3 Fly ash from pilot-scale co-combustion tests
3.2 Materials and methods
Demolition wood was supplied by Electrabel Nederland from power plant ‘Centrale Gelderland’
(Nijmegen, the Netherlands). The fuel was sampled after the pre-treatment with hammer mills and micro mills. However, it was still too coarse for direct combustion. Therefore, the material was treated by a so-called Ultrarotor (Jäckering Germany) to obtain further size reduction. Poultry litter from layers was derived from a poultry farm (Mr Wisse's poultry farm, Colijnsplaat, the Netherlands). The sample was taken from the transport belt leading from the shed. The moisture content of this poultry litter is reduced by drying it with warm air from the stable, which makes it more suitable for combustion processes. The grain size of the poultry litter was reduced with a hammer mill (type Hamex, 30 kW). A sample of SRF was supplied by Eco Energy Europe (Born, the Netherlands). The fuel was sampled after shredding and pelletizing (pellet diameter 10 mm). Because of the low pellet strength, the material was again pelletized and then treated by the ultrarotor (Jäckering Germany) to obtain sufficient size reduction. The coal was from Venezuelan source (Paso Diablo). It was milled and dried by IKO (Germany).
3.2.2 Combustion experiments
The combustion experiments were performed in the 1 MWth test boiler of KEMA. The experimental boiler installation basically consists of (see figure 3.1):
A pulverised coal silo and transport system.
A pulverised secondary fuel silo and transport system.
A pulverised coal burner and combustion chamber.
A flue gas cooler.
A fabric filter with an ash disposal system.
A low and high-pressure water-cooling system.
The pulverised coal is supplied in bulk cars in a nitrogen atmosphere (see also table 3.1). The pulverised coal is transported by means of nitrogen to one of the two storage silos, which can contain 45 m3 each. Coming from the silos, the pulverised coal is weighed and transported in dosed quantities to the burner of the vertically built combustion chamber. Transportation is performed by means of air (primary air) at about 60 to 65 °C. The burner, having a capacity of 1 MW thermal, has ample adjustment facilities and is suitable for several kinds of pulverised coal. The capacity can range from 100 to 40%. The combustion air is preheated in an air heater, which is placed in the second part of the flue gas cooler. The combustion air is supplied as so-called secondary combustion air to the burner and partly (15%) as tertiary combustion air into the combustion chamber. In the co-firing experiments, the secondary fuels are introduced by means of a special dosing installation. Coal and secondary fuel were introduced into the boiler by means of a special burner for co-combustion. In this burner, the secondary fuel (and transport air) is transported through the centre and the coal (and transport air) through an annular ring around the centre of the burner.
49
Figure 3:1: Flow diagram of KEMA Test boiler. Coal and secondary fuels are introduced in the combustion chamber by a special burner for co-combustion.
The generated fly ash is separated from the flue gas with a fabric filter.
The combustion chamber is composed of a number of detachable, cooled ring sections (15 rings of 30 cm high each), which means that the chamber can be varied in height. These ring sections are provided with entry ports for observation and sampling. On the fourth ring, the tertiary combustion air is supplied to the combustion chamber. Natural gas is used for starting up and preheating the installation, after which pulverised coal can be fired. The flue gases leave the combustion chamber at a temperature of about 1150 °C and are cooled down in the flue gas cooler by means of water-cooled pipes and the previously mentioned combustion air pre-heater. Filtering of the flue gas is performed with a fabric filter, which is installed behind the flue gas cooler. After removal of the fly ash, the flue gases are transported to the chimney by means of a ventilator.
The temperature of the flame was measured with a suction pyrometer. This pyrometer consists of a cooled tube that is put in the centre of the flame. An amount of gas is extracted from the flame and brought to a thermocouple, which serves to measure the temperature.
Each secondary fuel was co-fired in three percentages ranging from 10 to 33 %. A reference fly ash (REF) was produced from 100 % coal only. The ashes are coded using the co-combustion percentage, for example DW10 is the ash from co-combusting 10% (e/e) Demolition Wood.
50 Table 3.1: Selected design data for the KEMA Test Boiler
Design data Value Unit
Temperature of flue gases at boiler exhaust Intake temperature of fabric filter
3.2.3 Chemical characterization of (secondary) fuels and fly ashes
The chemical composition of the mineral matter of the fuels and fly ashes was analysed after total digestion in line with EN 15290 [3.2] and EN 15410 [3.3]. Digestion of the fuels was performed after incineration of the organic components at low-temperature conditions. The fly ash and the mineral matter from the incinerated fuel were each melted together with lithium tetra borate to produce a glass.
