3.3 conventional thermal treatment methods 3.3.1 overview ... · various processes which integrate...

OVERVIEW ON EXISTING THERMAL PROCESSES BECKMANN, M.; SCHOLZ, R. PAGE 1

3.3 Conventional Thermal Treatment Methods

3.3.1 Overview of Existing Thermal Processes

3.3.1.1 Introduction

Various processes which integrate pyrolysis, gasification and combustion funda-mental structural units are currently being applied and tested in the field of the ther-mal treatment of municipal waste and similar industrial waste. The main thermal pro-cesses can be broken down into a

• first unit for the conversion of the solid and pasty waste and a

• second unit for the treatment of the gas, flue dust or pyrolysis coke pro-duced in the first unit (Fig. 1).

The conventional thermaltreatment of residual waste,also often called „classicalcombustion of residualwaste“, represents a triedand true technology. Thesystematic classification ofprocesses is necessary inorder to improve the discus-sion and optimization ofprocessing technologicalpossibilities and to enable acomparison of the processesin the first place. First, theprocesses for thermal wastetreatment can be classifiedinto the so-called main ther-

mal processes and processes or plants for flue gas purification, energy conversion,ash treatment, production of supplementary agents, etc. The classical waste com-bustion can be viewed as a combustion-post-combustion process with a grate sys-tem in the first unit and a combustion chamber system in the second unit. Accordingto the classification in Fig. 1 the Fig. 2 shows the profile of a complete process as anexample.

Unit 1 Unit 2 processes and examples

A. combustion1) combustion combustion - post-combustion-process

(e.g. standard waste incineration) [2] to [9], [16]

B. thermolysis2) combustion thermolysis - post-combustion-process

(e.g. Schwel-Brenn-Verfahren by Siemens KWU) [11]

C. gasification3) combustion gasification - post-combustion-process

(advanced standard waste incineration) [10], [12]

D. thermolysis gasification thermolysis - post-gasification-process

(e.g. Konversionsverfahren by NOELL [13], Thermoselect-Verfahren [14] etc.)

E. gasification gasification gasification - post-gasification-process

(e.g. gasification and gas decomposition by LURGI [15])

1) here: includes the processes drying, degasification, gasification and combustion

2) here: includes the processes drying, degasification and pyrolysis

3) here: includes the processes drying, degasification and gasification

Fig. 1. Systematic description of main thermalprocesses [1].

Pöschl

Textfeld

Beckmann, M.; Scholz, R.: Conventional Thermal Treatment Methods. In: Ludwig, Ch.; Hellweg, S.; Stucki, S.: Municipal Solid Waste Management – Strategies and Technologies for Sustainable Solutions. Springer-Verlag, 2002, S. 78-101. ISBN 3-540-44100-X


The flue gas purification (so-called secondary measures) of the „classical“ processin particular has been improved in the last years. Plants equipped with state-of-the-art technology meet the statutory specified limits for the emission or dumping of pol-lutants into the air, water and soil.

The current priority is the development, modeling and optimization of the processcontrol (so-called primary measures) of the main thermal process. A considerablepotential for development is present in the area of the grate and post-combustion inorder to, for example,

• reduce the flue gas flow (flue gas purification plants, emission loads),

• improve the energy utilization and

• influence the characteristics of the residual material.

Considerable progress, triggered through the „new“ processes, have been made inthe development with regard to the „classical“ processes with grate systems. Theoptimization of the design of the combustion chamber ([2], [3], [7], [16]), flue gas re-circulation ([4], [5]), enrichment of the primary air with oxygen [6], water-cooled grateelements [7], further development of the control (e.g. IR-camera, [8], [9]), etc. shouldbe mentioned as examples.

thermal main process flue gas cleaning

10

4

6

5

32

1

9

11

7B 031 G0604.1997 Sami

SAUERSTOFF

1312

15

1416

17

18 19

20

8

7 21

combustion(grate)

Post-combustion boiler

legend

1 delivery 2 bulky refuse crusher 3 waste bunker 4 grab crane 5 charging hopper 6 reverse acting grate 7 charging equipment

8 primary air 9 ash discharge (wet)10 ash bunker11 secondary air12 spray absorber13 fabric filter14 sound absorber

15 induced draught - blower16 Venturi - scrubber17 radial flow - scrubber 18 wet - electric filter19 clean gas reheating20 analytical room (emission measurement)21 chimney

technical datapro unit:waste inputlower calorific valueproduced steamsteam pressure (boiler)steam pressure (superheator)superheated steam temp.feed-water temperature

=======

10,0 Mg/h7610 kJ/kg24,8 Mg/h54 bar40 bar400 C130 C

o

o

Fig. 2. Schematic representation of a waste power plant (classical household wastecombustion plant) [2].


In addition, new, altered process control with grate systems such as e.g.

• the gasification with air on the grate and the therewith connected

• independent post-combustion of the gases produced

is being tested on the pilot plant scale for future developments [10] and recently alsoin small industrial scale [43]. The desired improvements toward the above-mentioned objectives are then even more pronounced.

In the following section, the general points of the process control in grate systemswill be illustrated briefly. Then examples of process control in plants implementedpractically and results of experiments in industrial and pilot plants as well as on thelaboratory-scale will be discussed.

3.3.1.2 Process control in grate systems

The efficiency of the thermal process for the treatment of waste is primarily deter-mined through the process control of the main thermal process. The possibilitiesavailable for the reduction of pollutants and flue gas flows through primary measuresmust be exhausted. The expenditure necessary in the range of secondary measures,e.g. the flue gas purification, is then adjusted accordingly. Based on this, we will fo-cus on the thermal treatment of the classical waste combustion which can, as men-

tioned at the begin-ning, be divided intotwo units according tothe combustion-post-combustion concept.Combustion on thegrate takes place inthe first unit. Then, thegases and flue dustarising from the grateare combusted in thepost-combustionchamber in the sec-ond unit. Due to the

different tasks of the first and second units, the steps of solid conversion on thegrate (first unit) and post-combustion of the gases (second unit) will from now on betreated separately.

Each reaction condition is determined through the main influencing parameters (Fig.3). When discussing the partial steps of the solid conversion and the combustion of

Fig. 3. Main influential parameter [1].


the gases generated, the main influencing parameters mentioned in Fig. 3 (e.g. [1])must be considered with respect to the level and control possibilities through planttechnology (apparatus technology). This of particular value when regarding the indi-vidual reaction mechanisms (e.g. formation and degradation of pollutants). The vari-ous control possibilities along the reaction pathway are of particular importance forthe optimization of the process.

Solid conversion in grate systems

The solid conversion on a grate can more or less be divided into the following partialprocesses in the direction of the reaction pathway (Fig. 4):

• drying,

• degassing,

• gasification and

• residue burn out.

Since the oxygen supplied to thereaction gas of the grate (usuallyair) relative to the waste addedleads to overstoichiometric con-ditions (e.g. λ ≈ 1.3), it can beconsidered as „combustion on thegrate“.

In order to be able to discussprocess-engineering possibilities,the main influencing parametersmust be regarded [1]. Not onlythe actual level but also the distribution of the main influencing parameters along thereaction pathway must be considered. There are many different possibilities for thecontrol of the above-mentioned main influencing parameters in grate systems whichcan selectively influence individual partial processes along the reaction pathway(Fig. 5). These control possibilities are particularly advantageous when dealing withvariations in the reaction behavior due to fluctuating composition of the waste. Thefollowing representations of the process-engineering possibilities are also depictedin the overview Fig. 5 and the flowchart in Fig. 6.1

1 In this case, the main focus is on the process control for solid input material in grate systems. Whenconsidering the process control possibilities, the incineration of gaseous, liquid and dust-like feedmaterial must first be discussed since generally only one unit is necessary for these. Then, using this

material(refuse)+additive

combustion gas

Drying,Degasification

IgnitionGasification

burnout ofsolid matter

CO,CO ,CHH O,H ,N

2 X

2 2 2

CO, H CHH O,CO ,N

2 4

2 2 2

C+CO 2 CO2 →

CO,CO ,N2 2

CO ,O ,N

2

2 2

CO ,O ,N

2

2 2

CO+½O CO2 2→C+½O CO2 →

λtot < 1

BUW / Be / E-FU-VGAS.cdr / 03.07.00

CO+H O CO +HCO+½O CO

2 2 2

2 2

→ →

reaction gas(air, recycled flue gas etc.)

remnants(ash, valuable substances)

grate- division into zones along the grate path control of + concentration + temperature + residence time- grate elements (forward or reverse acting etc.) + residence time distribution + reactor behaviour

C H O NS W A

C A

A

Fig. 4. Schematic representation of the solidconversion in grate systems during thecombustion process.


It should be mentioned beforehand that the quench reactions at the surface of thewall should beavoided and there-fore a correspondinghigh wall tempera-ture through

suitable refractorylining should beguaranteed for bothunits (grate andpost-combustion).This also means thatthe heat transfermust be carried outseparately from thesolid conversion andthe post-combustionas far as possible,i.e. combustion andheat decoupling areto be connected inseries as shown inFig. 6. It can also beseen in Fig. 6 that

the two combustion units are drawn in series and separate.

This should make it clear that the different partial steps

• solid conversion on the grate and

• post-combustion process

can be optimized more easily, the more decoupled the two units are. In order toachieve this, not only a geometric separation is possible, but also, as will be ex-plained in connection with the post-combustion process, a fluid dynamic decoupling.

information as a basis, one can shift to the process control for lumpy and pasty materials which gen-erally require two units. Since the systematic representation would take up too much space, the pro-cedure and references are mentioned here (e.g. [1], [4], [45]).

