Available online at www.sciencedirect.
com
ScienceDirect
Procedia Engineering 83 (2014) 181 – 187
“SYMPHOS 2013”, 2nd International Symposium on Innovation and Technology in the Phosphate Industry
From lab to plant: first industrial experience of the new high efficiency
dihydrate hemihydrate process (DA-HF) process
Tibaut Theysa*, Dorina Fatia, Olivier Schrevensa, Agata Tarnowskab, Monika Zienkiewiczb
a
PRAYON S.A., rue J. Wauters 144, Engis 4480, Belgium
b
GRUPA AZOTY, ul. Kuznicka 1, Police 72-010, Poland
Abstract
During the first edition of Symphos, Prayon presented a newcomer in the family of phosphoric acid processes: the Prayon Dihydrate Attack –
Hemi Filtration process (DA-HF).
In this process all the dihydrate slurry from attack tank is converted into hemihydrate before filtration. Process conditions have been found such
as to make possible phosphoric acid production directly from hemihydrate conversion tank. Based on pilot test results with more than 6 different
phosphate rocks, with this process P2O5 recovery gains of about 1,5% - 2% were expected with respect to a dihydrate process. Slurry filterability
is also improved by about 10% - 20% and P2O5 content of the produced acid is increased by more than 6% compared to a Dihydrate process.
Based on such promising results Grupa Azoty and Prayon Technologies decided to join their forces to run two industrial tests in Grupa Azoty
plants.
One of the Dihydrate production plant was converted into DA-HF process by simple modifications, including piping modifications, pump
addition, steam and sulphuric acid injection pipes installation. During the tests, Egyptian and Moroccan rocks were processed successfully. The
joint project team showed that the process is stable, it can be stopped and started easily. The efficiency obtained is about 97,5% with Morocco
rock and the P2O5 produced above 33 %. This corresponds to an impressive gain in efficiency of about 4% P2O5 compared to the dihydrate
process.
The results obtained being very promising, Grupa Azoty and Prayon Technologies have agreed to move forward studying the conversion of one
Grupa Azoty production plant into this new recently patented technology.
Grupa Azoty, in cooperation with Prayon, is preparing to change the technology of phosacid production from DH to DA-HF. As a next step in
this project feasibility study, including process and technical issues, is being prepared.
© 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
© 2014 The Authors. Published by Elsevier Ltd.
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Scientific Committee of SYMPHOS 2013.
Peer-review under responsibility of the Scientific Committee of SYMPHOS 2013
Keywords: Hemihydrate , dihydrate; phosphoric acid processes, Filtration ;
1. Introduction
For years, little has happened in the field of phosphoric acid production processes. The five process routes (DH, DH-HH, HH,
HRC and HDH) are well known and consolidated (ref 1 Becker). All the previously patented process technologies for these routes
are essentially all in the public domain.
Almost all technology providers have concentrated their efforts on optimization of the existing processes and on equipment
developments (ref 2 PSA, South Africa).
* Corresponding author. Tel.: +3242739735.
E-mail address:
[email protected]1877-7058 © 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Peer-review under responsibility of the Scientific Committee of SYMPHOS 2013
doi:10.1016/j.proeng.2014.09.036
182 Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187
Despite the widespread belief that the field is mature and has reached a stable state, during recent years PRAYON has been
investing in R&D works in order to improve the efficiency and the profitability of its processes. One of the most interesting
developments is the discovery of the DA-HF process route. It is a dihydrate – hemihydrate process with a single filtration step.
Pilot testing of this process has been described in several recent publications. (Ref 3. South Africa 2010, Ref 4 Symphos 2011).
In comparison with a dihydrate process, this process has several advantages among which we can quote, the higher plant
efficiency, the higher P2O5 content in the acid produced.
Based on these observations GRUPA AZOTY and PRAYON organised an industrial test in April 2012 that showed very
promising results. In order to complete the observations made during the first test a second test of two weeks was organised during
the first half of December 2012.
The aim of this test was to:
1. Verify and complete the results observed during the first test.
2. Operate the plant with a “difficult” rock (Egyptian rock).
3. Check the scaling speed of the filtration section.
4. Check the sturdiness of the process i.e. plant stop and start-up, phosphate changes, plant operation.
5. Define the equipment to be added or modified in order to convert one of the phosphoric acid lines into DA-HF process.
This paper describes the results obtained during the second industrial test.
1. Description of the tests
1.1. Plant modification
The industrial test was performed on one phosphoric acid line of Grupa Azoty located in Police, Poland. The line was adapted
in order to ensure the test with the minimum investment.
Description of plant before the test
Figure 1 : simplified scheme of the production line used for the test.
The line has a nominal capacity of 150 tpd P2O5. It is fitted with 4 reactors of about 225 m³ each.
