D6.1 Optimised operation of an oxyfuel cement plant

Before proceeding to the design of a full oxyfuel cement pilot plant (TRL 7) the individual testing of key oxyfuel components (burner, calciner, cooler) has been executed based on the theoretical findings from the ECRA CCS project. Testing and demonstrating the oxyfuel calciner, clinker cooler and burner under industrially relevant conditions were undertaken in CEMCAP project (corresponding to reaching TRL 6). Due to the counter current flow in the clinker burning process and the recirculation of gases in the oxyfuel process, changes of the operational parameter in one process unit influence connected equipment units. An overall modelling of the oxyfuel operation is necessary. For this purpose the VDZ process model was adapted to the outcome of the CEMCAP prototype testing by including comprehensive data from the testing and restructuring the process modules. Based on the pilot testing results the calcination process of the raw material and the heat radiation profile in the kiln could be optimised. Moreover the amount of the false air ingress especially the air leakage from the cooler could be refined.
For further optimisation of the oxyfuel process model a parameter matrix was spanned. It included different operational modes, such as varying material and volume loads, combustion characteristics (flame length and shape), false air ingress and degree of heat exchange.
The simulation results of the optimised oxyfuel process model showed that by adaptation of the burner settings to the experimental results and by switching the oxygen supply from secondary to primary gas the heat transfer from the gas to the material could be enhanced. Thus, the heat transferred to the material in the sintering zone and the temperature profiles along the kiln are optimal to generate the required clinker phases. Furthermore it can be expected that the coating behaviour of the material in the kiln and the thermal load of the rotary kiln are similar in the optimised oxyfuel operation and the reference air case and therefore achieve an optimum operational mode. For the reference air case (optimum case) simulation results of a clinker burning process with best available techniques (BAT) and a clinker production of 3000 t/d was used. For equipment protection in the oxyfuel process with higher calcination temperatures the degree of calcination at the kiln inlet has been slightly decreased.
After the optimisation of the oxyfuel process model the waste heat recovery and the heat integration were evaluated with the help of an iterative procedure between the VDZ process model (PM) and the heat integration model. In order to evaluate the energy demand of the oxyfuel clinker burning process for different operation scenarios, the VDZ process model was applied for six different operation conditions, in which the false air ingress and the preheater stages were varied. Based on the process model results the process and energetic integration of the air separation unit (ASU) and CO2 purification unit (CPU) and options for power generation by an Organic Rankine Cycle (ORC) were investigated by SINTEF.
The simulation results showed that the thermal energy demand of the clinker burning process rises by 0.8 – 1.3% per 2% of false air ingress mainly caused by the heating of the additional air. Simultaneously the specific power consumption of the CPU increases by 2.7 – 3.5% per 2% of false air ingress in the considered range of 4.6 – 8.1% due to the dilution of the flue gas by air. As the electrical energy demand for the CPU is rising exponentially with increasing false air ingress, the maximum acceptable level of false air ingress is around 8%.
This shows that the requirement for regular maintenance is much higher for the oxyfuel clinker burning process than for the clinker burning process with air operation.
The influence of the number of preheater stages on the ORC power generation was evaluated. The simulation results showed that decreasing the number of preheater stages leads to higher energy consumption in the ASU and CPU. With the consideration of the increased energy consumptions of ASU and CPU the ORC efficiency with the additional fuel used in the case of 4 and 3 preheater stages is +5.6% and +15.2%. It is obvious that from energetic perspective reducing the number of preheater stages to increase the ORC performance is not productive.
The adaptation of the VDZ process model to the experimental results and the optimised oxyfuel process model simulation results showed that a retrofit of existing plants is possible. But in order to refine the experimental results and evaluations and for validation of the simulation results and conclusions, experiments at a full scale oxyfuel cement pilot plant (TRL7) will be necessary.