Starting materialsLumpy, also pasty when in connection with a solid or inert bed

Amount of oxygen availableLevel Usually overstoichiometric (combustion), understoichiometric (gasification) possible; making

independent post-combustion possible; oxygen exclusion (pyrolysis) not customaryControl along thereaction pathway

Very well adjustable (e.g. air/oxygen gradation, flue gas recirculation, etc.) when separatedinto individual zones; the partial steps of drying, degassing, burn out of the solid can beinfluenced in connection with temperature control

TemperatureLevel Bed surface temperature up to ca. 1000 °C and higher; average bed temperature lowerControl along thereaction pathway

Very good possibilities through separation into several zones as with the control of the oxygenconcentration (preheating of the air, flue gas recirculation, water/vapor cooling)

PressureAt standard pressure, generally only a few Pascals underpressure due to the plant construc-tion

Reactor behaviorSolid According to the movement of the grate elements, the individual zones can approach continu-

ously stirred reactor characteristic (e.g. reverse-acting grate) of a plug flow reactor character-istic (e.g. travelling grate); over the entire reactor length, a plug flow reactor characteristicresults

Gas a) Oxidizing agent etc. is forced to flow through the bed and is distributed evenly over the bedsurface, resulting in very good contact between gas and solidb) Flow control over the bed possible as counter and parallel flow, gas treatment in thefollowing process step necessary (e.g. post-combustion)

Residence timeLevel (averageresidence time)

In the range of several minutes to hours; adjustable through grate speed and mass flow andcan be influenced in the project design through total length and width

Control along thereaction pathway

Very good adaptation possible through separate speed regulation of the grate elements in theindividual zones; control through discharge roller if necessary for additional improvement ofthe burn out at the end of the grate

AdditivesAdditives for the binding of pollutants in the solid and influence of the characteristics of theresidue (ash, partly fused ash, slag); inert bed e.g. matrix for possible easily melted sub-stances (e.g. plastic)

Functional range (examples)For the solid conversion in the first step in household waste combustion plants; separation ofmetals from the composite at low temperatures and simultaneous understoichiometricconditions

Fig. 5. Characterization of grate systems [44].


The division of the grate area into several grate zones which are separate from oneanother with respect to the reaction gas supply must also be mentioned when evalu-ating the control possibilities of grate systems. This makes it possible to control theamount of oxygen available and the temperature in the bed along the reaction path-way through variation of the supply of reaction gas so that the individual steps, end-ing with burn out of the residue, can be matched to suit the characteristics of thestarting material, even with fluctuating composition.

An additional leewaywith regard to distri-bution of oxygen andtemperature isachieved when inertgas, recycled fluegas or additionaloxygen is given asreaction gas to cer-tain waste materialsinstead of air (down-draft). A preheatingof the reaction gascan also be consid-ered for waste mate-

rials with low heating values.

Through the variation of the distribution, the oxygen content and the temperature ofthe reaction gas, the burn out of residues from waste with varying heating valuesand mass flows in particular can be optimized on the one hand. On the other hand,the temperatures and residence times in the combustion chamber can be influenceddirectly over the bed with respect to the post-combustion process.

When considering the control of the gas flow over the bed, one can differentiatebetween parallel, center or counter flow principles. In the parallel flow principle, afirst post-combustion of the gases emitted from the beginning of the grate occurs. Inthe counter flow principle, the hot gases from the end of the grate in the drying anddegassing zone should convey heat to the beginning of the grate. The gases, flow-ing parallel or counter, can be post-combusted optimally independent of their condi-tion at the post-combustion chamber if the separated process control of the grate (1st

unit) and the post-combustion process (2nd unit) is available. This makes it possible

gasification (solid)stoker systemwaste

added fuelif necessary

if necessaryadded material

(additive)

reaction gas(air)

reaction gas(air)

heattransfer

gascleaning

λ 0,4~~λ 1,3~~

λ <1 λ 1≥

flue gas recirculation

λ1>< 1 λ3 > 1λ2 < 1

PFRCSRpost-combustion (gas/dust)

"hot" wall "hot" wall

*)

***)

*) offtake solid residues not shown

**)

***)

**) coarse and pasty waste***) gaseous, liquid and finely divided waste

exhaust gas

unit 1 unit 2

S

S,b S,e

tot

Fig. 6. Schematic representation of the separated processcontrol; combustion/gasification-post-combustion pro-cess with grate and combustion chamber system [4],[45].


to implement common primary measures for the reduction of NOX with a simultane-ous high burn out. Similarly, the temperature conditions in the drying and degassingstages are not only dependent upon the flow control. The heat transfer ratios can beinfluenced through the so-called secondary heating surface effect, i.e. though radia-tion exchange between the surrounding refractory-lined hot walls and the surface ofthe material. These things are considered in order to guarantee that no certain flowcontrol is preferential from the start.

An additional degree of freedom in the optimization of the course of combustion (ac-cording to the residence time and residence time behavior) is gained through theseparation of the movement of the grate elements into zones. The separated move-ment of the grate elements into the grate zones along the path of the grate means,on the one hand, an increased construction expenditure but on the other hand, theresidence time and bed height can be controlled independently in the individual re-action zones. The grate itself can be used as forward-acting, reverse-acting or rollergrate.

Through an additional water-cooling of the grate elements, which makes the con-struction more complex than the conventional grate, the reaction gas distribution,oxygen concentration and temperature can be varied over a wider range regardlessof the grate construction (cooling) in contrast to the grate cooled only with reactiongas (air). The above-mentioned measures and accompanying advantages can betransferred even more distinctly. The excess-air coefficient, the flue gas and fluedust flows can be reduced. The temperature in the burn out stage can be increasedto the point that the ash accumulates in a sintered state, which leads to a markedimprovement of the elution behavior of the residual waste [14], [36]. Furthermore, thewater-cooled grate results in a lower risk of wear through thermal influences forwaste materials with high heating values and a low ash sifting through the grate inparticular (low thermal expansion permits little clearance between the grate ele-ments).

The control of the combustion process on the grate through the use of an infraredcamera (IR camera) for the detection of the temperature at the bed surface was alsomentioned. The position of the main reaction zone can be identified through themeasurement of temperature fields. This zone can be shifted toward the beginningor end of the grate e.g. through the redistribution of the primary air. Through selec-tive change of the primary air (reaction gas input), a fast adjustment to fluctuations inthe waste composition, an avoidance of gas blow through and streaming, an im-provement of the residue burn out and the avoidance of emission peaks can be ex-pected.


Post-combustion process

In connection with the post-combustion process, the main influencing parameters(Fig. 3) and corresponding control possibilities should be discussed in the same wayas before for the solid conversion process. An overview is shown in Fig. 7.

First, the residence time in connection with the formation and decomposition reac-tions of pollutants is considered. In chemical process-engineering one differentiatesbetween two limiting cases with regard to the residence time behavior: the ideal con-tinuously stirred reactor (CSR) and the plug flow reactor (PFR) [20]. Fig. 8 shows thedecomposition of CO dependent upon the residence time for these two reactor char-acteristics. It can be seen that the plug flow reactor exhibits the best reactor char-acteristics for the decomposition. It is thereby assumed that the temperature and

concentration pro-files are constantthrough the cross-section of the reactorcross-section. Forthe conditions in thepost-combustionprocess, this meansthat stirred reactorsmust be connectedto ensure mixing(Fig. 6, second unit,first stage).

It should again bementioned here thatan amplified heatdecoupling com-bined with a quencheffect at a cold wallis to be avoided inorder to guarantee a

uniform temperature profile. Mechanical components can generally not be appliedfor mixing due to the high temperature, corrosion, etc. Two fluid dynamic stirringmechanisms can be used:

• so-called supercritical swirl flow with linear recirculation or recirculation ofhot gases ([22], [23]) for plants with low capacity or

Starting materialsGaseous, liquid, dust-like

Amount of oxygen availableLevel Overstoichiometric to understoichiometric; variable over wide ranges; if overstoichiometric at

reactor discharge: called “combustion chamber”; if understoichiometric at the end: called“gasification reactor”


Very good through gradation of oxidizing agent and fuel along the reaction pathway (intro-duced over stirred reactor elements)

TemperatureLevel Different combustion temperatures in the range from 1000 °C to 2000 °C or more; range is

very variableControl along thereaction pathway

In addition to the staging of oxidizing agent and fuel along the reaction pathway, interventionthrough flue gas recirculation, spraying of water etc. possible; indirect heat coupling anddecoupling through corresponding heating or cooling systems; many possibilities

PressureAt standard pressure, generally only a few Pascals underpressure due to the plant construc-tion; high-pressure combustion rare; pressure gasification more common

Reactor behaviorDust/gas Hydraulically, stirred reactor as well as plug flow characteristics can be approached for dust

and gas

Residence timeLevel (averageresidence time)

In the range of seconds (longer at higher pressure); adjustable through load conditions andcan be influenced in the project design through geometric dimensions


Very difficult; residence time distribution can be controlled over the reactor behavior

AdditivesAdditives, in particular, introduced over the stirred reactor elements, fin order to bind pollut-ants (e.g. sulfur dioxide, nitroxides) as well as to influence the slag characteristics and meltingpoints of the dust

Functional range (examples)Combustion of liquid residues; post-combustion of gas and dusts in the second step of thethermal treatment process; high temperature gasification of residues for the production ofprocess gas (low and high temperatures); certain combustion processes (e.g. recirculation ofchlorine as hydrochloric acid in the production cycle etc.)

Fig. 7. Characterization of combustion chamber systems [44].


• single or multiple jets arranged above or next to each other which act as in-jectors to draw in and mix the surrounding ([4], [24]).

The installation of such an intense mixing zone (stirred reactor element) at the en-trance of the post-combustion leads to the above-mentioned hydraulic separation asrepresented in Fig. 6. This is supported when the mixing zone is arranged in a nar-rowed cross-section. A relaxation zone (plug flow reactor as suitable reactor char-acteristic for the decomposition) is situated above the mixing zone.

A high intensity in the mixing zone must be achieved in order to guarantee thegreatest possible burn out (CO, hydrocarbons, flue dust). Plants with greater capaci-ties are generally equipped with injector jets. A good cross-sectional overlap, a suffi-cient injector jet penetration depth and an adequate suction of the residual current(flue gas flow from the grate) must be achieved through the proper layout of the jetgeometry, the number andarrangement of the jets aswell as the inlet impulse.

Measures for the control ofthe amount of oxygen avail-able along the reaction path-way are, for example:

• air staging,

• fuel staging,

• flue gas recirculation,

• oxygen supply

• etc.

When considering the controlof the above-mentioned maininfluencing parameters, sufficient possibilities are available to optimize the processcontrol with respect to the mechanisms of formation and decomposition of pollutants.At this point, the extensive experience and knowledge concerning the reduction ofnitroxides and the improvement of the burn out of gaseous, liquid and dust-like fuelsin particular should be mentioned. This experience should be transferred to the post-combustion.

0,00001

0,0001

0,001

0,01

0,1

1

0 0,5 1 1,5 2 2,5

mean residence time τ [s]

deco

mpo

sitio

n ra

te c

/c0

[%]

temperature T = 1100 Koxygen concentration ψ O2 = 2 Vol.-%water concentration ψ H2O = 18 Vol.-%

CSRPe = 0

PFRPe = ∞

real reactorPe = 10

Pe w LD

= ⋅

Fig. 8. Dependence of the CO decomposition on theaverage residence time and the mixing condi-tions in the reactor [1], [21].