The first reactor is fitted with 4 agitators. In this vessel, the sulphuric acid and the phosphate rock are added. The cooling of the
slurry is ensured by a slurry circulation through a flash cooler. From the compartment 4, the slurry is pumped to the filter and the
cake is washed. The cake is washed 3 times before its dry discharge. It is then transported by a series of conveyors on the gypsum
stack located at about 5 km from the plant. The filter has a total surface of 40 m². It has two vacuum circuits, one ensuring the
slurry filtration and the washes, the other dedicated to drain the cake.
Modifications performed
Figure 2 : simplified scheme of modifications.
Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187 183
For the test the following main modifications were performed.
x Installation of a line to pump the slurry from compartment 1 to the flash cooler.
x Installation of a steam line in compartment 4.
x Installation of a sulphuric acid line in compartment 4.
1.2. Main process parameters
The following process parameters were followed during the test:
x Residential time DH section: about 3h30 at 100% capacity.
x Solids content in DH section: about 25 to 30% w/w.
x Temperature in DH section: 74 to 76°C.
x SO3 DH section: 0.7 to 1 %.
x Residential time HH section at 100% capacity: About 1h20 minutes.
x Filter surface: 40 m².
x Temperature HH section: 97 to 99°C.
x SO3 HH section: 2.5 to 5%.
x % P2O5 in conversion section: 32 to 34% P2O5.
1.3. History of test: brief description
The second industrial test lasted for two weeks. After the two first days to check the plant in Dihydrate mode we started the
plant and operate it under DA-HF process continuously during 12 days.
During these twelve days we operated the plant during 160h processing about 1900 t of Egyptian rock and 850 t of Moroccan
rock.
During the DA-HF operation, the plant stopped and started about 25 times. Only 3 stops were due to mechanical failure the
other were organised to clean the filter and the conveyor hoppers not adapted to hemihydrate handling. After each stop plant start-
up was easy.
Special arrangements were organised to install a small plant laboratory to allow Prayon technician to perform filtration test, SO3
analyses and density measurements.
Operators followed a training course to understand the main difference between the first and the second test and to be aware of
the objective of this test.
We started the plant by partially emptying and conversion reactors. Then we started phosphate feeding using concentrated acid
(40% P2O5) as recycled acid. At the end of the filling the solids content was close to 30% and the P2O5 in the reaction liquid 36%.
Level in conversion was set-up at 1.5 m instead of 2 m for the next start-up operation step.
The sulphate content for dihydrate and conversion reactors were then adjusted and steam added in conversion compartment to
start the conversion. Once conversion started, conversion tank was fed with gypsum slurry during 30 min to increase its level
nominal level. After filling, filtration was started and plant capacity gradually increased.
2. Attack-filtration results
During the test we collected very valuable information on the performance of the test but also on the operation of the process
and the modifications to be performed to convert an existing plant into DA-HF technology.
The paragraphs hereunder present the most relevant findings related to:
x Quality of acid and calcium sulphate produced;
x Efficiency of the processes;
x Filterability;
x Plant operation
2.1. Analyses
2.1.1. Chemical analyses
The tables hereafter show the chemical analyses of phosphoric acid and calcium sulphate cakes obtained when the plant operates
in DA-HF mode.
184 Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187
Table 1 : Analyses of phosphate, acids and calcium sulphate when plant operated with Egyptian rock
Elements Unit Phosphate Rock Acid C1 Gypsum C1 Acid C2 Hemihydrate C2
Egypt (250°C basis) (250°C basis)
P2O5T % w/w 28,5 36.03 32.85 0,2
P2O5UN % w/w 1.16 0,19
P2O5 CO % w/w 0.78 0,29
Crystal water % w/w 19,60 6,22
Table 2 : Analyses of phosphate, acids and calcium sulphate when plant operated with Moroccan rock
Elements Unit Phosphate Rock Acid C1 Gypsum C1 Acid C2 Hemihydrate C2
(250°C basis) (250°C basis)
Morocco
P2O5T % w/w 30.7 37.4 33.8 0,2
P2O5UN % w/w 1.42 0,13
P2O5 CO % w/w 0.59 0,49
Crystal water % w/w 19,60 6,3
x Co-crystallised and Unreacted losses are strongly reduced during the conversion. This is a surprise in the case of the un-
reacted rock. Indeed, in pilot test we have never observed un-reacted losses reduction. This observation could be due to the
fact that sample was taken from the top of the reactor where phosphate rock from the phosphate feeder could accumulate.
x The acid dilution due to conversion is about 3.5 %. It is less than we would have expected. It is due to the lower steam
consumption.
2.2. Process results
2.2.1. Filtration test
During the industrial test in order to observe the slurry behaviour, several filtration tests were performed in the plant laboratory.
We had the possibility to perform 13 tests when the plant operated in DA-HF mode.
Comparing the values obtained during the first test on the filtration of the dihydrate slurry, one can see that the filterabilities
obtained are better in the case of the DA-HF. It is due to the better filterability of the cake due to the shape of HH crystal on one
hand but also to the fact that the water consumed for the cake wash is lower coming from the higher P2O5 content in the product
acid and to the lower crystal water content of hemihydrate crystals.