This glass was digested by the following procedure: a mixture of 500 g 65% HN03, 50 g 70% HCIO4
and 450 g 38% HF was made in 1 l PE bottle. 100-500 mg of the sample was mixed with 10 g of the acid mixture to digest the sample by heating in a bomb14 overnight at 190 °C. The elements Al, Fe, K, Mg, Na, P and Si were analysed by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES). Free lime was analysed in accordance with EN 450. Cl, F and Br in coal were analysed after pyro hydrolysis. Mercury was analysed after digestion with O2 bomb.
The amount of inorganic phosphorus and organic phosphorus which is soluble in acid was analysed by ICP after extraction, according to the method of Salomons and Gerrtisen [3.4] and Barnett [3.5].
This was only performed for poultry litter because of its high phosphorus contents.
3.2.4 Characterization of mineralogy of secondary fuels and fly ash
The mineralogical characteristics were determined using two different methods, namely X-ray diffraction (XRD) and sequential dissolution. The scan range of the XRD was 5-75 2θ and 6-130 θ
14 Materials which are not fully dissolved by acid-digestion at atmospheric pressure may require a more vigorous treatment in pressure vessels lined with polytetrafluoroethylene (PTFE) glass, silica or vitreous (glassy) carbon or in sealed silica tubes; this treatment is called bomb-digestion. The test sample and acids are heated in such a closed vessel, so that the digestion is carried out at higher temperature and pressure. Source: PAC, 1988, 60, 1461 (Nomenclature, symbols, units and their usage in spectrochemical analysis-X. Preparation of materials for analytical atomic spectroscopy and other related techniques (Recommendations 1988)) on page 1469.
51
respectively. The count time was 1-2 s. The XRD analyses were performed with a Philips diffractometer PW1820 (Co X-ray tube) or a Bruker D4 Endeavor (Cu X-ray tube). If the XRD was meant to quantify the concentration of the minerals, an internal standard was added (metallic Si) before the XRD analysis was performed. These XRD analyses were performed with the Bruker D4 Endeavor. The identified minerals were quantified using the Rietveld method and adjusted to the known mass percentage of the added standard. The percentage amorphous material is assumed to be 100% minus the sum of the mass percentages of the identified minerals. The X-ray diffraction measurements (XRD) were carried out by Wageningen University/TCKI and CORUS.
The glass composition was determined based on the determination method of reactive silicon [3.6].
This was combined with so-called sequential dissolution. A key feature of the analysis is the cascade approach, which involves removing in turn the fraction that is soluble in acid and the fraction that is soluble in potassium hydroxide, and analysing each separately. The dissolved fraction from these steps is analysed with ICP-AES for Al, Ca, Fe, K, Mg, Na, S, Si and Ti.
3.2.5 Scanning electron Microscopy (SEM)
The morphology of the fly ashes were investigated using a scanning electron microscope (SEM). The SEM was a JEOL 6300. The pictures were taken in the SE-mode (secondary electrons) with an acceleration voltage of 15 kV. The samples were sputtered with gold.
3.2.6 Grain size distribution of fuels and fly ashes
Grain size distribution of fly ashes and pre-milled coal was analysed with laser granulometry (Malvern). The other fuels were analysed with a combination of mechanical sieving (coarse fraction) and laser granulometry (fine fraction).
3.2.7 Properties of cement paste and mortar with fly ash
The following characteristics were used to assess the performance of fly ash:
Setting time was tested in accordance with EN 450. This test is based on the penetration behaviour of a standard needle in a cement paste or cement fly ash paste. This cement fly ash paste consists of 75% cement and 25% fly ash. The amount of mixing water is adjusted to obtain a standard consistency. See figure 3.2.
Pozzolanic behaviour was tested in accordance with EN 450. The pozzolanic behaviour is expressed as the Activity Index (see Box 1 in 2.2.4).
The soundness of cement fly ash paste was tested in accordance with EN 450 using the LeChatelier test. This cement fly ash paste consists of 50% cement and 50% fly ash. The amount of mixing water is adjusted to obtain a standard consistency. This fresh cement paste is to fill a ring with two needles (see figure 3.3). After 24 hours of hardening, the distance between the tips of the needles is determined with a calliper. After that, the hardened cement paste together with the ring is stored in boiling water for a quarter of an hour. After that the distance between the tips of the needles is measured again. The increase of the distance before and after boiling is a measure for the unsoundness.
52
Figure 3:2: Vicat apparatus for determination of setting time, which is measured as the penetration behaviour of a standard needle in
a cement (fly ash) paste.
Figure 3:3: LeChatelier test ring, filled with cement (fly ash) paste for determination of soundness. This is measured as the increase in distance between
the tips of the needles before and after boiling the sample in water.
53