3.3.1.3 Examples and Results

The explanations concerning the main influencing parameters and the principle con-trol possibilities through the process control make the discussion of already existingplant concepts possible while simultaneously making the potential for developmentclear. The discussion of the individual aspects of the process control without refer-ence to the main influencing parameters, for example in the form of questions ifcounter or parallel flow, forward-acting or reverse-acting grate are more suitable forwaste combustion, appears to be inappropriate due to the complexity of the proc-esses. It should be mentioned that e.g. plant conceptions with a forward-acting grateand those with a reverse-acting grate can both achieve good practical results. Thefollowing discussion of the effect of certain process controls is important to under-stand the processes taking place during waste combustion. However, no estimationof the total concept of the plant can be derived from the evaluation of the individualaspects.

The solid conversion on the grate will be covered in the first section and then thepost-combustion process.

Solid conversion on the grate

The main goals of the solid conversion on the grate are the utilization of the ash in ahigh burn out (low residual carbon content) and a small as possible (available) con-centration of heavy metals and salts.

These objectives are influenced, aside from the composition of the waste, mainlythrough the main influencing parameters temperature, oxygen concentration andresidence time as well as residence time behavior or mixing behavior. A circulationof „ignition cores“ ([8], [46]) and a stabilization of the temperature in the bed isachieved through the intensive mixing of the bed in the reverse-acting grate systemwith respect to the ignition of the bed in particular. Due to the early ignition of thebed in reverse-acting grate systems, the addition of primary air generally alreadyreaches its maximum in the second primary air zone. The maximal temperature inthe bed is also reached in this section. Reverse-acting grates can be approximatedin the individual sections as stirred reactor elements with respect to the residencetime behavior ([32], [47]). A forward-acting grate system can be considered as aplug-flow reactor with reference to the mixing behavior. The input of the primary airinto the drying and ignition phase must be regulated carefully in order to avoid abreak in the ignition front. The main air is generally added in the middle of the grate.The primary air in the burn out zone must be adjusted carefully for a high burn outregardless of the type of grate. A too high primary air current in the burn out zonecan lead to a „blow cold“ (extinguish) of the bed and a corresponding high carbon


content in the ash. The influence of the amount of inert material (ash) in the fuel onthe burn-out will be considered more closely. The inert material acts as „heat reser-voir“ which takes up heat corresponding to its capacity in the main combustion zoneat high temperatures. Due to this heat reservoir, the threat of „blow cold“ in the burn-out zone is reduced. This positive influence of the inert material should be consid-ered in connection with the demand of the separation of the inert material, e.g.through mechanical-biological processes aimed toward supplying less „ballast“through the incineration process.

The influence of the air staging and the inert material on the burn-out can be clearlyshown through calculations with the help of mathematical models for the solid con-version. Although the model cannot be discussed here in detail, it should be men-tioned that the solid conversion of unknown fuels in grate systems can be describedfor stationary and changing practical processes using this model [18]. The residualcarbon content with regard to the initial amount in the bed element over time forvarious boundary conditions is shown in Fig. 9 (see legend in Fig. 9). A comparisonof curves 1 and 2 in Fig. 9 clearly show the influence of the inert material as de-scribed above. A significantly better carbon conversion is achieved with inert mate-rial than without. Curve 4 shows the influence of a air staging for a fuel without inertmaterial in which the residual carbon content is as low as that for case 2 (without airstaging).

To ensure an even distribution of primary air, the pressure loss in the grate elementmust be much higherthan that in the wastebed, independent ofthe type of grate. If themain fluid resistancefor the reaction gas(downdraft) is locatedin the grate element,then an even distribu-tion of waste over thegrate surface resultsregardless of the ac-cumulation of thegrate element withwaste (fuel bed).

Different specifications, with respect to the different residence time behaviors ofbeds on forward-acting and reverse-acting grate systems, also exist for the control of

300 600 900 1200 1500 1800time t [s]

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

carb

on c

once

ntra

tion

mC

,Re/

mC

,0 [

- ]

33

22

11

44

curvecurvemRGmRGmC,0mC,0mInmIndeq,0deq,0Ψ O2,RGΨ O2,RGϑRGϑRGϑ0ϑ0

[kg/h][kg/h][kg][kg][kg][kg][mm][mm]

[-][-][oC][oC][oC][oC]

11

00

10001000

2 2 100 100 3,5 3,5 3,5 3,5 10 100,210,21 40 40 10001000

33

3,53,5

20 20

4-I 4-II 4-I 4-II 200 600200 600 10 10 0 0 20 20 0,21 0,21 40 40 1000 1000

II IIII

air stagingair staging

[without inert material in the bed and [without inert material in the bed and without ignition]without ignition]

[with inert material in the bed[with inert material in the bed

[with inert material in the bed and without ignition][with inert material in the bed and without ignition]

and with ignition]and with ignition]

[without inert material in the bed and [without inert material in the bed and without ignition]without ignition]

Fig. 9. Calculated carbon concentration vs period of timefor a given bed [18].


the residence time along the grate pathway. For forward-acting grate systems, acontrol of the grate speed in several grate zones independent of one another is gen-erally possible. However, for reverse-acting grate systems, such a differentiatedcontrol of the residence time is not carried out. A so-called discharge drum for theadditional control of the residence time in the burn out zone is found more often inreverse-acting than forward-acting grates [46].

The course of combustion is influenced, in addition to the distribution of the primaryair, through the temperature and oxygen concentration of the primary air (reactiongas). The primary air is generally preheated to temperatures of ϑ = 140 °C, which isparticularly advantageous for the drying and burn-out phase. Higher bed tempera-tures are also reached in the main combustion zone through a preheating of the air.The increase of the oxygen concentration to up to 35 vol.-% in the reaction gas iscarried out e.g. using the “SYNCOM Process” [6]. This guarantees the burn-out andthe criteria for the landfill class I [48] and causes a reduction in the flue gas flow (seebelow). In addition, the gas flowing through the bed can be reduced which leads to adecrease in the amount of flue dust and the diameter of the flue dust particles.

The temperature along the grate pathway is still influenced by the control of the gasflow over the bed. In the parallel flow principle, a combustion of the gases comingfrom the beginning of the grate is sought. The hot gases are then lead over the burnout zone. On the other hand, in the counter flow principle, the hot gases shouldtransfer heat from the grate end in the drying and degassing zone to the beginningof the grate. The flow control will be discussed later with regard to the post-combustion. At this point it should be mentioned that the temperature level in thedrying and burn-out phases is influenced through the design of the combustionchamber geometry (secondary radiation surface). This so-called secondary heatingsurface is of more importance in the burn-out zone of a counter flow combustion thana parallel flow combustion. In the parallel flow combustion, the design of the com-bustion chamber must avoid slagging - for wastes with high heating values in par-ticular.

The use of water-cooled grate elements leads to the above-mentioned advantagethat the primary air distribution is less dependent upon the cooling of the grate ele-ments. This can then result in further improvements with respect to the ash quality(increase in temperature and reduction of the oxygen concentration) in experimentson the pilot plant scale ([19], [36], [37]). In addition, the flue dust concentration (seeabove) is reduced through the lowering of the primary air. The water cooling alsohas the advantage [7] that the wear through thermal influences is reduced and that,even at high heating values, there is no threat of thermal overload and that the grate


siftings are decreased (less expansion allows for narrower slits between the grateelements).

The setting of the primary air supply, grate speed, waste feed, etc. is carried out us-ing firing power control in order to ensure an even course of combustion and highestpossible burn out [2]. The firing power control can be supported through additionaldetectors and measuring signals, e.g. IR camera [8], pyro-detectors [9], „heatingvalue sensors“ [49].

Whereas the setting of the process control for a high burn-out is state-of-the-art inthe practice, the optimization of the process parameters with respect to the selectiverelease or binding of heavy metals in the ash (primary measures for the improve-ment of the ash quality) is still a current research topic. Approximately 300 kg/Mgwaste

grate ash remain after the waste combustion, from which ca. 40 kg/Mgwaste scrap ironcan be separated. Currently, ca. 3 million Mg/a grate ash accumulate in Germany, ofwhich ca. 50 % to 60 % are used as secondary building material in road and pathconstruction and the rest is deposited in overground landfills [50]. The aim of futuredevelopments must be the improvement of the ash quality through primary measuresin order to make a high-grade utilization possible.

Dependent upon the combustion condi-tions, heavy metals, chlorine, sulfur andfluorine etc. are released from the fuelbed and bound by the ash. One candifferentiate between highly volatilesubstances such as Hg, Cd, As, Clmiddle volatile substances such as Pb,Zn, S and nonvolatile substances suchas Cu.

Investigations with various waste com-positions in a pilot plant have shownthat the evaporation of heavy metals(e.g. Cu, Zn and Pb) is correlated withthe bed temperature and chlorine con-tent (Fig. 10, Fig. 11; MW = municipalsolid waste, reference fuel; CSR = carshredder residue - light-weight fraction;ES = plastic scrap from electronics)[36].

Waste bed temperature in °C

Vola

tiliz

atio

n in

% o

f th

e in

ven

tory

woo

d

stra

w

MW

CS

RP

VC

ES

Fig. 10. Relationship between volatiliza-tion and waste bed temperature[36].

Cl-inventory in g/kgCl-inventory in g/kg

Vola

tiliz

atio

n in

% o

f th

e in

vent

ory

woo

d

stra

w

MW

CS

R

PV

C

ES

Fig. 11. Relationship between volatiliza-tion and Cl inventory [36].


