1. Selecting phosphate rock.
2. Selecting source of sulfuric acid.
3. Receiving and storing raw materials.
4. Grinding and otherwise preparing the rock.
5. Reacting the phosphate rock and sulfuric acid.
6. Filtering to separate phosphoric acid from gypsum.
7. Concentrating and clarifying the phosphoric acid.
8. Sludge treatment
9. Dihydrate Process Requirements

The description of phosphoric acid production will be divided into the steps in the contents list, followed by a summary of the requirements of the process.

Selection of phosphate rock

Many phosphoric acid plants are built in countries where phosphate rock must be imported. The plant is often designed on the basis of a single phosphate rock; however, it is often prudent to build into the plant sufficient flexibility to permit the use of rocks from different sources. This versatility will enable the producer to take advantage of competitive situations and to avoid disruption of supply when the intended source is inadequate or interrupted by hostilities, disasters, or other circumstances.

Many plants find it advantageous to use a blend of rocks from different sources. The extra expense of making the plant more versatile usually can be repaid many times over by savings resulting from freedom of choice in the world market. Some examples of steps to increase the versatility of the plant are:

1. Extra grinding capacity for harder rock.
2. Extra filtration capacity to provide for rocks that cause less rapid filtration or have lower P₂O₅ content.
3. Slurry handling systems that will cope with acid-insoluble impurities in the rock.
4. More corrosion-resistant construction for rocks that have corrosive impurities.
5. Extra sludge treatment capacity for rock with a higher Minor Element Ratio MER (Al₂O₃+Fe₂O₃+MgO)/P₂O₅.

When the plant is built at or near the mine, there is still the likelihood that the rock composition will vary. In addition, there are other questions to be considered relating to the economic optimum balance between cost of extra beneficiation as compared with the cost of using lower grade rock [8].

The selection of phosphate rock source is sometimes viewed as a simple matter of obtaining a given amount of P₂O₅ in the rock delivered to the plant at the lowest price. However phosphate rock is a complex raw material that affects plant operation in numerous ways, some of which may be unpredictable. Therefore, a thorough evaluation of all quality factors should be made before selecting a phosphate rock or changing from one source to another [9].

A complete chemical and mineralogical analysis of a phosphate rock is helpful in evaluating its usefulness for making phosphoric acid. However, this information is not sufficient in itself; trial runs in a plant or pilot plant are needed for a reliable evaluation unless the rock is one that has been used extensively in other similar plants with known results [10].

The following quality factors may provide a general guide for selecting phosphate rock for phosphoric acid production. The economic effect of many of the factors can be evaluated quantitatively to arrive at a comparative value of alternative sources of phosphate rock.

Starting with a standard grade of rock, the more common quality factors for wet-process phosphoric acid production and their effect are:

1. Lower grade (% P₂O₅) means that more tonnage must be purchased, transported, handled, and (usually) ground.

 2. An increase in the CaO:P₂O₅ weight ratio increases the sulfuric acid requirement. (Any CaO present as CaSO₄ should be excluded in calculating this ratio.)

3. Magnesium oxide increases the acid viscosity and can negatively affect the gypsum filtration rate. Magnesium oxide also forms precipitates with fluorine in the reactor, which may blind the filter cloth; therefore and then high MgO content is considered undesirable. When phosphoric acid is used to produce ammonium phosphates or polyphosphates, water-insoluble (but citrate-soluble) magnesium ammonium phosphate compounds may be formed. These compounds form troublesome impurities in liquid fertilisers.

4. Increases in the Fe₂O₃ + Al₂O₃ content above 2%-3% decrease the plant capacity, often decrease the P₂O₅ recovery, and cause post-precipitation problems (sludge). However, up to about 4% may be tolerable i.e. a MER [Minor Element Ratio (Fe₂O₃ + Al₂O₃ + MgO)/P₂O₅] of < 0.08. From the other side Al₂O₃ and MgO are a bonus in some way because they reduce the corrosivity of the acid by forming complex ions with free fluorine ions.

5. It is desirable to have enough reactive silica (SiO₂) to form SiF₄ and/or fluosilicates to avoid formation of free hydrogen fluoride (HF), which is very corrosive. Excessive silica or other acid-insoluble impurities may cause erosion of equipment and possible accumulations in digestion vessels, depending on particle size, character, and plant design. In addition a high percentage of silica in the rock would increase the required filter area.