Table 3 : Average values of Filtration coefficients measured during the operation of the plant in DH and DA-HF modes.
DH operation DA-HF operation Egypt DA-HF operation
rock
first test Morocco rock
Conversion data (C2):
Slurry temperature °C 76 95 95
Filtrate Density 20°C 1,308 1.444 1.460
Filtration rate :
Atm. Pressure (local) (mmHG) 760 760 760
Filter Vacuum (mmHG) 500 500 500
Filtration cycle (S) 180 180 180
Filtration rate (TPD P2O5/m2) 3,8 4.2 5.4
Cake thickness (mm) 70 43 60
Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187 185
2.2.2. Efficiency
Reaction efficiency (including un-reacted and co-crystallised losses) was fairly stable during the test. When the plant was
stabilised on targeted sulphate value average efficiency was 98%. This has to be compared with the average 91-94% Dihydrate
industrial yield with the same rock.
Washing efficiency when filter operate under normal DA-HF condition was identical to the efficiency observed with a DH
process (above 99%).
2.3. Comments and pictures about the tests
This industrial trial was a needful complement to the first one performed in April 2012. It allows us to collect very useful
information.
First observation was the confirmation that a plant designed under dihydrate process can be converted to perform an industrial
test.
The main difficulties during the test were linked to the design of the gypsum discharge system not adapted for HH. In order to
operate despite this characteristic we had to clean manually gypsum chutes, wash conveyor with water, use hydration inhibitor and
stop the filter regularly to clean the hopper.
Once the process parameters were established and initial trouble shooting solved, plant operation was stable and easy. Plant
shut-down and start-up were easy. Operators operate alone the plant the last 2 days of the test.
We could easily produce 32 - 35% P2O5 phosphoric acid.
Compared to DH operation, absolute pressure in the flash chamber had to be kept at a lower pressure to ensure slurry cooling.
Vacuum pump capacity had to be increased.
Picture 1 : typical dihydrate crystal during normal operation.
2.3.1. Conversion
To perform a test, low specific power of conversion agitator can be accepted. For the test we used an agitator of low specific
power (about 0.3 KW/m³ of slurry). It did not impede the gypsum conversion into hemihydrate
The initial conversion of gypsum to hemihydrate is easy provided that the total acidity is high enough. Increasing the P2O5
content before start-up to at least 38 % really ease start-up operations.
We had no difficulty to convert the DH into HH as long as we kept a high enough total acidity and a temperature around 98°C.
HH crystals stayed stable even if the plant is stopped for more than 8 hours.
2.3.2. Filtration
During the majority of the test the filtration of the slurry was easy. Cake discharge was dry. HH crystals were formed of
aggregate easy to filter.
186 Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187
Picture 2 : typical HH crystals observed during the test.
Picture 3 : HH cake before discharge
3. Industrial plant modification based on the lesson learned
Based on the lessons learned from both industrial tests, we developed several recommendations to convert an existing plant into
a DA-HF process to plant operation and to installed equipment.
In term of operation we developed a series of recommendation to start-up, stop and operate the plant. For example we found
that using concentrated phosphoric acid to start-up the plant ease really the plant start-up.
To convert a dihydrate plant to DA-HF process following changes need to be considered:
x Installation of a conversion tank.
x Check and,if required ,replacement of flash cooler vacuum pump. Indeed, lower absolute pressure will have to be reached.
x Adaptation of filter to permit hemihydrate handling. It includes:
x New filter through;
x New cells;
x New central valve (including a pre-wash section);
x Doubling of filtration line and filter seal tanks (to ensure partial washing when plant is in operation)
x Filter hopper replacement by adapted one.
4. Conclusion
Very close collaboration between Grupa AZOTY production and development teams and Prayon engineer and doctors made
possible the further development of the new DA-HF process showing a very promising future.
It has been clearly demonstrated that the DA-HF process can be operated on an industrial basis. It is a stable process easy to
operate and to start. Compared to single stage process it offers a significantly higher efficiency (+4%), higher acid strength (+6%
Tibaut Theys et al. / Procedia Engineering 83 (2014) 181 – 187 187
P2O5) and some increase in capacity. The increased plant profitability combined with the ease of revamping make this process a
very suitable candidate for all existing phosphoric plants.
Bibliography
[1]. P.Becker “Phosphate and Phosphoric Acid: raw materials, technology, and economics of the wet process”, 1989 Marcel Dekker INC
[2]. Phosphoric acid processes: where do we go from here? P.A.Smith, IFA Technical Conference 2010, Sun City, South Africa
[3]. New developments in Dihydrate – Hemihydrate processes, A Hoxha, D. Fati , IFA Technical Conference 2010, Sun City, South Africa
[4]. New developments in PRAYON Dihydrate – Hemihydrate processes T. Theys, D. Fati, A Hoxha , Symphos 2011, Marrakech Morocco
