1. Fluid mechanics and mixing
2. Burner Scaling
3. Gas Combustion
4. Oil Combustion
5. Coal Combustion
6. Fuel characterisation
7. Emissions & Reduction
8. Measurements equipment
9. Applied research
10. Mathematical Modelling
11. Other Research
12. Trials for customers and Members
13. Facilities

Starting in the sixties, combustion aerodynamics was one of the main theme of experiments. Various fuels were fired in IJmuiden's furnaces, several burners were investigated, and new flames were observed. As a consequence of the observations made initially in the fifties, attention was focused on jet flames, their entrainment and interaction with the furnace. Later on, swirl was applied to enhance the mixing between fuel and air. New flow patterns were observed invoking research in furnaces and isothermal models. Finally, the decade resulted in establishing means of generating flames possessing distinct fluid flow characteristics.

1.1 Aerodynamic and mixing experiments at IJmuiden

Overview table of aerodynamic and mixing experiments held at IJmuiden.

1.2. Aerodynamic and mixing in isothermal non-swirling flows

During the furnace trials performed in the sixties in IJmuiden's furnaces, it was observed that the near burner region was of primary importance for flame stability. The velocity and mixing measurements of isothermal flows of cement kiln burners demonstrated that this region extended up to about five primary nozzle diameters by which distance the primary jet, after some entrainment, had reached self-similarity.

1.3. Aerodynamic and mixing in isothermal swirling flows

The aerodynamics of flames observed in the third pressure jet oil performance trial in IJmuiden's furnaces was fascinating enough to render velocity and static pressure measurements in the flow region close to the burner which was mounted on a rig outside the furnace in isothermal conditions. A tangential entry swirl generator was used to provide the tangential velocity component. The ratio of the tangential to total flow rate was varied (0, 0.31,0.47,0.69) giving three degrees of swirl and a comparison with the no swirl conditions.

1.4. Comparison between isothermal and combusting flows

Aerodynamic and mixing studies at IJmuiden were directly linked with the furnace trials. Often in order to comprehend the aerodynamic picture of the flow of the gases inside the furnace, the velocity distributions were measured close to the burner exit with only air flowing through the burner inlets (non-combusting experiments). Additional experiments were carried out (BEE 62/1, BEE 62/2) in aerodynamic laboratories using small scale water and air models.

1.5. Near field aerodynamics of swirl stabilized burners

Results of the burner scaling trials (SAL 81, LAW 87/1) were frequently discussed at numerous technical meetings in particular at the Flame Aerodynamics Panel. Gradually it became apparent, that the inability in providing a complete interpretation of the burner scaling trials stemmed from the lack of understanding of the near burner zone aerodynamics and mixing.

1.6. Cold model studies

After the initial desk studies (BOR 83) which led to formulation of the NFA program, the first phase (NFA-1) was commenced. In this phase, a number of cold swirling flows were visualized and a few selected flows were measured using a hot-wire technique.

1.7. Inviscid flow analysis

An inviscid flow analysis was applied in interpretation of the NFA experimental results. A mathematical model based on Batchelor’s solution (BAT 85) of an inviscid equation of steady axi-symmetrical solid body rotation vortex, provided considerable help in understanding the flows.

1.7. Turbulent flow analysis

At the IFRF, turbulent flow analysis of the NFA isothermal flow cases started in 1985, initially through a cooperation (WEB 86) with the University of Sheffield. Later the same year, a copy of the Computational Fluid Dynamics code "Fluent" was installed at IJmuiden and the NFA data were used to validate several turbulence models (VIS 87, WEB 90).

1.8. Combusting swirling flows

When a fluid parcel enters the front of a swirling flame, it is rapidly accelerated. This acceleration is due to combustion and when assembled over a large number of fluid parcels, it alters the flow pattern and turbulence. This alteration to the flow pattern depends not only on the magnitude of the acceleration but also on the position of the flame front.

1.9. Diffusion type-2 and type-1 flames

In the NFA-3 experiments the conditions for generating type-2 and type-1 flames were established and compared with the NG-1 results. This was accomplished by varying the fuel jet momentum and examining whether type-2 and type-1 flames were generated.