6. Chlorine contents above about 0.03% cause increased corrosion of stainless steel especially in the case of the high-strength, high-temperature processes. More expensive alloys may tolerate Cl contents of 0.10% or perhaps higher.

7. High organic matter may increase foaming problems (by stabilising the foam), increase viscosity, and hinder filtration. The effect depends on both the character and quantity of organic matter. Some rocks must be calcined to remove organic matter to make them usable.

8. Carbonate and resulting carbon dioxide (CO₂) contributes to foaming and increases consumption of antifoam reagents. Presence of carbonate increases the rock reactivity (faster dissolution) and may improve the P₂O₅ recovery (less unreacted phosphate rock).

9. All commercial phosphate rocks contain fluorine (F); no special effect has been noted due to variations in fluorine content within the range of experience. Effects of fluorine on scaling, corrosion, and post-precipitation are related to other elements that combine with fluorine, including Na, K, Al, Mg, and Si.

10. Upon acidulation, some rocks that contain sulfides release hydrogen sulfide (H₂S), a toxic gas that must be neutralised in the gas scrubber. These sulfides tend to increase corrosion.

 11. Strontium and lanthanides (rare earths), found in igneous rocks, inhibit rehydration of the hemihydrate to gypsum, which can cause problems in certain phosphoric acid processes. Moreover strontium causes problems in the concentration sections because strontium sulfate has a minimum solubility in 40% P₂O₅ acid. An extremely thin film of SrSO₄ causes a very pronounced reduction in the capacity (thermal transfer loss in the heat exchanger) of the concentration unit.

12. High contents of toxic impurities in phosphate rock used (for example, cadmium and arsenic compounds) may render the resulting phosphoric acid unsuitable for fertiliser production.

13. Hardness is a factor since harder rocks require more grinding capacity.

14. The particle size of the rock as-received affects the amount of crushing and grinding required. Very fine particle size may lead to dust losses in handling.

15. Low reactivity of the rock may require finer grinding.

16. Filterability of the rock-acid slurry is one of the most important characteristics of a phosphate rock for use in phosphoric acid production. Factors influencing filterability are complex and not completely understood. However, if a plant is to be designed to use a specific rock, an acceptable filtration rate can usually be attained through experimental means by adjusting operating conditions, addition of crystal modifiers, or pre-treatment of the rock.

Table 1 gives the range of composition and median values of a group of 15 phosphate rocks from commercial sources. Although the group is representative, compositions outside this range have also been used. Assuming 94% overall P₂O₅ recovery, the amount of phosphate rock required per tonne of P₂O₅ recovered as phosphoric acid is given in Table 2.

Calculation of sulfuric acid requirement

Although the sulfuric acid requirement for the production of phosphoric acid from any given rock is best obtained experimentally, it is sometimes necessary to calculate it from the chemical analysis of the rock. For a first approximation the sulfuric acid requirement may be equated to that required to combine with the calcium in the rock to form calcium sulfate. This calculated value is often close enough for planning purposes. The requirement per tonne of P₂O₅ recovered should be adjusted according to the expected recovery. The overall recovery for a single DH process is seldom more than 94% if mechanical and sludge losses are included.

If a complete analysis of the rock is available, a more exact calculation may be made. The method is illustrated in Table 3 and is explained below:

1. Assuming 94% overall P₂O₅ recovery, 1,064 kg of rock P₂O₅ is required per ton of P₂O₅ recovered.

2. If the rock contains 33% P₂O₅, 3,224 kg of rock is required.

3. The CaO content of the rock is calculated. If the rock contains any other cation that forms an insoluble sulfate (such as barium), its CaO equivalent should be added.

4. The CaO equivalent of the SO₃ content (not total S) should be considered.

5. Typical filter cake contains about 3.3% of the input P₂O₅ in insoluble forms of which 1% may be unreacted rock and 2.3% CaHPO₄ co-crystallised with gypsum. The aggregate weight ratio of CaO combined with P₂O₅ is about 1.0.

