INFLUENCE OF THE ROCK IMPURITIES ON THE PHOSPHORIC ACID

PROCESS, PRODUCTS AND SOME DOWNSTREAM USES

by: Tibaut THEYS – Prayon Technologies S.A., Belgium

Summary
Production of phosphoric acid and phosphatic fertiliser is influenced by soluble impurities
contained in the phosphate rock. They will influence the process, the quality of the acid and the
gypsum produced as well as the downstream units using this phosphoric acid. This article,
through a review of existing publications, summarises their impact on the corrosion, filtration,
scaling, phosphoric acid, gypsum, and fertiliser qualities.

Introduction

Phosphate quality has always been a subject of concern and numerous articles have been
written treating the impact of some foreign ions on the production of phosphoric acid and
fertilisers.
Considering the production of phosphoric acid, all the elements that are not PO43- or Ca2+ are
considered as “impurities”. If not soluble, these impurities have a very limited impact on the
process; erosion of stainless steel equipment and an impact on the filtration of gypsum as these
are the only phenomena that can be detected.
Once the ions are in solution in the acid, other problems such scaling, corrosion, production of
sludge, increase of acid viscosity and process instability can occur. Such phenomena can be
produced by the presence of one single ion but also by the combination of several of them. For
example, if the conditions are favourable, a small increase of the potassium, aluminium or iron
content of an acid can produce a large quantity of sludge made of so-called X compound.
These effects are not always detrimental to the process. Silica and aluminium, for example, are
often added to the reactor to improve filterability. With such addition, filterability will not be
the only improved factor; corrosion speed will be reduced due to the reduction of free fluorine
in the reactor.
When the production of phosphoric acid was in its early days, rock quality was an issue in
order to find the raw materials giving the best performances. Nowadays, the task of the
production staff is to use the phosphate rock they have in their warehouse, to understand its
impact on the process and to choose the blend of rock or the appropriate additive that should
possibly be added to compensate for any deficiency.
The impact of impurities on the process will be considered through the following aspects:
process chosen to produce phosphoric acid;
corrosion;
filtration;
phosphoric acid quality;
gypsum quality;
scaling;

Presented at the IFA meeting of the Technical Committee on October 1, 2003 at the
Hilton Abu Dhabi
* Received: 05.08.2003
email address: [email protected]
 fertiliser production
At the end of the article, charts will summarise the different observations.

Process Choice

Choosing a phosphoric acid process has always been a difficult task. It is based on the overall
investment cost, the energy requirement, the calcium sulphate produced, the intended use of
the acid … and the raw materials. Impurity content of the phosphate rock can in some case
disqualify a process from the initial selection. It is the case, for example, of the lanthanide
content of the rock (La, Ce, Y) which, even in very limited quantities, slow down the conversion
of Hemihydrate to Dihydrate (Witkamp, 1989). Rocks containing this element thus disqualify
Hemihydrate – Dihydrate processes.
Phosphate containing a high Al content can also cause problems with Hemihydrate –
Dihydrate processes as it slows down the re-crystallisation of Hemihydrate to Dihydrate.
Indeed, aluminium fluoride (AlF52-) increases the HH solubility by up to 40% close to the DH
solubility (Martynowicz et al, 1996). To temper this problem silica is sometimes added to the
rock (Anonymous, 1993).
CO2 content can also be a limitation with a process having a vacuum flash chamber as the
reactor. Indeed, the high CO2 content in the rock produces a very large quantity of foam
inducing a slurry carry over that can hardly be controlled by the use of high quantity of
defoamer. Note that a limited amount of carbonate is often seen as a positive feature as it
“chemically” grinds the rock during the reaction increasing the speed of reaction and
preventing coating (Smith and Bourgot, 1991). A content of 3 to 5% of CO2 is often appreciated
in the rock even though it would result in higher sulphuric acid consumption.

Corrosion-Erosion
Impurities contained in the phosphate rock have positive or negative impact on the corrosion
of the stainless steel parts used in the process. Those increasing the corrosion or erosion are: F,
Cl, S or S2-, SO42- , insoluble SiO2 (erosion) and in some case ions (salts or complex) capable of
oxidation - reduction reactions as Fe3+.

