MODELLING OF DESULPHURIZATION PROCESS WITH THE [615843]

MODELLING OF DESULPHURIZATION PROCESS WITH THE
SCOPE OF SO 2 EMISSIONS DECREASE AND INCREASE OF
SYNTHETIC GYPSUM QUALITY
Tudor Prisecaru 1
1University POLITEHNICA of Bucharest, Faculty of Mec hanical Engineering and
Mechatronics, 313 Splaiul Independentei Street
2CEPROCIM S.A., 6 Preciziei Street, Bucharest, Roman ia
An experimental model for wet flue
presented. The model was used to study the modelling of desulphurization process from which
synthetic gypsum it is formed. Results further showed that the process must be contr olled in order that
gypsu m is settling in good conditions.

Keywords : fluid dynamics, mass transfer,

1. INTRODUCTION

Flue gases d esulphurization
for decreasing sulphur dioxide (SO
fossil fuels, with the scope of complying with the emission limit values (ELV).
desulphurization also allows obtaining high quality co
materials [4] in different fields. The objective of the study is the removal of SO2 th rough
chemical and physical procedures
and co-products [7-9] into the economic value chain. A dynamic model has been developed to
simul ate system performance using the operating data pro vided by a coal

2. METHODOLOGY

For performing laboratory tests an experimental ins tallation was acquired for
desulphurization of gases.
Figure 1. a) and b) Flue gas MODELLING OF DESULPHURIZATION PROCESS WITH THE
EMISSIONS DECREASE AND INCREASE OF
SYNTHETIC GYPSUM QUALITY

1, Malina Mihaela Prisecaru 1, Razvan Lisnic 2*
University POLITEHNICA of Bucharest, Faculty of Mec hanical Engineering and
Mechatronics, 313 Splaiul Independentei Street , Bucharest, Romania
CEPROCIM S.A., 6 Preciziei Street, Bucharest, Roman ia

ABSTRACT
An experimental model for wet flue gas desulfurization process with a limestone
presented. The model was used to study the modellin g of desulphurization process from which
Results further showed that the process must be con trolled in order that
m is settling in good conditions.
fluid dynamics, mass transfer, wet FGD, SO 2 emissions, synthetic gypsum
esulphurization [1-3] is a technique through which flue gases are treated
for decreasing sulphur dioxide (SO 2) concentration produced during combustion process of
fossil fuels, with the scope of complying with the emission limit values (ELV). Flue gases
esulphurization also allows obtaining high quality co -products that can be used as raw
The objective of the study is the removal of SO2 th rough
chemical and physical procedures [5,6 ], as well as the integration of waste water manage ment
into the economic value chain. A dynamic model has been developed to
ate system performance using the operating data pro vided by a coal -fired power plant.
For performing laboratory tests an experimental ins tallation was acquired for
a)
Flue gas desulphurization experimental installation MODELLING OF DESULPHURIZATION PROCESS WITH THE
EMISSIONS DECREASE AND INCREASE OF
University POLITEHNICA of Bucharest, Faculty of Mec hanical Engineering and
, Bucharest, Romania
desulfurization process with a limestone slurry is
presented. The model was used to study the modellin g of desulphurization process from which
Results further showed that the process must be con trolled in order that
gases are treated
concentration produced during combustion process of
Flue gases
products that can be used as raw
The objective of the study is the removal of SO2 th rough
], as well as the integration of waste water manage ment
into the economic value chain. A dynamic model has been developed to
fired power plant.
For performing laboratory tests an experimental ins tallation was acquired for
b)
experimental installation

116
For maximizing the efficiency of flue gases desulph urization and increasing of
synthetic gypsum quality influencing the following parameters was considered:
depressurization, interaction time (flowing rate of desulphurization solution) and oxidizing
atmosphere (oxidizing conditions) for ensuring the transformation of sulphite ions into
sulphate ions. Limestone dosing into the aqueous su spension was imposed at 20% and 30%.
Further measurements of gypsum quality will restric t dosing range.
To assess the interfacial mass transfer CFD models prepared were post processed and
reaction rates were represented alongside several a xes within the tank, as it is shown in fig 2.

Figure 2. Axes for which reaction rates were proces sed

These axes are located at three points at 100mm, 30 0mm and 400mm heights. From their
location point of view three of them are on the sym metry plan of the tank and three are in
lateral at 50 mm from the wall of the secondary rea ctor’s tank.
In figures 3-8 reaction rates for gypsum forming ar e presented after all six axes described
above.