1.10 References

Laboratory scale and semi-industrial scale experiments play an important role in development of burners and combustion systems. Results obtained on small and semi-industrial scale burners can be extrapolated to the full industrial scale when scaling rules are used to account for the differences in thermal input range. (Photo: One of a series of experimental burners manufactured for the CEMFLAME2 experiments)

2.1 Trials on gas and coal scaling

Overview table of trials on gas and coal scaling held at IJmuiden.

2.2. Basic investigations

2.3. Combustion system scaling

2.4. Scaling of natural gas flames (scaling 400 study)

2.5. Scaling of coal flames

2.6. References

The first trials on natural gas combustion took place in September 1968 and these were carried out by and for the Shell Research Station at Egham (UK). The trials were done by the Shell team using the IFRF furnace with assistance of IFRF furnace operators and technicians.

3.1. Trials on Gas Combustion at IJmuiden

Overview table of trials on gas combustion at IJmuiden.

3.2. Early Natural Gas Combustion Research

The first trials on natural gas combustion took place in September 1968 and these were carried out by and for the Shell Research Station at Egham (UK). The trials were done by the Shell team using the IFRF furnace with assistance of IFRF furnace operators and technicians. The objective was to explore optimal conditions of a specific type of reactor designed to produce highly radiative natural gas flames.

3.3. Recent Natural Gas Research

Over the last ten years or so, tremendous progress has been made in development of low-NOx burner technology for natural gas combustion. At the beginning of nineties the work at Tokyo Gasled to the development of an advanced fuel direct injection (FDI) concept (NAK 90/2;MAT 95).

3.4. Low-NOx Burner Designs at IJmuiden

In 1991, the IFRF designed a series of natural gas burners for SCALING 400 studies (WEB 91; WEB 93/2). The burners featured a central gas injector and individual gas spuds located on the burner circumference. Fig. 4.3 shows NOx characteristics of a 12 MW version of the burner tested at John Zink Co. Very substantial NOx reduction was achieved when either 80% or 100% of fuel gas was provided through the individual gas spuds.

3.5. Excess Enthalpy Combustion

The excess enthalpy combustion process (TAN 95) developed and promoted by a number of companies in Japan, including Nippon Furnace (NFK), is based on the principal that furnaces can be operated with furnace exit temperature being only slightly lower than process temperature. Such a modern furnace is equipped with an efficient system of heat recovery which allows the combustion air to be preheated to levels of 1300°C or higher.

3.6. Combustion of Natural Gas with Oxygen

Combustion of natural gas with oxygen is becoming increasingly popular with, among other industries, glass manufacturers and iron and steel producers. The reasons for this interest come from the improved efficiency of the processes that can be achieved using oxy-combustion.

3.7. References

4.1. Trials on Oil Combustion at IJmuiden

Overview table of trials on oil combustion at IJmuiden.

4.2. Non-swirling jet flames

In all the experiments on oil flames which had been carried out till 1960, blast atomization was used and either steam or air was applied for atomization. Developments in gas-turbine technology sparked a wider interest in pressure atomization. The first three trials with pressure jets (BEE 61/1; BEE 62/3; BEE 63/1; BEE 62/4; BEE 64/1; BEE 65/1) were carried out in the beginning of the sixties in furnace No. 2.

4.3. Swirl stabilized flames

The third pressure jet trial (BEE 65/1) set up the research scene for the years to come. For the first time in IJmuiden, swirl was applied to the combustion air stream by introducing a portion of the combustion air tangentially to the burner. No stabilizer disk was used. It was observed that a stagnation zone was formed in front of the burner and its presence enhanced the flame stability.

4.4. Reduction of NO_x emissions from oil flames - trials in the Air Pollution Series

The AP-8 and AP-9 trials were designed to investigate NOx reduction from oil and gas fired industrial furnaces by burner modifications. The objective of these trials was to provide the Dutch Government with the information needed for the forthcoming NOx emission legislation. Experimental data were generated in the AP-8 trials from the optimization of emissions from a number of commercial burners.

4.5. References

In trials on pulverized coal combustion, the double concentric jet type burners were used at IJmuiden in the fifties and in the early sixties, Fig. 2.1. Two basic flow conditions were investigated corresponding to boiler burners and cement kiln burners. In the boiler conditions the central jet velocity was considerably lower than the annular jet velocity.

5.1. Trials on Pulverized Coal firing

Overview table of trials on pulverized coal firing held at IJmuiden.