Impurities increasing the speed of corrosion-erosion

In the reactor, the presence of chlorides can cause severe increase in corrosion rates, especially
in the case of the high strength-high temperature processes. The chloride being increased "pro
rata” to the P2O5, if this is linked with a higher operating temperature, the corrosion rate is
greatly increased. PRAYON's PH range of processes have different operating parameters from
the other High Strength processes, operating at a lower temperature in the second zone of the
Hemihydrate Attack section, and as such are much less sensitive to the chloride level in the
phosphate (Smith, 1991).
In the case of under vacuum concentration unit %P2O5 and Cl content are not anymore linked.
Independently of the Cl content in the weak acid the chlorine content in a 54% P2O5 acid
leaving the concentration is always between 150 and 300 ppm.
The corrosive effect of chloride is greatly reinforced by the sulphate and fluor content in the
acid. F- and SO4-- ions have a synergical effect on the corrosion and, for a given chlorine
content, greatly increase the speed of corrosion of stainless steel alloys (Becker, 1997 and
Becker, 1980)

2
Considering corrosion, three groups of phosphates, according to their chlorine content, can be
defined: low, <300 ppm; medium, 300 to 700 ppm and high, > 700ppm (Schorr, 1993).
Stainless steel alloy withstand corrosion once they are passivated. The usual explanation of
Passivity is that a thin protective film – either metal oxide or chemisorbed oxygen- forms on
the surface of the metal, preventing further contact with the electrolyte (i.e. the phosphoric
acid) (Anonymous, 1965). This layer can be destroyed in the presence of reducing substance.
Certain rocks contain pyrite which is converted to H2S, HS-, S2- during the reaction. Some
calcined rocks contain organics matter not fully oxidised during calcination. These reducing
substances impair the corrosion resistance of the passive alloys utilised in phosphoric acid
service. In some cases MnO is added to maintain the passive condition (Schorr, 1993). When
Simplot went to calcined rock,( Bierman and Long, 1981) they had serious corrosion problems
(due to the change of redox potential of the acid) which were initially resolved by the addition
of MnO2 and permanently by the aeration of the reactor generated by the foam breaker blades
on the agitators.
If the SiO2 source is from a solid insoluble in phosphoric acid, erosion can be observed. It is
mostly observed once the insoluble silica level is high (Halaseh and Naber, 1996).
F- is also a very corrosive anion. Its corrosiveness is considered similar to the chlorine. Often it
is “passivated” by reacting with silica or aluminium (either contain within the phosphate or
add as an additive) and others forming much less aggressive complex ions.

Impurities decreasing the speed of corrosion-erosion

All forms of SiO2 are not equally soluble in the phosphoric acid (and thus able to react with F):
it depends on the mineralogical characterisation of the ore. Diatomite and kaolin are very
soluble while olivine, or muscovite are almost insoluble in phosphoric acid (Smith, 1991).
Soluble SiO2 has a very positive impact on the corrosion as it complexes free fluorine. The
reaction products H2SiF6, a strong acid with medium corrosive capability (Smith, 1991;
Halaseh and Naber, 1996). Besides silica, other elements such as Al, Mg can also form
complexes with fluorine reducing its corrosive capacity. The fluorinated complexes produced
are (AlFx) where x = 1 to 6 and MgF2.
Schorr (1993) has demonstrated the impact of the free fluorine on corrosiveness. He defined a
fluorine complexing ratio : F / (SiO2 + Al2O3 + MgO) and showed that the higher this ratio, the
higher the corrosion speed.

Filtration

The main impurities having an impact of filtration are silica, aluminium and fluorine.
Depending upon its content and under which form it is found in the filtration section silica can
have a positive or negative impact on the filtration. If the silica source is made of quartz or
other insoluble salts in the phosphoric acid and if its particle size is quite small (less than 50
µm), it can impede the filtration requiring larger surface of filtration (Halaseh and Naber,
1996). If soluble, by reacting with the free fluorine it will impede the production of gypsum
needles and promote formation of gypsum crystals, which are easy to filter (ratio length on
width of the crystals close to 1, cluster).
Combined with F, Na or K, silica can form fluosilicate salts scaling on filter cloth, pans and
piping reducing the filtration rate. Chukrovite is a precipitate also commonly found that has
similar scaling properties. Process parameters, mostly temperature can emphasis these scaling.

3
If it is linked with fluorine, aluminium improves filtration. It seems to form AlF52- salts that are
incorporated in the calcium sulphate crystal. This incorporation retards the growth of the
crystal along the fast growing direction promoting the formation of crystal having a square or
cluster shape, easier to filter (Martynowicz et al, 1996).
High aluminium containing rocks offer the best filtration rate. This filtration effect however
depletes somewhere beyond 1.4% Al2O3 content in rock (Becker, 1997). Considering several
sedimentary phosphate, Becker (2001) shows that the increase of filtration rate due to the
presence of Al is about 1 t P2O5/m²/day.
Magnesium in phosphoric acid has a dual effect on the crystallisation, initially combining with
the fluorine it improves crystallisation but above a certain threshold its effect in increasing the
viscosity of the acid phase causes a rapid deterioration of the filtration rate. This value depends
on the presence of other impurities but is of the order of 2-3% (Smith, 1991). With Florida
rock, the presence of 1.2% MgO already decrease the filtration rate and the recovery. Once the
content reach 1.83%, the reduction is very important (Leyshon, 2000).
Halaseh (1997 and 1998) defines a Fluorine Complexing Ratio as:

