Books and Book Editorships:
E. Schachermayer, G. Bauer, E. Ritter, P.H. Brunner:
"Messung der Güter- und Stoffbilanz einer Müllverbrennungsanlage (Projekt MAPE)";
A Material Balance of the Municipal Solid Waste Incineraton Plant "Spittelau" via the Residues from Incineration - Summary
(You can order this paper at Umweltbundesamt, A-1090 Wien, Spittelauer Lände 5: (paper is written in German))
First generation measures of environmental protection have been designed to minimize the impact of pollutants from the anthroposphere into the environment in order to avoid immediate damage. This so called "end of the pipe" strategy serves as a kind of filter on the rear end of the system with the net effect of transferring the waste disposal problem from one environmental compartment into another. Thus there is need for a second generation of environmental protection measures, which optimize the whole system, based on the precautionary principle: The material fluxes through the anthroposphere should be controlled in an environmentally sound manner in oder to minimize environmental impact and to ensure an optimized and sustainable use of the natural resources.
In order to implement these strategies, more advanced techniques have to be developed. So far the most efficient of them is the method of material flux analysis, which grew out of the well known "pathway analysis". Since the variety of matter used in the four anthropogenic compartments is very large and comprises around 10,000 to 100,000 chemically defined substances in different combinations and functions, reliable information on the material fluxes can only be obtained by keeping track of elements rather than of chemical substances. This is the main idea of the material flux analysis, which then provides the necessary global "material" overview for the implementation of the new generation of environmental protection measures.
The main goal of this project was to establish a material flux analysis of the state of the art municipal solid waste incineration plant "Spittelau" in Vienna. With its installed thermal capacity of 460 MW, 60 MW of which are provided by the process of waste incineration, this plant is the largest central heat power plant in Austria. Every year ca. 270,000 tons of municipal solid waste from private households and trade and commerce are treated there.
In more detail, the primary goal of of the project "A Material Balance of a Municipal Solid Waste Incinerator Via the Residues From Incineration" was to develop a method which allows to determine
· the flux of selected elements through the municipal solid waste incineration plant Spittelau - and -
· the chemical composition of waste from private households and municipal solid waste (containing waste from private households and commerce).
Even if the waste composition can be determined by the relatively homogenous products of incineration, we have to be aware of the fact that the measuring results can vary significantly. The more we know about the material transfer during the incineration and the flue gas cleaning the better we can analyze the measuring results and the less sampling is necessary. Therefore another goal of this project was to find out if the electrostatic precipitator (ESP) dust or other products provide enough information in order to obtain reliable results.
Municipal solid waste (MSW) is a clear indicator for the consumption fluxes into and from private households. Due to the products of the waste treatment we can estimate the consumption fluxes through the households . This is of major importance because in modern societies the private household plays a key role as far as product and material fluxes are concerned.
The project consisted of two parts: the pre-experiment and the main experiment.
The extremely heterogeneous composition of municipal solid waste called for an advanced mathematical model for the optimization of sample collection and the interpretation of the obtained data. Each measuring method has its uncertainties and limits. Methodical problems concerning the optimization of the sampling were solved by feeding the data of the pre-experiment into a statistical "variance component model".
During the pre-experiment several samples were taken from slag, ESP dust, waste water and filter cake which allowed the analysis of the element concentration by chemical means. The selected elements were carbon, phosphorus, sulphur, fluorine, chlorine, iron, lead, zinc, cadmium, copper and mercury. The data provided by these samples were fed into the statistical variance component model mentioned above, which provided an estimate of the minimal number of samples necessary in the main experiment in order to achieve a pre-determined accuracy of the measured concentrations. The variance component model also allowed a detailed investigation of the statistical influences of sample collection, sample processing and sample analysis.
In the main experiment waste from private households as well as municipal solid waste was incinerated seperately. During both incineration periods the same incineration products as during the pre-experiment were sampled and analyzed.
By the help of a material flux analysis the flow of the selected elements through the plant could be determined as follows: At first the input and output products were determined and product balances for both the pre- and the main experiment were established. Based on the product balance for the main experiment a special material balance was established for each selected element. The last step consisted of defining the transfer coefficients and their statistical uncertainties.