An additional parameter for the evaporation is the oxygen concentration. Resultsfrom experiments with radioactive tracers (69mZn) in a pilot plant [37] clearly show thedifferent behavior for the release of zinc under reducing and oxidizing conditions.The measured zinc evaporation for three different operation conditions is shown inFig. 12. The two parameters which were varied on the grate were the oxygen partialpressure and temperature. The redox conditions of the incineration process werechanged from combustion to gasification operation by decreasing the flow rate ofprimary air to one third. But even in the so-called combustion operation mode, re-ducing conditions dominate in the solid bed in the first primary air zone. The tem-perature variation was achieved by adjusting the water content in the waste material.Fig. 12 reveals that the trend of zinc evaporation is independent of the operatingconditions and the amount of evaporated zinc. The evaporation takes place very fasti.e. in a narrow area on the grate. However the location of the evaporation dependson the operating conditions. The evaporation always occurs at locations with hightemperatures and with reducing conditions and can reach up to 100 % evaporation.For the experiment in which only 50 % evaporation of the radio tracer was obtained(wet, oxidizing), one could expect the following: First, the location of evaporation isshifted to places further down the grate, due to a delayed in reaching the necessarytemperature. Second, the shift in the temperature profile brings the evaporation zonefor zinc into the region where oxidizing conditions prevail (air zone 2). More andmore zinc oxide is produced which is not volatile and therefore further zinc evapora-tion is stopped. The possibility that a thermal mobilization of heavy metals at hightemperatures is supported by the reducing effect of the hot carbon in the fuel bed isreferred to in [36].


It should also be men-tioned in connection withthe ash quality that amarked decrease in theextractability of pollutantscan be achieved througha sintered state in com-parison with the so-calleddry state. Through thetransition from the sin-tered to the melted state,only relatively small im-provements with respectto the extractability are

achieved [36].

Post-combustion

The main objective of the post-combustion process is primarily the degradation ofCO and organic trace elements as well as the burn-out of fly dust (a total high burn-out). The minimization of NO through primary measures is also being investigatedfor direct practical implementation. The process control of NO minimization with si-multaneous high burn-out in connection with the firing systems for gaseous, liquidand dust-like fuels is already state-of-the-art ([25] to [29]).

The main influencing parameters to achieve this objective are temperature, oxygenconcentration, residence time and residence time behavior. These parameters aredetermined through a series of individual aspects of the process control and not onlythrough the flow form which is often discussed in connection with an initial post-combustion of the gases (e.g. parallel flow) or the minimization of NO in the com-bustion chamber above the bed. Through an appropriate separation of the post-combustion zone (2nd unit) from the grate zone (1st unit), the gases, flowing parallelor counter, can be post-combusted optimally regardless of their state at the entranceto the post-combustion. The separation of the post-combustion from the grate asshown in Fig. 1 will be discussed in more detail in [1]. The separation does not haveto be geometric, it can also ensue hydraulically. A geometrical separation wouldhowever be optimal (for example over a tube).

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00

zinc

eva

pora

tion

[-]

oxidizing conditionswater content of waste w <wevaporation ~ 100 %

2 1

oxidizing conditionswater content of waste wevaporation ~ 50 %

1

reducing conditionswater content of waste wevaporation ~ 100 %

1

position on grate [m]

Fig. 12. Normalized Zn amount detected in flue gasscrubber after the pulsed addition of Zn tracer atposition 0. Accumulation is monitored whiletracer is transported along the grate (0.0 m to2.0 m). oxi.: oxidizing conditions; red.: reducingconditions (gasification); w: water content ofwaste; evaporation: % of tracer transferred toflue gas [37].


This can however generally not be carriedout for practical reasons. A hydraulic sepa-ration (Fig. 13, Fig. 14) is also very effectivewhen, for example, a good „overlap of thecross-section“, adequate suction of the re-sidual current to be mixed and sufficient„penetrating depth“ of the injector jets isachieved according to the jet depth theorythrough dimensioning of the jet geometryand penetration impulse. This generallyleads to fields of jets in several levels (Fig.15). If possible, fields of jets should be ar-ranged on all channel sides to ensure goodoverlap. The positioning of the injectionalong a narrowed cross-section in the post-combustion zone supports the mixing (Fig.

14, Fig. 15). This hydraulic separation also leads to the formation of the continuouslystirred reactor element (CSR) shown in Fig. 6. The settling section is then locatedabove the fields of jets (Fig. 14, Fig. 15) („plug flow zone"). The average velocity wx

of an injector stream which penetrates thecombustion chamber above the secon-dary air jets with a speed of w0 decreasesapproximately hyperbolically with thepenetration depth of the stream x. In thecombustion chamber, the stream is di-verted in the direction of the upstreaminggas y, depending on the ratio of the flowrate of the upstreaming gases wy to theexit velocity of the stream at the jet w0

proportional to the square of the penetra-tion depth x2. Through the reduction ofthe velocity wx and the diversion of thestream ∆y, the mixing is worsened corre-spondingly. The installation of a so-called prism in the combustion chamber hasserved to improve the mixing conditions both for old and new plants with large cross-sections in the area of the secondary air injection and the corresponding largepenetration depths in particular (Fig. 16). In this system, first installed in the MSW-Incineration plant in Bonn, the flue gas from the grate is divided into two partial flows“A” and “B” by a membrane-wall construction in the shape of a prism. The prism is

waste bed

continuouslystirred reactor

waste

ash

primary air

secondary airfuel

post-combustion

solid conversion(grate)

plug flowreactor

Fig. 13. Schematic representation ofthe grate and post-combustion zone [51].

1

2

3

4

5

6

7

8

910

11

1213

1 charging hopper2 shutting clack3 charging shaft4 charging grat5 main grate6 burn-out grate zone7 ash discharge (dry)8 ash discharge (wet)9 ash discharge (grate siftings)10 primary air11 secondary air12 tertiary air /

recirculated flue gas13 ignition burner

Fig. 14. Schematic representation of theforward acting grate and sepa-rated post-combustion zone [7].


water-cooled and protected with refraction-ing material. Secondary air is injected intothe divided flue gas streams “A” and “B” asindicated in Fig. 16, reducing the jet lengthover the cross-section correspondingly.Through the installation of the prism, amuch shorter and clearly defined burn-out ofthe flue gases just above the prism could beachieved [16].

The injection of secondary air should beavoided as much as possible. The substitu-tion of sec-ondary air

through recirculated flue gas reduces the flue gasmass flow (at the chimney) on the one hand and onthe other hand prevents temperature peaks in thepost-combustion through intense mixing when theinjector jets are positioned properly. This leads to adecreased thermal formation of NO. The influence ofthe flue gas recirculation in the post-combustion pro-cess of a waste combustion plant is shown in Fig. 17[5].

The flue gas recirculation was simulated through the addition of nitrogen (inert gas).The mixing of the flue gas originating at the grate is carried out at a high inert gasratio with injector jets which contain almost no oxygen. High temperature peaks canthen be avoided. Altogether, a decrease in the NOX concentration is achieved. Theconstant low CO concentration with regard to the inert gas ratio shows that the COconversion is not directly dependent upon the oxygen available in this case for acorrespondingly designed mixing capacity.

A different separation between grate and post-combustion as represented in Fig. 13is shown in Fig. 18. Here, the stirred reactor element for the post-combustion of thegases emitted from the grate is situated above the first half of a drum grate. Thepost-combustion zone now only consists of the „calmed burn-out zone“ (plug flowreactor zone in Fig. 13).

IEVB/GW502/21.03.1996nozzle stage 2

nozz

le s

tage

1nozzles

nozzlesinjection jet expansionof a single nozzle

suction of fuel gasesproduced at the gratefor mixing and post-combustion

Fig. 15. Possible arrangement offields of jets in two levels forthe formation of a „stirredreactor element“ [4].

A

B

Fig. 16. Cross-sectionPrism [16].


The European CombustionGuidelines, based on the17th BImSchV, set process-internal requirements for theminimum values of the oxygenconcentration, temperatureand residence time (6 vol.-% O2, 850 °C, 2 s) with theobjective of guaranteeing asufficient burn-out. This re-quirement forces the operatorto increase the temperature inthe combustion chamberthrough e.g. the reduction ofthe primary excess-air coeffi-cient. This intensification of the combustion requirements in connection with the in-creased heating values in recent years leads to higher thermal load and more corro-sion and therefore to less availability of the plant [38]. In connection with the princi-ple requirements for the post-combustion step (see above) as well as with the exam-

ple discussed, it has becomeclear that the main influencingparameters oxygen concen-tration, temperature, resi-dence time and reactor be-havior (mixing) cannot be dis-cussed individually, especiallywith respect to a high degra-dation of CO and organictrace elements. This relation-ship was confirmed in a prac-tical plant (MHKW Mannheim)within the framework of a re-search project [39]. Five ex-periments were carried out inthe waste reactor 2 of the

HMVA Mannheim in which the parameters combustion air and amount of waste werevaried in order to influence the oxygen concentration, temperature and residencetime. The experimental settings were accompanied by an extensive measurement,sampling and analysis program for the flue gas flow (gas and dust) and the grate

0

50

100

150

200

250

300

350

400

450

500

0 0,2 0,4 0,6 0,8 1

Inert gas rate N

OX

[mg/

m3 ] i

.s.s

.d. a

t 11

Vol

.-% O

2

0

10

20

30

40

50

60

70

80

90

100

CO

[mg/

m3 ] i

.s.s

.d. a

t 11

Vol

.-% O

2

NOx ConcentrationCO Concentration

Fig. 17. NO and CO gas concentration dependentupon the inert gas rate [5].

4

6

8

6

5

75

1

5

35

>1200°C

1050-

1150°C

950 - 1025°CMinimum temperature recording 17.BlmSchV 850°C

21 - Feeding hopper2 - Roll screen3 - Co-current flow4 - Primary air system5 - Secondary air system6 - Burner7 - Post-combustion zone8 - Ash outlet

Fig. 18. Drum grate with „Feuerwalze“ [3].


ash. The residence time range for the flue gas during the experiments at tempera-tures above 850 °C was between τ850 = 0.7 s and 2.27 s. An important result of theexperiments is that no direct connection is evident between the pollutant concentra-tion in the flue gas and the residence time in the post-combustion chamber. The in-fluence of the mixing (reactor behavior) with respect to the results shown in Fig. 19should also be mentioned here2. The experimental settings V5 (total load) and V2(partial load) result in almost identical requirements concerning heating value, fireposition, combustion chamber temperature and flue gas composition. They differhowever in the load conditions with respect to the flue gas flow and the residencetimes in the post-combustion process. A shorter residence time in experiment V5with better mixing results in CO, Corg., and PCDD/F flue gas concentrations almostidentical to those of the V2 experiment. The results of the practical plant supportthose from investigations carried out in the pilot plant ([40], [41], [42]).