%F
FCR = 19
% Al 2 O3 % SiO 2
+
17 15

He has tested several phosphates of different FCR ratios. He observed that for FCR values
between 1.378 and 0.42: as the ratio decreases filterability of the slurry increases, %P2O5 of the
acid increases and corrosion decreases. For ratio lower than 0.42 filterability decrease due to
the presence of un-dissolved silica salts and poor crystallisation.
Similar work was performed on Zin and Oron phosphates showing the positive impact of clay
addition (Schorr, 1996).
Kruger and Fowles (1998) showed that Na and K has an effect on the supersaturation curve of
calcium sulphate promoting the production of smaller crystals. On the other hand, they
observe that if aluminium initially promotes crystals growth and thus filtration rate, brings
also some disadvantages. At high aluminium to fluoride ratios, Al increases the acid viscosity
and density and therefore has a detrimental effect on filtration at those levels. Concerning Mg,
they observe that this element has little effect on the crystal morphology itself.
Organic matters present in the phosphate ore, those used for the beneficiation of the
phosphate, those added to the reaction section as defoamers or crystal habit modifier, can also
affect the gypsum crystallisation (Smith 1991, Theys and Roblin, 2002).
Rare earths in the slurry liquid phase above 0.18%, impair and also eliminate the advantageous
effect of aluminium fluoride complex on crystallisation of calcium sulphate and consequently,
deteriorate slurry filtration when using Kola rock (Halaseh and Naber, 1996).
In the case of igneous phosphate rock, potassium and iron seems to have a positive effect on
the filtration of gypsum within a narrow range of concentrations (Kruger et al, 2000). This
effect is not observed with Brazilian rock (Sinden, 2003).

4
Phosphoric Acid Quality

Once the soluble impurities contained in the phosphate are found in the acid they will have an
impact on the physico-chemical characteristics of the acid but also will influence the down
stream production.
Main soluble impurities as MgO, Al2O3, Na2O, CaO, K2O, Fe2O3, SO42-, SiF62- have an impact on
the density and the viscosity of the acid. This effect can be forecast from analysis of the acid
(Becker, 1997).
MgO is often found in the sedimentary rock as dolomite (Astley 2001). It increases the
viscosity of the acid produced. Acids with a higher MgO content have often a lower P2O5 in
order to ensure the filterability of the slurry. The amount of gypsum solid present in the slurry
is also lower to overcome this problem. For example if the high MgO Karatau phosphate is
used, to ensure the production of phosphoric acid, reactor must run at 20-23%P2O5 and at 90-
95°C in order to reduce the viscosity (Halaseh and Naber, 1996).
If the rock contains a high quantity of carbonate and some organic matters, the acid produced
can form foam (Halaseh and Naber, 1996). If these foams which appear in the flash cooling
section they can be carried over to the condenser causing a pollution of the cooling water. This
phenomenon can be reduced or avoided by the injection of a defoamer. Hemihydrate processes
are more susceptible to foaming as they have a more viscous acid and thus need a higher
defoamer addition rate.
If acid is being produced for animal feeds, such as dicalcium phosphate, low levels of fluorine
must be obtained by defluorination. It is usually done by an addition of Silica to the acid in
order to promote the stripping of SiF4. Nevertheless, if Al and Mg are present in the acid, they
form stable complexes with the fluorine, SiF4 formation will be reduced and defluorination will
be more difficult or even impossible. It explains why when feed grade acid is produced an
organic crystal habit modifier is preferred to clay.
Cadmium and other heavy metals can be an issue on the acid quality (Van Kauwenbergh 2002)
when downstream production is considered. The main impact is the level of heavy metals in
the fertiliser. Acid produced by the Hemihydrate route contains less Cd than by the Dihydrate
route. The Cd cocrystallise in the calcium sulphate. This uptake of Cd by co-precipitation is
even higher if anhydrite is considered.
Cd removal is possible by calcination of the phosphate, co-precipitation or ion exchange
extraction. Davister (1992) did an evaluation of these routes and their respective cost.
Depending upon the P2O5 content of the acid, several other salts beside calcium sulphate can
precipitate in the acid. In the case of 27-30% P2O5, major co-precipitation examples are:
fluosilicate of Na or K, fluo-aluminate (Ralstonite) [AlF6MgNa], Chukrovite
[(SO4Ca)x(SiF6Ca)y(AlF6CaX)z12H2O]. During / after concentration the major precipitate are:
X-compound [(Fe,Al)KH14(PO4)8 4H20]; Mg,Al (SO4)2F nH2O, fluosilicate of Mg, Ca, Al and
others (Becker, 1997).
Where acid must be shipped for international consumption, the acid must normally meet MGA
(Merchant Grade Acid) specifications. The suspended solids content should be less than 1% on
arrival. Florida acid, with its high Al2O3 content, suffers post-precipitation during several
weeks after concentration. A low Al2O3 content in the acid simplifies the clarification process
and reduces the inventory, as shorter retention times are required for ageing (Smith, 1991).
Some igneous rocks suffer of the same post-precipitation problems. In the case of the
extpansion of Foskor plant in Richards Bay, large desupersaturation and settling tanks were
installed for the merchant grade acid (Kirsten et al, 2002).