5.2. Non-swirling jet flames

In trials on pulverized coal combustion, the double concentric jet type burners were used at IJmuiden in the fifties and in the early sixties, Fig. 2.1. Two basic flow conditions were investigated corresponding to boiler burners and cement kiln burners. In the boiler conditions the central jet velocity was considerably lower than the annular jet velocity. For the cement kiln configuration the velocities of the central and annular jets were of the same order.

5.3. Swirl stabilized flames

Already in 1962, swirl was applied to pulverized coal burners (BEE 67/1). However, it was not until 1966-1967 then systematic investigations on the subject were initiated (LEU 67/2, LEU 70/2, HEI 69). The original objectives were to investigate how to produce early and stable ignition and high-intensity combustion in pulverized coal flames by applying swirl either to the secondary combustion air stream or to the primary coal-air jet.

5.4. Reduction of NOx emissions from coal flames - trials on the Air Pollution series

At the end of the sixties, nitrogen oxides (NOx) began to be regarded as atmospheric pollutants, and recognition of this fact resulted in public pressure to reduce their emissions from stationary combustion sources. The AP-1 trial was the first series of experiments at IJmuiden dedicated to the reduction of NOx emissions by burner modifications.

5.5. Low NOx burners for pulverized coal

A substantial body of work was concerned with the generation of knowledge required for both design and performance optimization of burners fired with pulverized coal. Several experiments were dedicated to performance optimization of the Externally Air Staged Burner. Two new burner concepts were developed: the Aerodynamically Air Staged Burner and the Air-Staged Precombustor Burner.

5.5.1. Externally Air Staged Burner
5.5.2. Air Staged Precombustor Burner
5.5.3. Aerodynamically Air Staged Burner

5.6. NOx Reburn Technology

Reburning is a combustion technology, which decreases NOx concentration in combustion products using CHi radicals as NOx reducing agents. The NOx reduction procedure is based on Myerson principle (MYE 57) that CHi radicals can react with NO converting it to HCN species, which subsequently can be transformed to molecular nitrogen.

5.6.1. Internally Fuel Staged Burner
5.6.2. In furnace NO_x reburning
5.6.3. Optimization of reburn jet penetration and mixing

5.7. Tangentially fired boilers

Most of the burner development work, which was carried out at IJmuiden, was relevant to wall-fired boilers. However, a desk study (VIS 91) and furnace trial (WOY 93/2) was dedicated to tangentially-fired boilers. In the furnace trial (WOY 93/2) the objective was to evaluate the combustion of a tangentially-fired burner cell by applying the low-NOx techniques commonly used in tangentially-fired boilers.

5.8. Combustion of coal blends in furnace trials

Studies on coal blends included several furnace trials (NAK 90/1). The AP-20 trials were the first (NAK 90/1) IJmuiden experiments on the effect of coal blends on the performance of the Aerodynamically Air Staged Burner under staged and unstaged burner operation. The experiment was executed in two phases. In the first phase, a semi-anthracite and a high-volatile bituminous coal were blended and fired.

5.9. Combustion of pulverized coal with oxygen and recycled flue gas

Fossil fuel power stations have been targeted for application of carbon dioxide removal technologies. Ultimate solution with respect to mitigating carbon dioxide emissions from conventional pulverized coal fired combustors involve separation of the carbon dioxide from the flue gas with subsequent non-atmospheric disposal.

5.10. References

Through the eighties, coal combustion research formed a significant fraction of activities at IJmuiden. Various burner development projects, as well as mathematical modeling activities, called for a technique to characterize specific coals. This demand led to the development of one of the most successful series of experiments: the Coal Characterization (CC) series.

6.1. Trials on fuel characterisation

Overview table of trials on fuel characterisation at IJmuiden.

6.2. Coal characterisation

Through the eighties, coal combustion research formed a significant fraction of activities at IJmuiden. Various burner development projects, as well as mathematical modeling activities, called for a technique to characterize specific coals. This demand led to the development of one of the most successful series of experiments: the Coal Characterization (CC) series.

6.3. References

At the end of the seventies, regulations of sulfur dioxide emissions for coal and oil fired power plants were under intensive discussions. The issue was whether the proposed emission regulations could be met by in-furnace sulfur capture using calcium based sorbents, as an alternative technology to flue gas scrubbing.