Figure 1 shows the material balance and the system analysis for the plant, which consists of the processes "incineration", "electrostatic preciptator" and "flue gas control system" (including denox, ESP and waste water treatment). The location of the places where samples have been taken are indicated by small circles.
By chemical analysis of the incineration products like slag, ESP dust, waste water, filter cake and exhaust gas it has been possible to determine for the first time in Austria the chemical composition of "waste from private households" of an average Austrian household, and in addition the composition of municipal solid waste, including waste from trade and commerce.
Due to the very heterogenous composition of waste we expected enormous variations in the material concentration of the different incineration products. It was not until statistical methods were used that these uncertainties could be quanitfied and that the reliability of the data could be estimated. By imposing a normal distribution the statistically analyzed data from the pre-experiment provided a so-called "confidence interval" for the concentration of each selected element. Thus it was possible for the first time to determine the "mean concentration" of the selected elements within a specified error bound, of 10 to 20% depending on the specific element under consideration. The statistical approach has augmented the relability of the results considerably. The number of samples was chosen in a way that the "real value" of the material concentrations in each incineration product lay with a probability of 95% within the confidence inerval of the "estimated mean material concentration" measured during the pre-experiment. The interval width was fixed at +/- 10% of the estimated mean concentrations.
As an example fig. 2 shows the hierarchical structure of the slag sampling procedure. The three distinct levels "collective sampling", "sample processing" and "laboratory sample analysis" are described by three independent stochastic variables in the variance component model.
It has turned out that a product and material balance of a MSW incineration plant can be established at affordable costs. For future projects where a balance is to be established for a longer period of time it is necessary to monitor time fluctuations. In order to be able to compare different results the uncertainty has to be indicated by means of confidence intervals for which the description of the sample treatment, the sampling procedure and the sampling frequency are of major importance.
Fig. 3 shows as an example the graphical representation of the transfer coefficents for "MSW waste" and "waste from private households" for copper, zinc and mercury. For all other selected elements similar diagrams exist.
In addition the following conclusions have been drawn from the obtained data:
The atmophilic metals like cadmium and zinc were concentrated mainly in the ESP dust. mercury was accumulated in the neutralization sludge. Carbon was found up to 98% as CO2 in the exhaust gas, and iron was almost completely removed as iron scrap from the slag by magnetic separation.
The average composition of waste from households and municipal solid waste is presented in Table 1. These values agree very well with similar data from material balances in Switzerland and other European cities. With the exception of carbon, the incineration with suitable secondary treatment of the incineration products is a sink for the elements selected above.
Table 1: Average composition of solid waste from private households and municipal waste (dry matter)
By using statistical methods a detailed plan for an optimized sampling procedure in municipal solid waste incineration plants has been developed. This sampling plan is a first step towards the construction of an effective and powerful monitoring instrument for the entire waste mangement system. In the long term this monitoring instrument is designed not only to help keeping track of the material fluxes from the anthroposphere into the environment via the release of liquid and solid waste, but to become in addition a useful tool for policy makers by providing reliable data for legal measures concerning waste minimization.
Changes in the material input into the MSW incineration plant lead to changes in the output. If we presume that the transfer coefficients do not vary and that the proportion of an element leaving the plant depends more or less always on ist proportion when entering the plant, examination of the ESP dust will be for most atmophilic elements the best way to monitor material fluxes through the incineration plant since ESP dust is more homegenous and can therefore be more easily sampled and treated than slag for example. Online sampling will be feasible. As the variations of the concentration values are much lower in ESP dust, it is ideal for keeping track of the trends of most elements.
For the survey and optimization of the material balance in Austria this project is of great importance, since for the first time it is possible to quantify the flux of the above selected elements, especially metals, through MSW incineration to the sink "deposition of filter cake, slag and ESP dust". By complementing this data with similar data from waste water treatment, and by comparing it with data on the consumed goods, the diffuse entry of metals into the environment can in principle be estimated.
Electronic version of the publication:
Project Head Paul H. Brunner:
B (MAPE) Online-Messung der Stoffbilanz auf der MVA Spittelau - Phase B, Messjahre 2001 bis 2004
Created from the Publication Database of the Vienna University of Technology.