5
Acid can have a dark colour due to organic matters. Depending on the use of the acid, this
colour has to be removed in some case. This removal can be done by the addition ofc to adsorb
the carbonaceous impurities followed by filtration (Schorr, 1993). Another option is to use
active carbon or oxidising agents such as nitric acid, H2O2...
H2S can cause health hazard if present in the acid. PCS Phos acid units operating on the
calcined N.Carolina rock monitor the presence of H2S on the filter (Sinden, 2003).

Calcium Sulphate

Phosphogypsum can in some cases be used in the construction industry. The four main uses
are:
- as a retarder in the production of cement
- for the production of wallboard/plasterboard
- for the production of gypsum blocks
- for the production of Stucco plaster

Depending each of the usage gypsum quality requirement varies. The main impurities that are
considered are: P2O5, F, Cd (ppm), Na2O, pH, organic matters, radioactive elements (Smith
and Theys, 2000).
Even if the produced calcium sulphate is not used for other applications, its impurities content
can have an impact on the Phosphoric acid production. Al2O3 and Fe2O3 form salts that contain
P2O5 (Halaseh and Naber, 1996). They thus increase the P2O5 co-crystallised loss. In the case of
aluminium this effect can mainly be observed with Hemihydrate process.
For Becker (1997), the major ions entering the calcium sulphate by co-precipitation are: HPO4--
, AlF52-, Cd++, Rare earths and Y+++. In the case of Cd, the portion of this ion that precipitates is
a function of the calcium sulphate form. In the case of Dihydrate, about 20% of the Cd of the
phosphate co-precipitates. This value increases to about 50% for Hemihydrate and 95% for
Anhydrite. The Prayon phosphoric acid plant of Engis is a Central-Prayon Process (Dihydrate
– Hemihydrate process) using mainly igneous rock. They observe that about 85% of the rock
Cd is contained in the Hemihydrate leaving the filter.
In the case of Dihydrate processes, aluminium has a stabilising effect on the dihydrate phase.
High Al containing rock (1-1.5% Al2O3) can still operate within Dihydrate precipitation phase at
temperature over 90°C and with a phosphoric acid concentration around 29% P2O5 (Becker,
1997).
Considering the radioactive elements, about 80% of the Ra-226 follows the phosphogypsum,
while about 86% of the U-238 and about 70% of the Th-232 are found in the phosphoric acid
(figures vary with the process). Due to the chemical separation the radioactivity of phosphoric
acid is significantly lower than that of the phosphogypsum (Anonymous, 2000). Radioactivity
level of gypsum can be an issue especially if the gypsum is used undiluted for down stream
application according to the radioactivity level of its phosphate source.

6
Scaling

Silica in the acid is considered as a positive feature as it combines with fluorine reducing the
corrosion potential of the acid, improving the slurry filterability and enhancing defluorination.
Nevertheless, if in too high content (or the amount of fluorine too low) it can cause scaling
problems in degassing ducts or cooling water systems (Anonymous, 2001). Indeed, silica reacts
with fluorine to form H2SiF6 that can evolve from the acid in the form of SiF4. Once this
product enters into and contacts with water it is solubilised and forms H2SiF6 and SiO2
precipitates. In degassing ducts these silica solids can form very hard scaling especially if the
condensation occur in a place where the temperature is above 60°C. Hemihydrate plants are
especially susceptible to this phenomenon.
When this absorption is done by the condenser cooling water, depending on the quality of
water, several types of scaling can be observed. If the water contains a high amount of Na or K
(like in the case of sea water) fluosilicate scale will be form. They are essentially formed at the
inlet of the condenser where the flow of water is high enough to bring the sodium and
potassium ions and small enough to be under the solubility constant of fluosilicate salts. If
process water is used silicate salts (CaF2 or fluosilicate) can be formed that can accumulate in
the cooling towers. This point emphasises the interest in using Fluorine recovery tower in
concentrations units to avoid the fluosilicate precipitation solids in the cooling system.
Some plants that use pond water containing Na or K to stack the gypsum face fluosilicate
scaling in their return water piping system. These scaling are remove by swapping the slurry
pipe and the return water pipe on a regular basis, the erosive effect of gypsum allows the
removal of the scaling.
Fluosilicate salts of sodium and potassium can also lead to the formation of scaling in the
filtration circuit. This is mainly due to the sudden cooling of the slurry during the filtration
(Halaseh and Naber, 1996).
To slow down this problem table filters are fitted with hot and high pressure cloth wash and,
sometime, a direct injection of steam on the filter cloth. In Prayon filter, the scaling are
reduced by a fast separation of the gas and the liquids and the possibility to inject steam in the
central valve to maintain the temperature in the filtration circuit. Moreover, the Prayon
“Digestion” section design ensures that the slurry fed to the filter is fully de-supersaturated
(Smith, 1991).
If the soluble SrO content exceeds 0.5% it can cause serious problems of scaling in
concentration heat exchanger section (Halaseh and Naber, 1996). These scalings can hardly be
removed by washing and needs mechanical cleaning to be removed. Prayon’s plant in Engis
had been confronted with that problem in the past. It was solved by using stainless steel pipe
submerged in an acid bath, that scale from the outside of the tube and thus could easily be
washed with high pressure water.
All the Brazilian igneous rocks contain significant levels of SrO, 0.5-1.1% but it is present in the
form of the mineral celestite (the natural SrSO4). As this form is a highly insoluble mineral
they do not face scaling problems.
One of the most surprising scales is chukrovite: (SO4Ca)x(SiF6Ca)y(AlF6CaX)z12H2O. These
crystal have the shape of a small double size pyramids (two pyramids on an common basis),
which precipitates from filter grade acid while cooling but essentially clogging filter cloths
(Becker, 1997).
Another crystal having a similar form is a solid having the composition: Ca3Al3(RE)SO4F13.
Where RE can be formed from rare earths or Yttrium. In some case Al+++ can be replaced by
other cations as Ca2+ or Si4+ and F- by OH- (Robinson, 1978).