6. The empirical assumption is that 15% of the fluorine combines with CaO to form CaF₂. Actual reactions are much more complex; Ca₄SO₄SiF₆AlF₆ (OH) •12H2O is an example of a complex insoluble compound found in filter cake.

7. Items 4, 5, and 6· are totalled.

8. Item 7 is subtracted from 3 to give the net amount of CaO for reaction with H₂SO₄.

9. The H₂SO₄ equivalent of the CaO is calculated.

10. The amount of excess H₂SO₄ is calculated by assuming that the 30% P₂O₅ acid contains 1.5% free H₂SO₄.

11. Total H₂SO₄ requirement is item 9 plus item 10.

12. For comparison, step 12 of Table 3 shows the H₂SO₄ requirement based simply on total CaO.

In Table 4 the sulfuric acid requirement for median-grade rock is taken from Table 3 (2.78 tonnes of H₂SO₄ per tonne of P₂O₅), and requirements for rocks of other CaO:P₂O₅ ratios are estimated in proportion to that ratio.

Source of sulfuric acid

Some aspects of the source of the acid will affect plants for phosphoric acid production. Most, but not all, phosphoric acid plants have onsite facilities for producing sulfuric acid from sulfur or pyrites. In this case, heat is recovered from the sulfuric acid plants in the form of steam, which is available for concentrating phosphoric acid and other uses.

Finally, sulfuric acid from pyrites, smelter operations, or other by-product sources may contain impurities that may or may not be deleterious for phosphoric acid production. In at least one case, zinc in smelter acid proved useful since the fertiliser produced from phosphoric acid contained enough zinc, mainly derived from the smelter acid, to improve crop yields in zinc-deficient areas. The same benefit applies to another micronutrient, copper.

Receiving and storing raw materials

An efficient system for bulk handling and storing phosphate rock and other raw materials is necessary for a modern phosphoric acid plant. The criteria to be met are:

a. Rapid unloading of ships or other delivery units;
b. Negligible loss of rock;
c. Easy storage with the ability of separating shipments or blending shipments as desired;
d. Efficient retrieval from storage;
e. Protection against wind, rain, snow and freezing weather;
f. Protection from contamination with other raw materials, windblown dust, soil, etc.; and
g. Provision for expansion if future needs warrant.

When phosphate rock is received dry, it is usually desirable to keep it dry by covered storage, especially if it is to be used in a dry grinding system to avoid the expense of redrying. If open storage is used, wind or heavy rains can cause losses of rock that may amount to several percent if it contains many fines. However, relatively coarse rock can be stored in open piles, particularly if it is to be wet ground. The storage capacity should be at least 1.5 times the largest shipment to allow for delays. Even larger storage capacity may be advantageous for blending shipments.

Rock grinding and preparation

The choices in rock grinding are dry grinding, wet grinding, or no grinding. Several processes claim the ability to use rock without grinding if it is finer than 35-mesh or in some cases 20-mesh (Tyler) screen size (approximately equivalent to the standard 425 and 850 mm sizes.

Most of the older plants and some of the new ones use dry grinding. Pendular roller mills or ball mills are often used with air classification. The power requirement naturally depends on the initial size of the rock, its hardness, and the desired particle size. For grinding Florida rock to 55% minus 200-mesh, a requirement of 15-20 kWh/ t of rock has been suggested, including air classification and pneumatic conveying to ground rock storage. Softer rocks may require one-half to two-thirds as much power.

There is a general tendency toward wet grinding in newer plants that are located near the mine. The wet grinding is done in a ball mill; a slurry containing 60%-70% solids is produced and fed to the digester via a surge tank. Advantages of wet grinding are 30%-40% reduction in power requirement and elimination of dust losses, atmospheric pollution by dust, and the necessity of drying the rock. The main disadvantages are somewhat faster wear of the balls and mill lining, a more complex water balance for the phosphoric acid plant with a potential increase of effluent flow, and a decrease in the amount of recycled wastewater that may be used in phosphoric acid production and which may contain some ions of value.

It is necessary to maintain reasonably close control of the water: solids ratio in grinding. The power requirement for wet grinding 163 tph of Florida rock to minus 35-mesh was given by Shearon as 1,865 kW for closed-circuit grinding and 2,835-2,984 kW for open circuit grinding [13]. The corresponding energy consumption is 11.4 kWh/t and about 18 kWh/t for closed and open-circuit wet grinding, respectively.