In the field of classical wastecombustion, measures for theprimary reduction of NO are stillbeing investigated [52]. Thesemeasures are connected morestrongly with the solid conversionin the bed and the conditions di-rectly over the bed in the com-bustion chamber than those forthe decomposition of CO and or-ganic trace elements. At the be-ginning of the grate, volatile ni-trogenous compounds (NHi, Norg.),which can be converted to NO when sufficient oxygen is available, escape from thefuel bed. The conversion of NHi radicals to HCN proceeds with increasing release ofvolatile components and formation of CH4 in particular. These reaction pathways arewell-known from batch grate and other experiments [31] and support the assumptionthat the NO formed can be degraded to N2 in the presence of NHi during the NO re-duction in the grate unit. However, a higher temperature is required in the gratecombustion chamber for this so-called internal “Exxon Process” than for the SNCRprocess. In order to suppress the conversion of NHi radicals with CH4 to HCN, tem-peratures above 1000 °C are necessary which result in the degradation of CH4. Itcan then be concluded that the temperature ϑG,gI is an important influencing pa-

2 Refer to [39] for more details.

0

2

4

6

8

10

0 0,5 1 1,5 2 2,5 3

τ850 [ s ]

0

10

20

30

40

50

V 5

org. C[mg/Nm3]

PCDD/F[ngTE/Nm3]

CO[mg/Nm3]

1.71 2.230. 70

2.27

V1V3

V4

COorg. CPCDD/F

O2 = 11%

V2

Fig. 19. PIC concentrations (CO, org.C,PCDD/F) of the flue gas in dependenceof the residence time of the flue gas τ850

above 850 °C [39].


rameter with respect to the release of volatile compounds and the degradation ofCH4 at the beginning of the grate. In addition, the oxygen supply or excess-air ratiohas an increasing influence on the determination of an optimal NO/NHi ratio in thefirst grate zone.

The effects of the main influencing parameters ϑG,gI and λgI described above wereconfirmed through tests in a pilot plant.

The results are shown inFig. 20. As expected, theNO values decrease withincreasing temperatureϑG,gI. In addition, localminima in the range of0.4 < λgI < 0.6 result forconstant temperatures.Reference points support-ing these results can bederived from investigationsin connection with the NOreduction in a fixed bedgasifier with separate post-combustion [30].

As shown below, an inde-pendent optimization of theNO reduction, in addition to

the optimization of the solid conversion together with other advantages, is achievedwhen the process control of the post-combustion is clearly separated from that of thegrate. Another degree of freedom for an independent post-combustion results fromthe gasification on the grate and the generation of a combustion gas (possibility ofair staging for an independent post-combustion for NO reduction, Fig. 25).

Gasification - Post-Combustion Process

In the separated process control, as shown in Fig. 1, the solid conversion on thegrate (1st unit) can be operated understoichiometrically (e.g. λ ≈ 0.4). Due to theabove-mentioned control possibilities of grate systems, a complete burn-out of thegrate residue (ash) can also be achieved in this mode of operation. The difference tothe classical incineration process is the generation of a gas which can be combustedindependently in the post-combustion process in the second unit. The second unitcan therefore be designed as an independent firing unit. This gasification-post-

0,05

0,06

0,07

0,08

0,09

0,10

700800

900

1000

11000,3

0,40,5

0,60,7

0,80,9

1,0

stan

dard

ized

NO

-con

cent

ratio

n /

X

NO

,exp

NO

,theo

,max

yy

XX

temperatureof zone I

[°C]JG,g I

air ratioin zone I

[ - ]lg ICUTEC / Ne / B-NOX-3D-eng.cdr / 08.04.1998

Fig. 20. Dependence of the NOX concentration on thetemperature ϑG,gI and the excess-air coefficientλgI during the combustion on the grate [53].


combustion concept, currently examined on a test-size scale, appears promising,as, in comparison to the conventional incineration processing in grate systems,

• the flue gas mass flows are significantly reduced (Fig. 21),

• combustible gases which enable an independent post-combustion processare generated,

• the post-combustion process itself can be optimized regardless of the proc-ess on the grate with the help of familiar primary measures for reducing theNOX-emissions and at the same time achieving high burn-out results,

• emission loads can be reduced considerably.

These aspects are nowexplained in detail, withreference to first resultsat a pilot plant. It shouldbe mentioned, that alsoan industrial scale plantwith a thermal power of15 MW has been com-missioned at the begin-ning of 2002 [43]. Theconsiderations for theoptimization of the pro-cess control begun overten years ago [12] andthe results achieved in

pilot plant in the mid 1990s ([10], [18]) are supported by results of the recently com-missioned industrial plants.

The diagram in Fig. 21 shows that for a separated process control with a stoichi-ometric ratio of λ ≈ 0.4 to λ ≈ 0.6 for solid conversion on the grate and λ ≈ 1.2 toλ ≈ 1.8 for the post-combustion, total stoichiometic ratios of λ ≈ 1.1 to λ ≈ 1.4 result.The total stoichiometric conditions achieved during the separated gasification-post-combustion process are significantly lower than those for the classical gasification-post-combustion operation (λ ≈ 1.6 to λ ≈ 2.0) resulting in a slower flow velocity inthe fuel bed (dust discharge) and a reduction in the flue gas mass flow in compari-son with the classical combustion process.

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1stoichiometrical ratio of the grate λg

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2

tota

l sto

ichi

omet

rical

ratio

λto

t

1.01.0

1.11.1

1.21.2

1.31.3

1.41.4

1.51.5

1.61.6

1.71.7

1.81.8

1.91.9

2.02.0stoichiometrical ratio in thestoichiometrical ratio in the

post combustion chamber λpcpost combustion chamber λpc

parameter of the curves:parameter of the curves:

Ne / b-lambd.ar2 / 26.03.1997

Fig. 21. Dependence of the total stoichiometrical ratio onstoichiometrical ratios of the grate and the post-combustion chamber [18].


As reported elsewhere [18] the combustiblegas composition differs from values resultingfrom balance calculations. The major com-bustible component is found to be carbonmonoxide (ψ CO ≈ 8..15 vol.-%). The hydrogencontent (ψ H2 ≈ 2..5 vol.-%) is far below the cal-culated equilibrium concentration. This maybe attributed to the fact that the water contentin the combustible material evaporates at thebeginning of the grate (Fig. 4). Thus, a het-erogeneous decomposition reaction betweensteam and the hot coke bed is not normal inconventional grate systems. Furthermore, theequilibrium of the homogeneous reaction ofCO and H2O to form CO2 and H2 is not at-

tained. Based on the assumption that the CO formation in the combustion bed of agrate essentially takes place via the heterogeneous gasification reaction of carbonwith oxygen and, depending on the height of the bed, additionally via the so-calledBoudouard reaction, a hot coke bed should follow soon after the successful ignitionof the combustible. Due to the decreasing carbon content along the length of thegrate, less reaction air is required in the following grate zones for gasification. Thisfact is confirmed by the results presented in Fig. 23, obtained at a pilot reverse-acting grate (0.5 MWthermal) ([10], [18]). For evaluating the influence of air staging,three distinctly different air distribution settings (in each case with constant massflows for both fuel and total air) have been tested. The main air supply is in zone 1for the first setting and in zone 4 for the second. An even distribution over the zones1 to 4 is approached for the third setting. Fig. 22 shows that the leveling off of hy-drogen and methane concentrations ψ is more or less independent of the selectedair staging settings. The wood used as a model fuel already ignites in the first stageof the grate. When shifting the main air supply from the beginning to the end of thegrate, the CO concentration in the combustion gas is reduced. However, an increaseof the grate bar velocity, which leads to a more intensive mixing and stoking of thecombustion bed, causes an increase of the CO concentration in the example given(Fig. 22).

The distribution of air depends on the fuel or waste material gasified. Refuse-derivedfuel (RDF) with a significant content of synthetic material requires careful degassingand ignition (Fig. 23). Supplying the main part of air at the beginning of the gratecauses the degassing products to be burnt immediately. This leads to high tem-

0

5

10

15

20

com

posi

tion

of c

ombu

stio

n ga

s ψ i

[Vol

.-% i.

N.tr

.]

ψ CO2ψ CO2

ψ COψ CO

ψ H2ψ H2

ψ CH4ψ CH4

0

25

50

dist

ribut

ion

ofre

actio

n ga

s [%

]

11 2233

44

55

test 960430-1λg = 0.4

test 960426-1λg = 0.4

test 960430-2λg = 0.4

test 960503-1λg = 0.4

zone

1zo

ne 1

zone

2zo

ne 2

zone

3zo

ne 3

zone

4zo

ne 4

zone

5zo

ne 511

11

11

22 222233

333344

4444

55

55

55

grate element velocitylow (ωHP = 30 Hz)



grate element velocityhigh (ωHP = 80 Hz)

CUTEC / Ne / e-rvgl1.ar2 / 17.12.1996

Fig. 22. Comparison of the com-position of combustion gaswith varying distribution ofreaction gas along thegrate path and grate ele-ment velocity (model fuelwood) [10].


peratures of the bed and consequently to a caking and fritting of the bed. An evenflow through the bed is hindered.

As for incineration, residual carbon contents ξC of about1 ma.-% or less can be attained under gasification con-ditions; thus, coke-like solid residues are avoided. Forexample, results of combustion and gasification of woodfrom railway sleepers treated with coal tar (contami-nated, and, thus, waste wood of no further use) areshown in Fig. 24. With respect to the remaining burn-outof the ash it must be mentioned at this point that with anoverall understoichiometrical operation of the grate pro-cess (gasification), local overstoichiometrical conditionsregarding the remaining carbon can nevertheless beadjusted in the area of the burn-out zone, if necessary.

Marked differences result between process conditions ofthe gasification and incineration mode with regard to theformation of flue dust. The significantly lower mass flowof air resulting from the operation under gasificationconditions compared to those resulting from the super-stoichiometrical operation causes a corresponding re-

duction of flow velocities through the combustion bed. The expected tendency, that adecreasing stoichiometric ratio causes the formation of less flue dust, is confirmed inFig. 24.

The gases generated by gasification in the grate process are supplied to the post-combustion chamber. The heating value is about hn ≈ 1500 kJ/kg to hn ≈ 2500 kJ/kgand temperatures reach levels of ϑ ≈ 750 °Cto ϑ ≈ 1000 °C depending on the stoichi-ometric air ratio of the grate process. If thepost-combustion chamber is well insulatedwith a refractory jacket, the post-combustionprocess runs independently without addi-tional fuel. Well-known primary measures tominimize pollutants can then be applied(e.g. [25] to [29]). Consideration is givenhere to NO minimization with a simultaneousreduction of the CO concentration. First of

0

5

10

15

20

com

posi

tion

of c

ombu

stio

n ga

s ψ i

[Vol

.-% i.