7
Fertiliser Production

Depending upon the type of fertiliser produced, impurities can directly influence the quality of
the product or the production itself. Hallsworth and Enriquez (1980) show the impact of the
main impurities on the DAP quality.
MgO, Fe2O3, and Al2O3, also called MER, affect (Palm 1986, Astley, 2001) the DAP grade by
taking the place of H+ ions thus reducing the room available for nitrogen or phosphate (if they
are linked with other anions than PO43-). Moreover, MgO by its impact on the viscosity can
increase the scaling speed of pipe reactor.
Although both R2O3 and MER are used frequently they have little scientific value as they do not
fully express the different effects of Al, Fe and MgO.
MER (minor element ratio) is defined by the formula :
Iron + Aluminium + Magnesium

P2O5 in phosphate.
The higher the ratio, the higher the risk problems during the production of DAP. Astley (2001)
considers that this ratio should be corrected depending upon the type of acid used. If 54%P2O5
acid with low solids content is used to produce DAP, part of the iron is removed with the
sludge, then the formula should be corrected.
The new formula becomes:
0.5 Iron+ Aluminium+ 1.5 Magnesium

P2O5
The higher ratio on the MgO showing its higher impact on the production of DAP is due to its
total solubility and its low tendency to produce post-precipitate.
Iron produces phosphate salts that are citrate insoluble during the production of DAP (Lloyd,
1988). The iron salts can have the form of the X compound or a pseudomorphs of
FeNH4(HPO4)2. Another compound that could be formed due to presence of impurities is
MgAl(NH4)2H(PO4)2F2.
The iron phosphates are only citrate insoluble if they are kept at low Mole Ratios for significant
time 10-20minutes at elevated temperatures. It has been possible to produce MAP/DAP and
ASP with very low CI's using very contaminate acids (Sinden, 2003)
MAP, due to its lower ammoniation requirement, is less influenced by the presence of
impurities in the acid. Al and Fe are considered as positive impurities in this production as it
improve the mechanical resistance of granules (Brownlie, 1978). Sinden (2003), does not
consider that iron has this positive effect. For him aluminium (with at least 0.2%) and MgO are
the impurities improving the mechanical resistance.
Impurities content in the phosphoric acid used for the production of TSP can be an issue as
they take the place of protons and reduce the water-soluble content of the phosphoric acid.
Indeed, this lack of H+ reduces the ability of the phosphoric acid to react with the phosphate
rock (Sinden 1993). In some cases this problem can be overcome by adding some Sulphuric
acid.
Phosphoric acid analysis is thus essential for the production of TSP. Sinden (1993) considers
that a good TSP can be done with a “poor” phosphate provided that the purity of the acid is
good enough. Beside iron and aluminium, MgO is also considered as a major impurity. These
elements have an impact on the physical properties of the TSP.