7.1. Trials and studies on SO₂ capture

Overview table of trials and studies on SO₂ capture at IJmuiden.

7.2. Direct sulfur dioxide capture

At the end of the seventies, regulations of sulfur dioxide emissions for coal and oil fired power plants were under intensive discussions. The issue was whether the proposed emission regulations could be met by in-furnace sulfur capture using calcium based sorbents, as an alternative technology to flue gas scrubbing.

7.3. References

8.1. Projects on measurements equipment at IJmuiden

Overview table of projects on measurements equipment.

8.2. Laser doppler velovimetry in semi-industrial flames

Nowadays, LDV measurements in laboratory scale flames (a few kW thermal input) do not cause any difficulties and are routinely performed. The receiving optics, kept outside the flame, is positioned either on the opposite side of the transmitting optics (forward scatter mode) or on the same side (back scatter mode). In the former case, the resulting signals are much stronger, and the signal to noise ratio is high.

8.3. LDV water-cooled probe

The introduction in 1986-1987 of laser velocimeters with fiber links provided new opportunities for measurements in large scale combusting flows. Most et al. (MOS 90) used a 30mm LDV probe enclosed in a 3m long water-cooled jacket while Ereaut and Gover (ERE 91) used a 25mm probe enclosed into a 70mm outer diameter, 5m long water-cooled jacket.

8.4. Laser sheet visualisation (Mie Scattering)

In fluid mechanics literature several visualization techniques have been used for decades. The Laser Sheet Visualization enables a flow and mixing visualization in a plane cutting through the flame. It is an attractive alternative to more restricted and slower methods commonly used in the industry such as smoke visualization. The Mie scattering visualization technique has been originally developed and tested in laboratory flames (BEC 67; SHA 77).

8.5. Visualisation equipment and setup

Planar measurements are obtained by illuminating the near burner zone of Furnace No. 1 with a laser sheet, Fig. 12.4. The light sheet, which can be up to 1m high, is produced by expanding the beam of a pulsed ND:YAG laser through a cylindrical lens. The laser produces 25 pulses per second at a wavelength of 532 nm and an energy of 190 mJ per pulse. The laser sheet is introduced, Fig. 12.4, from below the furnace outlet and its vertical plane crossed the burner and the top and bottom jets of the burner.

8.6. Image analysis

The series of visualization images, Fig. 8.5, transferred to the computer hard disk is automatically analyzed with a specially developed mixing analysis software. In order to relate the image gray level to a flow (particles) concentration, the software first rescales the images gray levels so that the maximum signal intensity measured in the jet potential core corresponds to 100% and the minimum signal intensity corresponds to zero.

8.7. Example of visualisation

The LSV visualization technique can be used in both engineering and scientific work. In engineering projects the technique is an excellent tool for visualizing: jets, sprays, Fig. 12.6, flow streams in burners and their interaction. For these applications, no quantitative information is needed and the engineer may quickly obtain information about flow symmetry, jet half-expansion angle, its trajectory. By performing a simple image analysis it is possible to identify regions of intensive and delayed mixing.

8.8. Spray characterisation

Knowledge of droplet size, velocity and concentration of oil droplets leaving the atomizer is essential for designing low-NOx oil burners. While performing numerical simulations of oil flames, the initial spray parameters (momenta, velocities, drop size distribution) have to be known accurately. At IJmuiden, a Phase Doppler Particle Analyser (PDPA) (BAC 80; BAC 84) is used for characterization of non-reacting sprays.

8.9. Temperature measurements

At the IFRF, suction pyrometers are used for measurements of in-flame temperatures. They are typically equipped with a type-B thermocouple (Pt 6% Rh/Pt 30%Rh) which has a melting point of 1830°C. The pyrometers are available in different versions and sizes. Over the last three years, substantial efforts were allocated to development of a technique that would allow for measurements of temperatures well in excess of 1800°C.

8.10. Chemistry and radiation measurements

At IJmuiden, water-cooled probes are used to sample the flame gases for subsequent chemical analysis. Several gas sampling probes are available. Recently, a special gas sampling probe for measurements in oxy-fuel flames was developed (LAL 96/1).