N.tr

.]

ψ CO2ψ CO2

ψ COψ CO

ψ H2ψ H2

ψ CH4ψ CH4

0

25

50

dist

ribut

ion

ofre

actio

n ga

s [%

] zone

1zo

ne 1

zone

2zo

ne 2

zone

3zo

ne 3

zone

4zo

ne 4

zone

5zo

ne 5

11 22 33

44

55



test Ro-26-01λg = 0.4

test Ro-27-01λg = 0.4

CUTEC / Ne / e-rvgl2.ar2 / 17.12.1996

Fig. 23. Comparison ofthe composi-tion of com-bustion gaswith varyingdistribution ofreaction gasalong the gratepath (refuse-derived fuelRDF) [10].

100

200

300

400

500

600

flue-

dust

con

cent

ratio

n χ

D,E

G,p

c [ m

g/m

3 (i.s

.s.d

ry)

refe

rrin

g to

3 v

ol.-%

O2

]

0.5

1

1.5

2

2.5

3

loss

on

igni

tion

of th

e re

mna

nts

ξ loi,r

e,g

[ma.

-%]

χD,EG,pcχD,EG,pc

ξloi,re,gξloi,re,g

λg=λtot=1.6 λg=λtot=1.8λtot=1,12λg=0,6

λtot=1,12λg=0,8 CUTEC / Ne / e-staub1 / 24.03.1997

Fig. 24. Flue dust concentration be-fore flue gas cleaning andloss of ignition for differentstoiciometric ratios in thestoker system [18].


all, the stoichiometric ratio of thegasification process affects theNO emission.

Fig. 25 shows the NO and COemissions of a staged combustionof the combustible gas generatedat the grate for a primary air ratioof λg ≈ 0.4 and λg ≈ 0.6. For thehigher primary air ratio, signifi-cantly lower NO emissions areobtained. An increase of the pri-mary air ratio is accompanied bya temperature increase fromabout ϑ ≈ 700 °C to about

ϑ ≈ 1000 °C. Furthermore, an increase of the oxygen supply results. With this, ahigher decomposition of volatile nitrogen components such as HCN and NHi via NOand a reduction of already formed NO via the “NO recycling“ path is more probable.In addition to the influence of the primary air ratio, the NO concentrations in Fig. 25show the typical course for the formation of NO from the nitrogen contained in thefuel. The NO concentrations decrease with a falling air ratio in the first stage of thecombustion chamber whereby the total air ratio is kept constant λtot ≈ 1.3. For a pri-mary air ratio of about λg ≈ 0.4, the NO concentration drops from about 450 mg/m3 fora single-staged post-combustion to 200 mg/m3 for a twofold-staged post-combustion(air supply in stage I and II) with λPC I ≈ 0.5, whereby the CO concentration is far be-low 10 mg/m3. In the same manner, the NO concentration can be reduced in thecase of a primary air ratio of λg ≈ 0.6, from 200 mg/m3 to about 120 mg/m3. For atwofold-staged post-combustion, a further NO reduction below 100 mg/m3 resultswhen air is fed into the first and the third stage of the combustion chamber wherebythe CO concentration remains below 10 mg/m3 [32]. The further NO reduction here isdue to the longer residence time in connection with higher temperatures and loweroxygen concentrations in the understoichiometric first stage. Furthermore, if air is fedto the third stage, nearly plug flow conditions prevail in the understochiometric partafter the enlargement of the post-combustion chamber, which is preferable for NOreduction steps [29].

Efficiency

In order to conclude the section covering the classical waste combustion, the deter-mination of efficiencies for the individual partial processes and for the total processwill be discussed. Within this framework, the representation can only be summarized

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2air ratio λpc I

0

50

100

150

200

250

300

350

400

450

500

NO

2-co

ncen

tratio

n ψ N

O2,

EG,p

c[m

g/m

3 i.s.

s.dr

y re

ferr

ing

to 1

1 vo

l.-%

O2]

0

20

40

60

80

100

120

140

160

180

200

CO

-con

cent

ratio

n ψ C

O,E

G,p

c[m

g/m

3 i.s.

s.dr

y re

ferr

ing

to 1

1 vo

l.-%

O2]

mG,g mRG,pc I

mEG,pc

parameter of the test

mf

λg

λtot

60 kg/h0.40.6

mRG,pc II

mRG,pc III

NO2

CO

position ofair staging

1.3to1.5

1.2to

1.4

pc I and pc II

waste wood 2

supply of combustible (G)and reaction gas (RG) (air)(postcombustion chamber)

NO2- and CO-concentration versus air ratio λpc I referring to different air ratio λg

CUTEC / Ne / NOCO-LR.ar2 / 21.03.1997

Fig. 25. NO2 and CO concentration versus airratio λpc I (different air ratio λG in thegrate unit) [18].


here. The description of operations through the determination of efficiencies canalso be applied correspondingly to the process assessment using other evaluationcriteria such as the balance of the specific CO2 or pollutant discharge. The efficiencydescribes the relationship between use and expenditure. Depending on what is re-garded as use or expenditure, very different efficiencies can result. It is thereforevery important to differentiate between certain efficiencies according to the accom-panying balance limits for the assessment of processes and process chains. Thedifference between certain efficiencies is illustrated in Fig. 26.

The waste energy Hw3) and the supplemental energy required ΣEadd is added to the

balance room TP (Treatment Plant, TP) of the plant being in consideration. The plantthen releases the effective energy Hut and the sum of the energy lost ΣEloss,TP. Theso-called „plant efficiency” is then

addw

utPlant SEH

H+

=η (3-1)

3) Generally, energy is represented as flow or stream. In the following considerations, the absoluteenergy per 1000 kg waste (MJ/Mgwaste) and the capital letters (H, E, etc.) are chosen.

PrimaryEnergy

*)ConversionProcess

ConsideredThermal

Treatment Plant

Balance CP

Balance CU*

Balance CU

*)E.g. generation of oxygen, electrical energy etc.

Balance TP

Σ Hp (Return)

Σ Hp

Σ Hp

Σ Eloss,TP Σ Eloss,TP

Σ Eloss,CP

HwHw Hut Hut

ηn = =H - H

Hut pΣ

w

HH

n

w

Hw

Σ Eloss,CP

Σ Eloss,TP

Hnet

ηp =

HH

ut

w + HΣ p

HH

ut

w + EΣ addη

plant =

Fig. 26. Simplified energy balance to illustrate the plant efficiency ηPlant, the primary effi-ciency ηp and the net primary efficiency ηn.


When, in addition to the energy supplied for the operation of the plant ΣEadd, the en-ergy lost ΣEloss,CP during the corresponding conversion process is regarded as ex-penditure (balance room CP), i.e. the total primary energy ΣHp required for the op-eration of the plant is viewed as expenditure, for example, primary energy for thesupply of electricity, for an eventual oxygen generation, for the production of oper-ating substances and additives, etc., then a cumulative balance room CU and theso-called primary efficiency from the ratio of success to expenditure results

pw

utp SHH

H+

=η , (3-2)

which can be significantly lower than the plant efficiency ηPlant. When considered inthis manner, we are dealing with a so-called „cumulative“ representation of the mostimportant process units connected in front or perhaps afterwards. Starting points forthe cumulative consideration are the necessary primary energy and the raw materi-als for the thermal treatment of the waste. The end point of the consideration is thenat the balance limit where the environmentally acceptable deposition of remainingmaterials and/or the implementation of residual valuable materials in a recycling pro-cess is possible.

The primary energy consumption necessary for the treatment could be saved if nowaste existed. A „waste use“ Hn can be created in which the primary energies nec-essary ΣHp are subtracted from the effective energy Hut. The primary resource isthen replaced by the „self-generated“ effective energy or substituted or mentally re-circulated4, as shown by the dashed-lines in Fig. 26. Only the supplied waste energyHw remains as expenditure and the net energy Hn as use so that the so-called netprimary efficiency results

0)SH(Hcase the in HH

HSHH

putW

n

W

putn ≥−=

−=η (3-3)

For the primary energy substitution (recirculation), the balance limit is marked withan „*“ (Fig. 26 also „CU*“).

4) It must be mentioned here that the substitution of one type of energy (e.g. natural gas enthalpy)through another (e.g. generated vapor enthalpy) can generally not be carried out at a 1:1 ratio. Thismeans that the amount of energy to be substituted (e.g. bound to the natural gas) can generally notbe substituted through an equal amount of substitute energy but through a higher amount (e.g. boundvapor). This behavior is described through the so-called energy exchange ratio ER as the ratio of sub-stitution energy to energy to be substituted. Usually ER > 1 and is dependent upon many factors. Forclarity, ER = 1 in order to illustrate the principle of substitution or the energy recirculation. The deter-mination of ER is covered in [1], [33], [34], [35].


In the case of a high necessary primary energy supply, the difference (Hut - ΣHp) forthe formation of the net energy can also be a negative value, i.e. more energy is re-quired for the treatment of the waste than the waste itself adds to the process. In thiscase, the definition of the efficiency is negative. Therefore, such cases are regardedas degrees of expenditure. As reference value, Hw – Hn is chosen, i.e. the de-gree of expenditure a is defined as

0)SH(H case thein HH

HSHHH

SHHa put

nw

n

putw

put <−+

=−+

−= (3-4)

, so that the smallest value for the degree of expenditure assumes amin = -1 for rea-sons of formality.

In the following section, these efficiencies will be calculated for the example of theclassical waste combustion. A significant prerequisite for the process evaluation isthe systematic representation of partial processes with the determination of the bal-ance limits and the respective material and energy flows which enter and leave inthe balance circles. For thermal waste treatment processes, it has proved suitable todivide the first step of the total process into the partial steps of the so-called mainthermal process and the flue gas purification and energy conversion (see Fig. 2).With the help of the overview in Fig. 1 for the systematic description, the main ther-mal process of the classical waste incineration can be illustrated as a combustionunit (1st unit) and post-combustion (2nd unit). A generation of effective thermal energyin the reactor is coupled with both units with respect to the balancing. It can be prac-tical to include a possible post-treatment of the residual material in the first balancecircle when comparing processes with an integrated melting of the residues (e.g.high temperature gasification) for example. This then leads to a first balance circle A,as shown in Fig. 27, which contains the main thermal process with effective thermalenergy.