8
A less known impurity known to have a negative impact on the production of TSP is Titanium.
Some Brazilian rocks contain a high level of this element. It reacts with the phosphate to form
a calcium-titanium phosphate salt lowering the water soluble P2O5 content of the TSP (Sinden,
1993).
Cd and other heavy metals are becoming a quality concern in the fertiliser production. SSP
only contains the Cd inherited from the phosphate, in DAP the Cd is included in the acid and in
some case the sludge (containing a high content of Cd due to coprecipitation), MAP as it is
often produced from sludge contains also a relatively high content of heavy metal, even higher
than that in the DAP produced from the same acid. TSP has the Cd content of the acid and of
the phosphate (Van Kauwenbergh, 1992).
Legal limitations on Cd in fertiliser start to appear. These limitations vary wildly from one
country to the other and are subject to regular changes. European Community has not yet
define a legal limit. Belgium and Luxembourg voluntarily limit the Cd content in fertiliser to 90
mg Cd/kg P2O5. In California (Anonymous, 2001), the limit will be reduced to 400 mg / kg
P2O5 on Jan.1, 2004 ! Indeed, a recent report (Anonymous, 1999) on fertiliser risk assessment
shows that Cd is not an health issue in fertiliser.
Besides Cd, almost all the impurities contained in the fertiliser have greater or less impact on
the health and / or the environement. Dylevskaya (2002) as well as Kiiski and Milbone (2003)
lists the main one as well as their potential negative impacts.
Depending upon the content of radioactive elements in the fertiliser, radioactive element as
Rd, U3O8, K40, Th could be regulated in the future (van der Westhuizen 2002). In 2000, under
the EU legislation, none of the phosphoric acid or phosphate in the world will see their
commercialisation restricted by their radioactivity (Anonymous, 2000).
Organic matters contain in the acid or the phosphate can cause odour problems during the
production of fertilisers. These odours are mainly due to the Mercaptan (MSH where M is an
organic molecule) evolving. They can be strongly reduced by an addition of an oxidising (as
NaClO) in the scrubbing system.

Conclusion

Considering their impact on the process, the product qualities, their interaction between them,
impurities contained in the phosphate can be considered either as positive or negative factors.
It is thus always a difficult task to evaluate a phosphate rock and to define its implication on
the process. One can then understand why phosphoric acid and fertilizer productions have
always been considered more as an art than a science.

Acknowledgement

I would like to thank all the gentlemen who helped me to write this article. I am very grateful to
Prayon R&D, production and engineering teams for their help. I would also like to warmly
thank Messrs. Karim Halaseh of JPMC, Paul A Smith of P Smith & Associates, John Sinden of
JSA ltda, Rob Fowles and Gert van der Linde of Foskor. Without their contribution, this article
would contain much less information. Finally, I would thank Arlette Coudron of PRAYON
Technologies for her help in “exhuming” data from our files and her patience reading my
drafts.

9
Overview charts
Impurities Process choice Corrosion Filtration Acid quality Gypsum quality Scaling Fertiliser
production
CO2 Produces
slurry carry
over if
phosphate
reacts with
sulphuric acid
in a vacuum
chamber.
Lanthanide Impedes the Decreases Rehydration of Causes scaling
conversion of filterability. Hemihydrate to
Hemihydrate Dihydrate is
to dihydrate. longer.
MgO Reduces corrosion Low content: Increases viscosity and Impact on DAP
(complex with F) improves filterability density of the acid. and TSP
by complexing with Keeps F in the acid. capacity and
F. Can cause post quality.
High content precipitation.
decreases filterability
(viscosity increase).
Cd Anhydrite or Quality issue in some Quality issue in Quality issue in
Hemihydrate countries some countries if some countries.
crystallisation gypsum is sold.
to remove Cd
from acid.
Al2O3 Impedes Reduces corrosion Low content: Increases viscosity and Stabilises Causes scaling Content should
Hemihydrate (complex with F). improves filterability. density of the acid. Dihydrate. (Chukrovite and be limited if
to Dihydrate Very high content: Retains F in the acid. Form Phosphate others). DAP and TSPare
conversion. decreases filterability Can cause post salt that co- produced.
(viscosity increases). precipitation precipitate in the Positive impact
calcium on MAP
sulphate. production.

10
Impurities Process choice Corrosion Filtration Acid quality Gypsum quality Scaling Fertiliser
production
Fe2O3 Neutral or Can improve filterability Increases viscosity Content should
negative and density of the be limited if
(depending acid. DAP or TSP is
oxydation stage). Can cause post produced.
precipitation. Positive impact
on MAP
production.
Na2O Fluosilicate scaling Increases viscosity Quality issue if Causes
reduces filterability and and density of the gypsum is sold fluosilicate
promote small crystal acid. for plaster. scaling.
(negative impac.t on Cause post
filtration) precipitation.
K2O Can improve filterability Increases viscosity Causes
but considered having a and density of the fluosilicate
negative impact due to acid scaling.
Fluosilicate scaling and Can cause post
promotion of small precipitation
crystals
SrO Impede Hard scaling on
conversion of heating
HH to DH elements.
F Increases Free F decreases H2SiF6 increases Quality issue if Causes
corrosion speed. filterability because viscosity and density gypsum is sold fluosilicate and
elongated gypsum of the acid. chucrovite
crystal. Can cause post scaling.
precipitation.
Cl Increases
corrosion speed.
SO42- Increases Increases viscosity Form calcium
corrosion speed – and density of the sulphate and
synergy with Cl. acid Chucrovite
scalings.
S2- Increases Health hazard (H2S)
corrosion speed.

