Contents Issue 1 (1999)

If you are interested in ordering back issues, please use this form

Michael Rziha and Rainer Wulff:

Cycle Chemistry in Combined Cycle Units – The Siemens Experience

The development of the combined cycle plants during the last decade had brought up a wide range of variations in respect of plant and boiler design and operational aspects (base load - weekly start / stop - daily start / stop). This wide range of different design and operation features is also reflected in the water chemical procedures for the unit treatment. In classical conventional drum boilers, either AVT (all volatile treatment) or a solid alkalization of the boiler water by trisodium phosphate or caustic soda had been usually used. In combined cycle plants, the choice for the chemical treatment is not as easy as it is for the conventional plants. For typical IPP plants, daily start-stop operation can be seen as "normal operation". Therefore, a solid alkalization cannot be recommended under the operational point of view. On the other hand, for the LP-boiler a solid alkalization might be required in order to control flow accelerated corrosion effects. Consequently, a mixed treatment (AVT for the HP & IP boiler and solid treatment for the LP boiler) becomes increasingly common. However, also with AVT in LP boilers good experience had been made in the past. The Siemens experience of the last 8 years with combined cycle plants different chemical operating conditions is presented.

PowerPlant Chemistry 1999, 1(1)

Digby D. Macdonald, Iouri Balachov, and George Engelhardt

Deterministic Prediction of Localized Corrosion Damage in Power Plant Coolant Circuits

The need to avoid unscheduled downtime, and the desire to extend the operating lives of thermal power plants, because of the difficulties in licensing new plants, has led to the development of advanced technologies for predicting the accumulation of corrosion damage in the heat transport circuits. However, power plants, even of the same design, and like all complex machines, become unique systems after only a short time of operation, because of differences in operating history and philosophy. Accordingly, it has proven to be extraordinarily difficult to accumulate sufficient empirical data, even over a large population of plants, to formulate effective empirical models for predicting damage, partly because corrosion-induced failures are rare (but costly) events. The alternative philosophy is determinism, which employs mechanistic models, whose outputs are constrained by the natural laws (conservation of mass, energy, momentum, charge, etc.). Deterministic models are characterized by a minimal need for calibrating data from a plant, because the model parameters normally can be derived by independent experiment. Furthermore, important relationships (e.g. the dependence of crack growth rate on conductivity), which are established as a consequence of the imposition of the natural laws in a deterministic model, might be impossible to establish empirically to the desired accuracy, because of the inadequacy of experimental systems. In this paper, we outline the application of deterministic models to the task of predicting the accumulation of damage in the primary heat transport circuits of Boiling Water Reactors.

However, we emphasize that the methods that we describe are generally applicable to all systems that employ water or an aqueous solution as a coolant, including PWRs, CANDU reactors, and RBMKs.

PowerPlant Chemistry 1999, 1(1)

Gernot Repphun, Andreas Hiltpold, Irene Mailand, and Bernhard Stellwag

Influence of Metal Addition to BWR Water on Contamination and Corrosion of Stainless Steel

The aim of this investigation is to minimise the activity build up (especially of Co-60 and Co-58) in the oxide layers of austenitic steels under simulated BWR conditions by variation of pH and dosing of metal ions. Stainless steel samples were exposed to water in 1l autoclaves in flow through mode (1l/h) over a period of 5 days at 288 °C with addition of metal ions and radioactive tracers. In addition to the activity measurements the semiconducting properties of the oxides and oxide layer thickness were determined by photo-electrochemistry and secondary ion mass spectroscopy. Correlations of activity uptake with layer thickness, type of semiconductor and band gap energy of the corrosion scales is shown. The presence of different metal ions in the oxidation process implies changes of the semiconducting properties of the oxides and different susceptibility of activity uptake into the oxide layers.

PowerPlant Chemistry 1999, 1(1)

James C. Bellows:

Chemical Processes in Steam Turbines

Turbines can last for many years without significant corrosion. Some do. Some do not, depending upon their environments. The thermodynamic processes and kinetic processes that control turbine corrosion are reviewed. Thermodynamics indicates that for most impurities, corrosion will be near the transition from superheated to wet steam. Homogeneous nucleation processes are too slow to be a significant process in transferring corrodents to a turbine.

Heterogeneous nucleation can be expected to be too rapid to limit turbine corrosion. Mass transfer of corrodents from bulk steam to the turbine surface is the most likely corrosion rate limiting process. Impurities can be deposited quickly but can be removed only very slowly. Control of steam borne corrodents must start at plant commissioning and extend through the life of the turbine.

PowerPlant Chemistry 1999, 1(1)

Joachim Fahlke:

RO Applications

Since 1982 makeup water treatment plants with integrated reverse osmosis were operated. The paper describes the technique of both plants (one from 1982 and one from 1996) in detail and shows some aspects of the development over a time period of 13 years. The advantages and disadvantages during the daily practical use were described.

Waste Water Reuse

Recycling of cleaned ash cooling water for process water in the FGD is principally possible and done for 8 years now, as is demonstrated by the example of Unit 7. The RO concentrate is recycled since November 1997. Since the start of this approach a total amount of 799,288 m ³ ash cooling water and 198,778 m ³ RO concentrate were recycled in Units 7 and 8. For Units 3 and 4 the recycled quantities of ash cooling water and RO concentrate amount to 688,982 m ³. This equals DM 168,704 of saved well water extraction fees since the start of the recycling approach and the natural well water resources were conserved.

PowerPlant Chemistry 1999, 1(1)

Daniel E. Meils:

Performance Assessment of Chemistry On-line Process Instrumentation

The electric power industry is relying more on installed chemistry on-line process instrumentation analysis and less on traditional bench top chemistry analysis for power plant operations. As the industry relies more on these on-line instruments to operate their power plants, it becomes increasingly important to insure they provide results that meet the accuracy and precision requirements of the users. Requirements for both accuracy and precision must be clearly specified. When establishing these requirements, consideration should be given to the importance of the results to safety, regulatory limits, criticality to operation, consequences of using bad results, and any other appropriate considerations. Requirements should be strict enough to assure results generated would be useful in deciding whether a parameter is within limits and in deciding whether the result is increasing or decreasing relative to the limits of operation. However, requirements should not be so strict that they require excessive analysis time or cost.

This article presents several performance assessment practices. These practices can be used to statistically evaluate the performance of chemistry on-line process instrumentation during initial instrument commissioning and for ongoing instrument performance assessment. By applying these methods, users can avoid the costs associated with inaccurate results and the costs associated with unnecessary maintenance.

PowerPlant Chemistry 1999, 1(1)