8.11. Slag deposit probes

Several IFRF projects were related to mineral matter transformation and slagging in pulverized coal combustion. The primary objective of these trials was to relate the nature of the slag deposits to the changes in the minerals prior to deposition (BRI 89; BRI 94/1; BRI 94/2).

8.12. References

The goal of all research carried out at the Research Station is to find application in industrial heating processes. Regularly Applied Research Programmes are undertaken as an interim step prior to ultimate industrial application. In this chapter, we present examples of applied research carried out at IJmuiden.

9.1. Applied research trials at IJmuiden

Overview table of trials on applied research at IJmuiden.

9.2. Natural gas combustion in glass melting furnaces

Glass melting furnaces produce high levels of thermal NOx, typically in the range 900 to 2800 ppm at 3% O2. When considering the increased utilization of natural gas in the glass industry, NOx reduction techniques require evaluation and the effect of their implementation on heat transfer should be established.

9.3. Research for cement kiln applications (CEMFLAM)

In the cement industry, approximately 40-50% of the costs encountered in clinker production is the cost of the energy. Consequently, maintaining high combustion efficiency in the cement kiln is of paramount importance. The flames must meet stringent requirements with respect to heat release profiles and flame shape to ensure acceptable product quality and overall process efficiently.

9.4. References

Overview table of other research trials held at IJmuiden.

11.2. Multiple burner trials

Three trials investigating multiple burners were carried out in the period 1971-1976. The first trials, MJ-1, were preliminary experiments to assess how the proximity of other flames would affect the performance of a given burner. The trials were primarily directed towards investigation of the effects of burner spacing, burner array, swirl intensity and swirl direction on flame stability, turndown and flame length.

11.3. Blast furnace trials

The development of the blast furnace process has been long influenced by economic factors, particularly coke cost. Efforts have been made to reduce the coke rate per ton of hot metal. Metallurgical coke is an expensive specialized product and partial replacement of the coke injection by cheaper fuels has become one of the main research topics on blast furnaces.

11.4. Combustion of "off-specification" fuels

For the first time coal-water slurry was combusted at IJmuiden in 1983, in conjunction with the HO-2 project (ENG 84) carried out for Hoogovens. The subsequent investigations CWS-1 and CWS-2 (ENG 86; BOR 84) dealt with burner design aspects for production of stable, unsupported flames burning coal water slurries in thermal conditions representative of boilers, which were originally designed for heavy fuel oil firing.

11.5. Combustion of lean gases

Two investigations (SCH 85/3; SCH 85/4), the G-2 and G-3 trials, were concerned with the utilization of gases on the steel industry. The trials were based on an extreme future scenario of gas supply, in which, it was assumed that the gaseous effluent from oxygen steel making processes could be mixed into the existing fuel supply systems.

11.6. Combustion of biomass and waste materials

There has been a considerable interest in recent years in the pyrolysis of biomass for the production of liquid fuels and chemicals. The liquid produced from the pyrolysis of biomass (bio-oil) is a mixture of oxygenated hydrocarbons and residual char particles.

11.7. Biomass combustion

The objective of the BIO-1 investigation (KAM 92/2) was to evaluate the combustion and pollutant formation characteristics of two bio-oils derived from slow pyrolysis processes and compare the performance with heavy fuel oil. The experiments were executed in water-cooled Furnace No. 2. The furnace was modified to simulate the thermal environment of a small industrial boiler or process heater by partial refractory lining of the furnace.

11.8. Co-firing of biomass and waste materials with pulverised coal

Substantial amounts of biomass wastes and municipal sewage sludges are produced. Co-firing of waste biomass with pulverized coal in utility boilers can help in disposal and assist in reducing CO₂ and NO_x emissions. The combustion technology required to co-fire these biomass wastes with coal has not been fully developed and effects related to NO_x, SO_x and unburned hydrocarbons emissions, particulates emissions and the incidence of slagging, fouling and corrosion remain unknown.

11.9. Multi-fuel burner programme

The Multi-Fuel Burner (MFB) program commenced in 1990, with the objective to generate engineering information required for designing low-NOx multi-fuel burners capable of firing natural gas and/or oil with acceptable carbon monoxide and particulates emissions. Several burner concepts were tested during the program execution and these include a natural gas burner, and two burners for heavy fuel oil combustion.

11.10. References

12.2. References