11
Impurities Process Corrosion Filtration Acid quality Gypsum Scaling Fertiliser
choice quality production
SiO2 If insoluble: If insoluble: Cause
increases erosion decreases fluosilicate
speed. filterability. scaling.
If soluble: If soluble: improves
reduces corrosion filterability.
(complex with F).
Organic Depending their Produce foam. Quality issue if Odour
matters composition, Colour the acid. gypsum is sold.
filterability can be
increased or
decreased.
Radioactive Potential Potential Potential Potential
elements environmental and environmental environmental environmental
health problems. and health and health and health
problems. problems. problems.
Quality issue if
gypsum is sold.
TiO2 Impact on DAP
quality.

12
Bibliography

ANONYMOUS, (1965) RESISTANCE TO CORROSION, Inco Alloys International Inc.

ANONYMOUS, (1993) CHINESE ROCK IN HYDRO PHOSACID AND TSP PROCESSES, Phosphorus &
Potassium No. 185, May-June 1993
ANONYMOUS, (1999) ESTIMATING RISK FROM CONTAMINANTS CONTAINED IN
AGRICULTURAL FERTILIZERS DRAFT REPORT, Prepared by the Office of Solids Waste U.S.
Environmental Protection Agency and Centre for Environmental analysis Research Triangle Institute,
August 1999.
ANONYMOUS, (2001) HOW DOES ROCK PHOSPHATE QUALITY AFFECT OUR ACID ?, Indian Ocean
News, June 2001
ANONYMOUS B, (2001), CALIFORNIA TO PHASE IN PHOSPHATE METAL LIMITS, Green Markets,
Volume 25, Number 36, September 3 2001.
ASTLEY, V., STANA, R., IMPACT OF PHOSPHATE ROCK IMPURITIES ON THE MANUFACTURE OF
PHOSACID AND GRANULAR PHOSPHATES, IFA Technical Committee Meeting,, Thessaloniki, Greece,
Octobre 2001.
BECKER, P, (2001) COMMERCIAL PHOSPHATE ROCK FROM SYRIA, Fertilizer International N° 383,
July/August 2001.
BECKER, P., (1997) PHOSPHATE RAW MATERIALS, THEIR IMPURITIES AND RELATED EFFECTS
ON COMMERCIAL PRODUCT QUALITIES, AFA Tenth Annual Technical Conference, Amman, May
1997
BECKER, P., DUTHOIT, M., GAURON, M., DE VILLELE, H., (1980) FACTEURS QUI REGISSENT LES
VITESSES DE CORROSION DES ACIERS INOXYDABLES AU COURS DE LA PRODUCTION D’ACIDE
PHOSPHORIQUE PAR VOIE HUMIDE, ISMA Conference, 1980
BIERMAN, L.W., LONG, G.L. (1981) ACUTE CORROSION IN A PHOSPHORIC ACID PLANT
PROCESSING CALCINED ORE, Phosphorus and potassium N°.113, May/June 1981.
BOURGOT, A., SMITH, P.A., (1991) BASIC PHILOSOPHY OF PHOSPHORIC ACID PRODUCTION,
PRAYON TECHNICAL SYMPOSIUM ON PHOSPHORIC ACID TECHNOLOGY – May 1991
BROWNLIE, J.A., DAVIDSON, E., DICK, T.R., MARTIN, I.S.E., (1978) EFFET DES IMPURETES SUR
LES QUALITES DU MAP PRODUIT, ISMA Conference 1978.
DAVISTER,A., (1992) INVENTAIRES DES ETUDES ET RECHERCHES SUR LES PROCEDES
D’ELIMINATION DU CADMIUM DANS L’ACIDE PHOSPHORIQUE, Report for the European
Communities and the Institut Mondial du Phosphate (IMPHOS), 1992
DYLEVSKAIA, N., (2001) APPROACHES TO LIMITING THE CONTENT OF ENVIRONMENTALLY
HARMFULL IMPURITIES IN PHOSPHATE FERTILIZERS, IFA Technical Conference, Chenai, India,
September 2002.
HALASEH, K., NABER, G., COMPARATIVE EVALUATION OF JORDANIAN PHOSPHATE AS RAW
MATERIAL FOR WET PROCESS PHOSPHORIC ACID PRODUCTION, JORDAN PHOSPHATE MINES
CO, Fifth International Fertilizer Seminar, 14-17 April 1996 Amman-Jordan
HALASEH, K., (1997) WPA PRODUCTION USING JORDANIAN PHOSPHATE ROCKS, The Arab
Fertilizer Association – AFA, Tenth Annual Technical Conference Phosphate rock, Phosphoric Acid and
phosphatic, Potassic & Compound Fertilizers Technology , Amman – Jordan, May 8-5 1997
HALASEH, K., DUKHGAN, H., (1998) OPERATING EXPERIENCE OF PROCESSING LOW GRADE
JORDANIAN PHOSPHATES IN THE MANUFACTURE OF WPA PRODUCTION, IFA Technical
Conference, Marakesh, Septembre 1998.
HALLSWORTH, J.A., ENRIQUEZ, J.M., (1980), INFLUENCE DES IMPURETES DU PHOSPHATE SUR
LE TITRE DE L’ENGRAIS FINAL, ISMA conference, 1980
KARHUNEN, J., VERMEULEN, S., (2000) NATURAL RADIOACTIVITY OF PHOSPHATES AND
PHOSPHOGYPSUM, Fertilizer International No. 378 September/October 2000

