwork and piping to connect such small sources to existing abatement

equipment. Processes emitting large volumes of air where the VOC content is less than

25% of the lower explosive limit and where the VOC content tends to be consistent and

predictable are especially well suited to catalytic oxidation because the lower operating

temperature requires significantly less fuel and the smaller unit size requires much less

startup and cooldown time than thermal oxidation systems.

Catalytic units can often economically achieve destruction objectives for organic

compounds but, at the same time, can create other problems. For example, high

concentrations of methyl ethyl ketone can be reduced by, say, 98% and thus meet VOC

destruction requirements. However, for many catalyst combinations, the oxidation

proceeds by a mechanism that generates low concentrations of a highly odorous aldehyde

as a product of incomplete combustion that can cause significant odor nuisance problems.

The potential for inadvertent operational consequences of this type should be carefully

considered in selecting a catalytic unit and in the preparation of purchase specifications for

this type of VOC control device.

C. Potential Applications

The emission of combustible pollutants, either gaseous or particulate, is common to a wide

variety of chemical and manufacturing processes. The release of these materials may be

constant, intermittent, or cyclical in volume, temperature, and degree of contamination.

The effluent may consist entirely of compounds of one specific type, although they are

usually heterogeneous. In some cases, secondary pollutants, such as sulfur dioxide,

nitrogen oxides, and hydrogen chloride, may be emitted following incineration. Under

such circumstances, secondary effluent treatment methods such as alkalized wet scrubbers


may be desirable. Table 3 lists processes that are typical sources of organic gases and

particulate.

The list is by no means inclusive but suggests the wide variety of sources that can be

considered for afterburner-type pollution control. Quantification of the characteristics of

the effluent from these processes for afterburner control will be made more difficult by the

highly variable concentration due to dilution in the fume collection system. The informa-

tion regarding composition of these pollutant streams as they relate to afterburners would

include consideration of inlet temperature, concentration, toxicity, secondary pollutant

formation following combustion, photochemical reactivity, heating value, nuisance value,

and the degree to which the species may inhibit or enhance the combustion or catalysis

processes.

Catalytic incinerators are inherently higher in capital cost than both direct thermal

oxidizers and most recuperative oxidizers. However, because of significantly reduced fuel

expense, catalytic systems are usually lower in life cycle cost, although consideration

should be given to the continuing capital investment for catalyst replacement that, in many

cases, arises biannually. Regenerative thermal oxidizers have high initial capital costs, but

fuel use is usually the lowest of any of the afterburner alternatives. Also, they are prone to

plugging if the gases passed to the RTO have a significant particulate loading.

III. OPERATIONS AND SAFETY

Thermal oxidizers for liquids and gases are useful means of disposing of combustible

organic wastes. However, safe operation of these units is not inherent, and safety

considerations must be incorporated into both design features and operating practices.

Control systems must ensure personnel, equipment, and environmental safety with

minimum supervision under both normal and upset operation conditions.

In liquid- or gas-burning incinerators, the combustion temperature should be slightly

above the theoretical flame temperature of the waste stream’s lower flammability limit

(LFL). If necessary, auxiliary fuel should be burned to maintain this temperature goal. If

noncombustible gases or dilute aqueous wastes are to be burned, they should be introduced

downstream of the principal flame into a well-developed, high-temperature region.

Table 3 Applications for Incinerators Burning Gaseous Wastes

Industrial dryers Organic chemical production

Food product ovens Synthetic rubber manufacturing

Solid waste incineration Asphalt processing

Coke ovens Fat rendering

Enamel baking furnaces Fat frying of foods

Foundry core ovens Petroleum refining

Paper coating and impregnation equipment Gasoline distribution

Paint, varnish, and lacquer manufacture=use Degreasing

Dry cleaning Resin curing ovens

Coffee roasting Electrode curing ovens

Grain milling Sewage sludge drying

Carbon black manufacture Refuse composting


In some instances, combustible gas analyzers can be used to determine if flammable

vapor–air mixtures are outside their flammable limits. National Fire Protection Association

(NFPA) documents NFPA - 86 (498) and NFPA - 69 (499) do not require a flammability

control system for streams up to 25% of the LFL. However, if flammability is controlled by

either addition of dilution air (to keep below the LFL) or natural gas addition (to keep

above the upper flammability limit, or UFL), the stream should incorporate such

instrumentation. In that case, the maximum operating point is set at 50% of the LFL

with an initial alarm at 55% LFL and automatic shutdown at 60%.

For systems burning gaseous wastes, flame and detonation arresters are necessary to

quench flashback, prevent sustained combustion, and halt flame propagation. Detonation

arresters with automatic shutoff and explosion relief should be installed upstream of the

incinerator to isolate flammable mixtures from potential ignition sources (497). The types

of devices used include dry flame arresters, liquid flame arresters, and detonation arresters.

Safety interlocks recommended (497) to trigger shutdown and critical alarms of the

system are summarized in Table 4. All such interlocks and their supporting instrumenta-

tion and control logic should fail in a clearly defined safe position in the event of a power

failure. If possible, gaseous waste streams should be bypassed to a vent system with

appropriate high temperatures and residence times. The instruments should be kept simple

but accurate, responding automatically to system and operator failures as well as

significant upset conditions.

Table 4 Safety Interlocks for Incinerators for Waste Liquids and Gases

Interlocks triggering shutdown for incineration system

Power failure or loss of instrumentation air

Failure to recognize flame by flame safety system

High or low fuel-oil or natural gas pressure

Low combustion air pressure

Low atomizing-media (steam, air, mechanical pressure) pressure or flow

High oxidizer chamber temperature

Low quench water or recycle flow

High quench temperature

Low water levels or abnormal steam pressure in any waste heat boiler

Interlocks Triggering Alarms and Eventually Shutdown for Incineration System

Low oxidizer chamber temperature

Low excess oxygen for combustion

High CO, stack particulates, or emitted pollutants (e.g., SO2, NOx)

pH of scrubbing liquid out of range

Abnormal waste stream pressures

Source: From (497).


incineration systems for liquid and gaseous wastesdl.mozh.org/up/dk1860_10.pdfviscosity=density) of...

Documents