The balance limit B encompasses the balance room A with an internal recirculationof the effective thermal energy, e.g. for the preheating of the air [1]. The flue gas andwaste water treatment are included in the balance circle C. Together, the balancecircles of the main thermal process B and the flue gas/waste water purification Cform the balance circle of the total process D with the aim of the decoupling of theeffective thermal energy. If D is expanded to included the balance circle E, whichcontains the steam process, then the total process, with the goal of the electrical de-coupling of the effective energy in balance circle F, results. Processes in balancecircle F arranged in series before or after are of significance during the evaluation ofwaste treatment. These so-called cumulative viewpoints can be achieved throughthe expansion of the balance circles and the fictional supplementation of each proc-ess (e.g. the energy conversion in a power plant or the oxygen generation in an air-separation unit). Next, the material balances (material, mass and energy balance)must be created for the individual balance circles. The energy balance then formsthe basis for the determination of individual and total efficiencies. Fig. 28 showssuch an energy flow diagram for the case of a waste material with a low heatingvalue (hn = 6 MJ/kg). As mentioned before, the balance limit F was expandedthrough the addition of the processes for the oxygen generation and the primary en-ergy conversion. The balance limit K then contains the total process with effectiveelectrical energy and the consideration of the additional primary energy expenses.

Based on the systematic determination for the balance circles and the correspondingenergy balances, it is now relatively easy to form the characteristic single and totalefficiencies (basic description in Fig. 26). With the current definition of the efficiency:

input Nr. output

Balance boundary F ( )complete process with electr. ut. energy

Balance boundary D (complete process with therm. ut. energy)

Balance boundary C

Balance boundary Eelectr. energy transmutation

Balance boundary B

G~ηe,plant,E =

30%

F13additives FGC

F18 flue gas

flue gas

F14electr. energy FGC

F17 residues FGC

F15additiv fuel FGC

F16air FGC

F10 dissipation heat

F1electr. energy

F20 electr. energy

F2additives

F12 residues

F3air

F4oxygen

F5waste

F6additiv fuel

Σ input Σ output

flue gas cleaning(FGC)

waste water cleaning(WWC)

and

after-treatment

1. u

nit

2. u

nit

boile

r

Nr.

D13 C13 C17

D14 C14C18

D15 C15

D16 C16

C10C9

D5 B5 A5

D6 B6 A6

D4 B4 A4

D3 B3 A3

D2 B2 A2

D1 B1 A1 A9 B9

A10 B10 dissipation heat

steam

steam preheater

steam preheater

A12 B12

D10

E10

E20

condensate

D18

D17

D12

Fig. 27. Block flowchart of a classical household waste incineration (here with oxygenenrichment of the combustion air) [1].


η = use/expenditure, the following thermal plant efficiency for the balance room Aresults:

%72,2HH)HbisS(H

Heexpenditur

useRO,O

O

A8A7A6A1

A11Aplant,t, =

++==η , (3-5)

In the same way, the thermal plant efficiencies for consideration of the preheating ofthe air (balance circle B) and for the total process (balance circle D) can be deter-mined. The values from Fig. 28 are plotted in Fig. 29 as an example. Using the effi-ciency of the energy conversion from high-pressure steam to electrical energy (inthis case ηe,plant,E = 30 %, balance circle E), the electrical plant efficiency for the totalprocess (balance room F; decoupling of effective electrical energy) results as:

%19,9)Hbis(HHbisH

Eeexpenditur

use

F16F13F6F1

F20

S)S(Fplant,e, ===η+

(3-6).

Assuming an electrical efficiency for the electricity supply of ηe,n,I = 40 % and a spe-cific expenditure for the oxygen generation of 1.08 MJ electrical energy / kg oxygen,the additional expenditure in the form of the thermal and electrical primary energyefficiency can now be determined:

K13

K18 flue gas

K17 residues FGC

K27 dissipation heat

K28 flue gas

K29 nitrogen

K26 residues

K15additiv fuel

K14fuel

K16air

K22air

K21air

K10 dissipation heatK1fuel

K20 electr. Energy

K2

K12 residues

K3air

K24air

K4air

K23fuel

K5waste

Σ input7257 Σ output 7257

0

6000

K6additiv fuel0

1311

292

0

0

0

0

313

0

0

0

339

313

4638

640

0

0

0

0

energy input Nr. energy outputNr.

668

Bal. bound.JPower plant

ηe,n, I =40%

ηe,n, I =40%

ηe,n,I =40%

Balance boundary F (complete process with electr. ut. energy)

Balance boundary D (complete process with therm. ut. energy)

Balance boundary C

Balance boundary Eelectr. energy transmutation

Balance boundary B

G~ηe,plant,E

=30%

F13 D13 C13 C17

D18 F18

D17 F17

D12 F12

D10

F10

D14

D15

C14C18

C15

D16 C16

C10

D1 B1 A1

D2 B2 A2

D3 B3 A3

D5 B5 A5

D6 B6 A6

A12 B12

E20 F20

A10

E10

A9

B9

C9

B10dissipation heat

B9flue gas

steam

steam

steam preheater

condensate

F14

F15

F16

F1

F2

F3

F5

F6

Σ input6589

6000

4551

0

610

1145

181

0

flue gas cleaning(FGC)

waste water cleaning(WWC)

and

B10

after-treatment

1. u

nit

2. u

nit

boile

r

0

0

0

125

0

339

125

0

electr. energy FGC

additiv fuel FGC

air FGC

electr. energy

additives

air

waste

additiv fuel

additives FGC

el. energy 117

D4 B4 A4F40 oxygen

Balance boundary K (complete process with electr. ut. energy in consideration of the expenditures from the primary energy resources)

Bal. bound.I

Power plant

Bal. bound.I

Power plant

Bal. bound.I

Fig. 28. Simplified energy flow diagram of the classical household waste incineration withregard to the primary energy resource expenditure (here with oxygen enrichmentof the combustion air) [1].


%1,18)HH(SHHSHHS

?H

K24K21K16K13K6K1

E,a,eE11K,p,e bis)bis()bis(

O =⋅

=η++

. (3-7).

In the above-mentioned example, the primary thermal efficiency for the balance roomK results under the assumption that the electrical energy conversion (balance room)is omitted.

The primary energy could be saved or could supply a higher quality utilization inother processes, if no waste to be treated existed. This consideration leads, togetherwith the explanation to Fig. 27, to the net primary efficiency. For the example in Fig.28, the following net primary electrical efficiency results:

%14,0H)HbisS(H

EE?)HH(H

13K5K2K

20F20EEa,e,11D8A11AKn,e,

***

R*R**RO,*RO,*O*

* =+

−−⋅−−=η . (3-8)


A waste with a verylow heating value(hn = 6 MJ/kg) ischosen for the ex-ample in

Fig. 28. The heatingvalue of the wastehas a critical influ-ence on the processcontrol of the wasteincineration plantand affects the effi-ciency of the plantcorrespondingly.

Fig. 30 shows thedependence of effi-ciency determinedabove upon the initialwaste heating valuehn,W. In order tomaintain the requiredtemperature (herethe process tem-perature of the post-combustion) for de-creasing initial wasteheating values, the

measures „λ reduction“, „oxygen enrichment“ and „supplementary fuel“, often ap-plied practically, are compared with one another as shown in Fig. 30. The primaryelectrical efficiency ηe,p,K = 19 % and the net primary efficiency ηe,p,K* = 17 % can beread out of the diagram in Fig. 30 for initial waste heating values of hn,W = 10 MJ/kg,which is normal for middle European cities.

Balance circle Effi-ciency

Energy use Specific fluegas mass

Specific CO2

mass[MJ/kg W] [Mg/Mg W] [Mg/Mg W]

Balance circle AMain process with therm. ut.energy

ηt,plant,A

=72,2 %

4551 4616 641

Balance circle BMain process with therm. ut.energy and steam recirculation

ηt,plant,B

=71,3 %

4370 4616 641

Balance circle DComplete process with therm.ut. energy

ηt,plant,D

=66,3 %

4370 4780 660

Balance circle KComplete process with electr.ut. energy in consideration ofthe expenditures from theprimary energy resources

ηt,p,K =60,2 % 4370 5223 748

Balance circle K*Complete process with therm.ut. energy and recirculation ofthe expenditures from theprimary energy resources

ηt,n,K* =46,8 % 2807 4633 641

Balance circle FComplete process with electr.ut. energy

ηe,plant,F

=19,9 %

1311 4780 660

Balance circle KComplete process with electr.ut. energy in consideration ofthe expenditures from theprimary energy resources

ηe,p,K =18,1 % 1311 5223 748

Balance circle K*Complete process with electr.ut. energy and recirculation ofthe expenditures from theprimary energy resources

ηe,n,K* =14,0 % 842 4633 641

Boundary conditionsInitial net calorific value of the waste hn,W = 6 MJ/kgPreheating of the reaction gas ϑA = 120 °CO2 concentration Ψ O2,RG = 23.9 vol.-%Stoichiometric ratio λ = 1.9Treatment temperature ϑG = 1000 °CFlue gas temperature after boiler ϑG,B = 220 °CFlue gas temperature after flue gas cleaning ϑG,FGC = 60 °CFlue gas temperature after reheating ϑG,RH = 120 °CEfficiency of electrical energy conversion ηe,plant = 30 %

Fig. 29. Exemplary representation of the evaluation of wastetreatment (for examples see Fig. 28 and text) [1].


It should be mentioned, with regardto the cumulative balances, thatnegative energy balances can alsoresult, e.g. the processes consumesmore energy (with reference to theprimary energy) than is generatedthrough the waste treatment. It is inthis case no longer a degree of effi-ciency but a corresponding degreeof expenditure [1].