Figure 3. Reaction rate for gypsum forming after as ymmetrical axis 1

117

Figure 4. Reaction rate for gypsum forming after as ymmetrical axis 2

Figure 5. Reaction rate for gypsum forming after as ymmetrical axis 3

Figure 6. Reaction rate for gypsum forming after ce ntral axis 1c

Figure 7. Reaction rate for gypsum forming after ce ntral axis 3c

118

Figure 8. Reaction rate for gypsum forming after ce ntral axis 2c

Analysing the variation graphs in figures 3-8 it wa s observed that reaction rates are
ranging between 10 -5 – 10 -3 kmol/(m 3s) values. Considering that superior values are nor mal
and taking into consideration the concentrations of reactants for reaction 1, the need to
accelerate mixing processes was observed in areas w here reactions slow down below 10 -3
kmol/(m 3s). This means introducing a geometrical or mechani cal agitation system. In the first
phase a geometrical system will be used through app ropriate wall profiling considering also
the flow spectrum in respective areas.
Generally, it is observed that reaction rates are h igher in the first 300 mm of the tank.
Also, this area is split in two regions depending o n the axis out of which values were
registered. This means that in the rest of areas ei ther reactants were depleted either the way of
mixing is poor. On the two subzones samples will ne ed to be extracted in order to identify the
causes that lead to their differentiation. Also, fl ow spectrums of reactants will need to be
overlapped on the variation graphs of reaction rate to highlight the causes of divergence
between the current lines of different components.
It was observed that the process in the second tank is completely without reactions.
This conclusion can lead also to the idea of reduci ng its volume depending on the results that
will be obtained from solving the first conclusions .
Currently only reaction tank 1 was considered, the one that produces calcium sulphite,
as being determining and strictly necessary for sec ond reaction taking place. Because of this
reason it was considered that in this phase of the research it is sufficient to analyze the mass
transfer just for the first reaction then after opt imizing this process the process governing the
second reaction is also optimized.
Desulphurization installation includes a first step that is represented by a vertical
reactor that also contains a scrubber and then the second step that is represented by a tank
meant for finalizing the chemical reactions and set tling the water compared with the gypsum
solution.
The issues that are followed by the numerical model ling – in this step – are
represented by determining the flow rate, temperatu res and pressure regime that must be kept
in the second step of the installation in order to obtain higher concentrations of gypsum in
aqueous solution.
Modelling the tank that represents the second step was done according to figure 9.

119

Figure 9. Geometrical model of reactor’s
second step

Figure 10. Modelling with hexahedral
synthesis elements of reactor’s second step

Modelling was based on a discretization in a number of approximately 1.5 million pyramidal
elements. For a reduction of calculation volume, it was chosen to group the pyramidal
elements into hexahedral elements according to figu re 10.
The process inside step two of the installation con sists in the penetration of a gas-
water emulsion composed of nitrogen, carbon dioxide , oxygen, sulphur dioxide, and water as
liquid.
This emulsion was introduced with a rate of 0.7 m/s to not generate a higher
turbulence and to disturb the settling process. The emulsion flow was correlated for an
average equivalent regime of the pilot burning poin t of 750 kWt. After performing numerical
simulation tests the following results were obtaine d regarding gypsum formation, as it is
observed in figure 11-13.

Figure 11. Late formation of gypsum.

Figure 12. Rapid formation of gypsum at high turbul ence.

120

Figure 13. Normal formation of gypsum.

3. CONCLUSIONS

In figure 11 we can see the gypsum is formed late a nd settling will be difficult to be achieved.
In figure 16 gypsum is formed immediately but this is due to a turbulence increase, which
makes the separation difficult. In figure 17 a norm al gypsum formation process is observed
which allows its settling in good conditions.

Acknowledgement
This work was supported by a grant of the Romanian National Authority for Scientific
Research and Innovation, CNCS/CCCDI – UEFISCDI, pro ject number PN-III-P2-2.1-PED-
2016-0867, within PNCDI III.

References
[1] R. Lisnic, S. I. Jinga, Study on current state and future trends of flue gas desulphurization
technologies: a review, Romanian Journal of Materia ls 2018, 48 (1), 83 – 90
[2] Patricia Córdoba, Status of Flue Gas Desulphurisati on (FGD) systems from coal-fired
power plants: Overview of the physic-chemical contr ol processes of wet limestone
FGDs, Fuel, Vol. 144 , March 2015, 274-286
[3] Sectorial Guidelines for EIA: Projects for Flue Gas Desulphurization Installations
in large combustion plants, Joint Assistance to Sup port Projects in European Regions,
Romania, 2014
[4] A-M Dragomir, R. Lisnic, T. Prisecaru, M. M. Prisec aru, C. A. Vîjan, D. C. N ăstac, Study
on synthetic gypsum obtasined from wet flue gas des ulphurisation in thermal power plants,
Romanian Journal of Materials 2017, 47 (4), 551 – 556
[5] M. Ploscaru, A.S. Sandulescu, Studiu privind ut ilizarea tehnologiilor de desulfurare la
CTE din bazinul Oltenia, SERMSTudo, CD-ROM, ISBN:97 8-606-10-0592-5, 2013
[6] Flue gas desulpurization wet limestone-gypsum p rocess, KC Cottrell Co., Ltd., 2016
[7] M. C. Caillahua, F. J. Moura, Technical feasibi lity for use of FGD gypsum as an additive
setting time retarder for Portland cement, Journal of Material Research and Technology, 2017
[8] G. Tzouvalas, G. Rantis, S. Tsima, Alternative calcium-sulfate-bearing materials as
cement retarders: Part II. FGD gypsum, Cement and C oncrete Research, 34 (11), 119-125,
2004
[9] R. P. Borg, C. Briguglio, V. Bocullo, D. Vai čiukynien ė, Preliminary investigation of
geopolymer binder from waste materials, Romanian Jo urnal of Materials 2017, 47 (3),370-378