13
KIISKI, H., MILBORNE, R.J., (2003) PRODUCT STEWARDSHIP APPLIED TO FERTILISERS, The
International Fertilizer Society, Proceedings 508, April 2003.
KIRSTEN, K., TYTGAT, J., THEYS, T., EXPANSION OF THE RICHARDS BAY DIHYDRATE
PHOSPHORIC ACID PRODUCTION COMPLEX, IFA Technical Conference, Chennai, India, September
2002.
KRUGER, A., FOWLES, R., THE EFFECT OF EXTRANEOUS SOLUBLE IONS IN IGNEOUS ROCK
PHOSPHATE ON CRYSTALLOGRAPHY OF GYPSUM DIHYDRATE AND THUS PHOSPHORIC ACID
PRODUCTION, IFA Technical Conference, Marakesh, September 1998.
KRUGGER, A., FOCKE, M.W., KWELA, Z., FOWLES, R., (2000) EFFECT OF IONIC IMPURITIES ON
THE CRYSTALLIZATION OF GYMPSUM IN WET-PROCESS PHOSPHORIC ACID, American Chemical
Society.
LEYSHON, D., (2000) KEEPING ABREAST OF THE MAGNESIUM PROBLEM, Fertilizer International,
N°378, September/October 2000.
LLOYD, G. M., ACHORN, F.P., SCHEIB, R.M., (1988)DIAMMONIUM PHOSPHATE QUALITY,
American Institute of Chemical Engineers Convention Clearwater, Florida, May 1988
MARZO, L.M., LOPEZ-NINO, J.L. POWERED OR GRANULAR DAP : THE USE OF FLORIDA ACID IN
A PIPE REACTOR, American Institute of Chemical Engineers Convention Clearwater, Florida, May 1986.
PALM, G., (1986) DAP QUALITY SEMINAR, Florida Institute of Phosphate Research, pages 8-11, August
1986
ROBINSON, N., (1978) EXPERIENCE DE FISONS SUR L’INFLUENCE DES IMPURETES DU
PHOSPHATE BRUT SUR LES PERFORMANCES DES UNITES D’ACIDE PHOSPHORIQUE, ISMA
Conference 1978.
SCHORR, M., (1993) MINERAL MODIFIERS IMPROVE WPA PRODUCTION, Phosphorus & Potassium
No. 187 – September-October 1993
SCHORR, M., LIN, I.J., WET PROCESS PHOSPHORIC ACID PRODUCTION PROBLEMS AND
SOLUTIONS, Industrial Minerals, April 1997.
SCHORR, M., LIN, J.,L., (1996) MINERAL MODIFIERS IMPROVE WPA PRODUCTION, Phosphorus &
Potassium No. 202, March-April 1996
SINDEN, J., (1993) EFFECT OF ROCK QUALITY ON TSP AND SSP PRODUCTION, Phosphorus &
Potassium No. 186 – July-August 1993
SINDEN, J., (2003), Personal communication.
SMITH, P.A., (1991) WHAT IS A HIGH GRADE PHOSPHATE ? ON ARE IMPURITIES A BONUS OR A
PENALTY ?, 3rd International Fertiliser Seminar, Amman, 1991.
SMITH, P.A., THEYS, T., (2000), GYPSUM, A PROFITABLE, SALEABLE PRODUCT, IFA technical
conference, New Orleans, September 2000.
VAN DER WESTHUIZEN, A.J., AN OVERVIEW OF RADIATION LEGISLATION THAT IMPACTS ON
THE PHOSPHATE AND FERTILIZER INDUSTRY, IFA Technical Conference, Chennai, India,
September 2002.
VAN KAUWENBERGH, S.J., (2002) CADMIUM CONTENT OF PHOSPHATE ROCKS AND
FERTILIZERS, IFA Technical Conference, Chennai, India, September 2002.
WITKAMP G.J. (1989) CRYSTALLIZATION OF CALCIUM SULFATE AND UPTAKE OF IMPURITIES,
Kanters bv, Alblasserdam, PhD Thesis, Delft University of Technology.

14