6 7 8 9 10 11 12

12

10

16

14

20

2246

50

54

58

62

66

70

74

78

18

initial waste thermal value in MJ/kg Wastehn,W

el. p

lant

, pri

mar

y an

d ne

t pri

mar

yef

ficie

ncy

fact

or

in %

η e,p

lant

,F ,

, η

ηe,

p,K

e,n,

K*

ther

m. p

lant

in

%, p

rim

ary

and

net p

rim

ary

effic

ienc

y fa

ctor

η t

,pla

nt,A

, ,

, η

ηη

t,pla

nt,D

t,p,K

t,n,K

*

ηe,plant,F

ηe,p,K

ηe,n,K*

λ limit curve for -reduction oxygen concentration of the reaction gas additiv fuel natural gas (essential energy condition = natural gas-/waste energy) uncoupling energy in the waste heat boiler and in the 1 unitst

ηηη

η

e,n,K*

e,n,K*

e,n,K*AF

e,n,K*

E

1,9

λ

Ψ O ,RG2

23 Vol.-%

22 Vol.-%

21 Vol.-%1,82,0

2,1

0%

15%

EAF

10%5%

ηt,plant,A

ηt,plant,D

ηt,p,K

ηt,n,K*

Fig. 30. Electrical plant, primary and net pri-mary efficiency (ηe,plant,F; ηe,p,K andηe,n,K*) dependent upon the initialwaste heating value hn,W for variousmeasures (classical waste incinera-tor; flue gas temperatureϑG,PC = 1000 °C) (explained in thetext) [1].


List of Symbols

SYMBOLS

a inert gas ratea degree of expenditureA/B partial flow A and BA-K balance circle number, effi-

ciencyCSR continuously stirred reactorE energy, energy exchange ratioh specific enthalpy (heat of com-

bustion)h calorific valueH enthalpym& mass flowMg 1000 gram = 1 tonMW municipal solid wasteoxi. oxidizing conditionsp pressurePe Peclet-NumberPFR plug flow reactorT temperature [K]ut utilizationV number experimental settingvol. volumew water-contentw velocityx stream xy stream y∆ differenceϑ temperature [°C]λ stoichiometric ratio,

excess-air coefficientξ concentration (mass-related)ψ concentration (volume-related)ω frequencyτ residence timeχ concentration

INDICES

0 starting valueI / II grate zoneadd additiveA air, reaction gasA-K balance circle number, effi-

ciencyb beginningB fuelB boilerC carbonCP conversion processCU cumulativeD duste electricale endeq equivalentexp experimentalF flue dustFGC flue gas cleaningg grateG fuel gas/flue gasHP hydrogen pumpi ignitionin inerti.s.s.d./i.s.s.dry

under standard conditions, drylo/loss

lossloi loss of ignitionmax maximummin minimumn net (lower) heating value, net

(usable) Enthalpy, net primaryefficiency

p primarypc/PC post combustion chamberplant plantR ratio (energy exchange ratio)re remnantsRe residueRG reaction gasRH reheatingS solidt thermaltheo theoretical


tot totalTP treatment processu usew/W waste

x stream x (post-combustion)y stream x (post-combustion)* recirculation of generated en-

ergy for substitution fuel


Notations

[1] Scholz, R.; Beckmann, M.; Schulenburg, F.: Abfallbehandlung in thermischenVerfahren. B.G. Teubner, Stuttgart, Leipzig, Wiesbaden. 2001. ISBN 3-519-00402-X.

[2] Emissionsminderung bei Müllverbrennungsanlagen. Endbericht eines Verbund-vorhabens zwischen Firma MARTIN GmbH, München, NOELL GmbH, Wür-zburg, L.&C. Steinmüller GmbH, Gummersbach, Projektträger UBA-Berlin,1994.

[3] Christmann, A.; Quitteck, G.: Die DBA-Gleichstromfeuerung mit Walzenrost.VDI-Berichte 1192, VDI-Verlag GmbH, Düsseldorf, 1995.

[4] Scholz, R.; Beckmann, M.: Möglichkeiten der Verbrennungsführung bei Rest-müll in Rostfeuerungen. VDI-Berichte Nr. 895, VDI-Verlag GmbH, Düsseldorf,1991.

[5] Scholz, R.; Beckmann, M.; Horn, J.; Busch, M.: Thermische Behandlung vonstückigen Rückständen - Möglichkeiten der Prozeßführung im Hinblick auf Ent-sorgung oder Wertstoffrückgewinnung. Brennstoff-Wärme-Kraft (BWK) /TÜ/Umwelt-Special 44 (1992) Nr. 10.

[6] Knörr, A.: Thermische Abfallbehandlung mit dem SYNCOM-Verfahren. VDI-Berichte 1192, VDI-Verlag GmbH, Düsseldorf, 1995.

[7] Lautenschlager, G.: Moderne Rostfeuerung für die thermische Abfallbehand-lung. GVC-Symposium Abfallwirtschaft Herausforderung und Chance, Wür-zburg, 17.-19. Oktober 1994.

[8] Martin, J.; Busch, M.; Horn, J.; Rampp, F.: Entwicklung einer kamerageführtenFeuerungsregelung zur primärseitigen Schadstoffreduzierung. VDI-Berichte1033, VDI-Verlag GmbH, Düsseldorf, 1993.

[9] Schäfers, W.; Limper, K.: Fortschrittliche Feuerungsleistungsregelung durchEinbeziehung der Fuzzy-Logik und der IR Thermografie. In: Thomé-Kozmiensky, K.J. (Hrsg.): Reaktoren zur thermischen Abfallbehandlung,EF-Verlag für Energie- und Umwelttechnik GmbH, Berlin, 1993.

[10] Beckmann, M.; Scholz, R.; Wiese, C.; Davidovic, M.: Optimization of Gasifica-tion of Waste Materials in Grate Systems. 1997 International Conference on In-cineration & Thermal Treatment Technologies, San Fransisco-Oakland Bay,California, 12.-16. May, 1997.

[11] Berwein, H.-J.: Siemens Schwel-Brenn-Verfahren - Thermische Reaktion-sabläufe. In: Abfallwirtschaft Stoffkreisläufe, Terra Tec `95, B.G. Teubner Ver-lagsgesellschaft, Leipzig, 1995.

[12] Scholz, R.; Schopf, N.: General Design Concept for Combustion Processes forWaste Fuels and Some Test Results of Pilot Plants. 1989 Incineration Confer-ence, Knoxville, USA, 1989.

[13] Göhler, P.; Schingnitz, M.: Stoff- und Wärmebilanzen bei der Abfallverwertungnach dem NOELL-Konversionsverfahren. In: Energie und Umwelt' 95 mit Ener-gieverbrauchsminderung bei Gebäuden, TU Bergakademie Freiberg, 22.-23.März 1995.

[14] Stahlberg, R.; Feuerrigel, U.: Thermoselect - Energie und Rohstoffgewinnung.In: Abfallwirtschaft Stoffkreisläufe, Terra Tec `95, B.G. Teubner Verlagsgesell-schaft, Leipzig, 1995.


[15] Albrecht, J.; Loeffler, J.; Reimert, R.: Restabfallvergasung mit integrierterAscheverschlackung. GVC-Symposium Abfallwirtschaft Herausforderung undChance, Würzburg, 17.-19. Oktober 1994.

[16] Périlleux, M.; Creten,G.; Kümmel, J.: Improving combustion and boiler perform-ance of new and existing EFW plants with the seghers-ibb-prism. 6thEUROPEAN CONFERENCE on INDUSTRIAL FURNACES and BOILERSINFUB Estoril - Lisbon - Portugal, 02.-05. April 2002.

[17] Scholz, R.; Beckmann, M.; Schulenburg, F.; Brinker, W.: Thermische Rück-standsbehandlungsverfahren - Aufteilung in Bausteine und Möglichkeiten derBilanzierung. Brennstoff-Wärme-Kraft (BWK) 46 (1994) 11/12.

[18] Beckmann, M.: Mathematische Modellierung und Versuche zur Prozessführungbei der Verbrennung und Vergasung in Rostsystemen zur thermischen Rück-standsbehandlung. CUTEC-Schriftenreihe, 1995.

[19] Hunsinger, H.; Merz, A.; Vogg, H.: Beeinflussung der Schlackequalität bei derRostverbrennung von Hausmüll. GVC-Symposium Abfallwirtschaft Herausfor-derung und Chance, Würzburg, 17.-19. Oktober 1994.

[20] Levenspiel, O.: Chemical Reaction Engineering. John Wiley and Sons. NewYork, 1972.

[21] Dryer, F.L.; Glasman, I.: 14th International Symposion On Combustion.Combustion Institute Pittsburgh, 1973.

[22] Leuckel, W.: Swirl Intensities, Swirl Types and Energy Losses of Different SwirlGenerating Devices. IFRF Ijmuiden, November 1967, G 02/a/16.

[23] Carlowitz, O.; Jeschar, R.: Entwicklung eines variablen Drallbrennkammersys-tems zur Erzeugung hoher Energieumsetzungsdichten. Brennstoff-Wärme-Kraft(BWK) 32 (1980).

[24] Scholz, R.; Jeschar, R.; Carlowitz, O.: Zur Thermodynamik von Freistrahlen.Gas-Wärme-International 33 (1984) 1.

[25] Malek, C.; Scholz, R.; Jeschar, R.: Vereinfachte Modellierung der Stickstof-foxidbildung unter gleichzeitiger Berücksichtigung des Ausbrandes bei einerStaubfeuerung. VDI-Berichte Nr. 1090, VDI-Verlag GmbH, Düsseldorf, 1993.

[26] Kolb, T.; Sybon, G.; Leuckel, W.: Reduzierung der NOX-Bildung ausbrennstoffgebundenem Stickstoff durch gestufte Verbrennungsführung.4. TECFLAM- Seminar, Oktober 1990, Heidelberg, 1990.

[27] Kolb, T.; Leuckel, W.: NOX-Minderung durch 3-stufige Verbrennung -Einfluß von Stöchiometrie und Mischung in der Reaktionszone.2. TECFLAM- Seminar, Stuttgart, 1988.

[28] Kremer, H.; Schulz, W.: Reduzierung der NOX-Emissionen von Kohlenstaub-flammen durch Stufenverbrennung. VDI-Berichte Nr. 574, VDI-Verlag GmbH,Düsseldorf, 1985.

[29] Klöppner, G.: Zur Kinetik der NO-Bildungsmechanismen in verschiedenenReaktortypen am Beispiel der technischen Feuerung. Dissertation, TU Claus-thal, 1991.

[30] Keller, R.: Primärseitige NOX-Minderung mittels Luftstufung bei der Holzver-brennung. Brennstoff-Wärme-Kraft (BWK) 46(1994) 11/12.

[31] Starley, G. P.; Bradshaw, F. W.; Carrel, C. S.; Pershing, D. W.; Martin, G. B.:The Influence of Bed-Region Stoichiometry on Nitric Oxide Formation in Fixed-Bed Coal Combustion. Combustion and Flame 59: 197-211 (1985).

3.3 conventional thermal treatment methods 3.3.1 overview ... · various processes which integrate...

Documents