There is some difference of opinion about the need for fine grinding of very unreactive rocks such as igneous apatite. Lutz and Pratt [14] suggest that such rock should be ground to 80% minus 200-mesh; whereas Somerville considers that rock reactivity is not a major factor. According to Somerville, unground Meramec (Missouri) apatite (minus 150-mesh) is satisfactory. The explanation of this difference of opinion probably lies in the type of digester used [11].

Calcining of phosphate rock usually is considered part of beneficiation. However, some phosphoric acid producers who purchase rock also calcine it to eliminate organic matter or to decrease carbonate content or both. One purpose in calcining is to improve the colour of products such as liquid fertiliser or nonfertiliser products such as sodium tripolyphosphate. A saving in foam-control reagents is another advantage.

Reaction system – acidulation stage

There are many types of reaction systems in use. The objective in designing the reaction system is to carry out the reaction between phosphate rock and sulfuric acid so as to recover a maximum percentage of the P₂O₅ from the rock as product phosphoric acid in the simplest and least expensive manner. Since the filtration step is the most critical and expensive step in the process, a primary objective in the reaction step is to form gypsum crystals of such size and shape that the filtration and washing can be carried out rapidly and efficiently.

Maximising recovery means minimising losses. Three types of P₂O₅ losses are recognised:

• unreacted phosphate rock,
• P₂O₅ cocrystallised with gypsum through isomorphic substitution of HPO₄^– for SO₄^–, and
• phosphoric acid lost in the gypsum due to incomplete washing.

Perhaps a fourth source of loss should be mentioned – mechanical losses due to spillage, leakage, washing of filter cloth, piping, and equipment for scale removal, and losses as sludge.

The purpose of the reaction step is not only to extract the phosphate from the rock but also to ensure slow growth of gypsum crystals to a relatively large size. To attain this goal, reaction systems are designed to prevent direct contact between the two reactants, phosphate rock, and sulfuric acid. A high concentration of free sulfuric acid would result in coating the phosphate rock with calcium sulfate reaction product, thus blocking further reaction.

A serious case of ‘reaction blocking’ in a phosphoric acid plant can take several hours or even days to correct. On the other hand, a high concentration of calcium ions (low sulfate) in the slurry will increase the amount of phosphate cocrystallised with the gypsum. Hence, the aim of designers and operators of reaction systems is to maintain a uniform composition of the slurry, avoiding pockets of high sulfate or calcium concentration. The liquid phase usually consists of phosphoric acid (about 25-30% P₂O₅) with about 2-2.5% free sulfuric acid; the optimum concentration of free sulfuric acid varies with rock composition. The solid phase is mainly gypsum. The proportion of solids in the slurry is about 30%-35%. Phosphate rock particles introduced into this slurry dissolve rapidly in the phosphoric acid in the liquid phase, which causes supersaturation with calcium sulfate and results in the growth of gypsum crystals.

To approach this ideal situation, the incoming streams of sulfuric acid and phosphate rock are mixed with the slurry (directly or indirectly) as rapidly and completely as possible, and the slurry in the reaction system is agitated to ensure homogeneity. The phosphate rock may also be premixed with recycled reaction slurry. Sulfuric acid may be sprayed on to the surface of the slurry in the reactor vessel or premixed with recycled weak phosphoric acid.

In many older plants with reactor air cooling, the sulfuric acid was diluted, sometimes to 55% – 60% H₂SO₄ and cooled in a heat exchanger before use. Most modern plants, with flash cooler cooling, use the sulfuric acid at the concentration at which it is received, usually 98-98.5% H₂SO₄.

When strong sulfuric acid is premixed with weak recycled phosphoric acid, much heat is released, and this is accompanied by evaporation of water and volatilisation of fluorine compounds (mainly SiF₄ and HF). A rich stream of fluorine compounds is provided in the vapour, which can be recovered as fluosilicic acid for sale or further processing.

The various dihydrate processes differ essentially in the reaction step. The reactor design generally falls into one of two categories, multicompartment reactors or single-tank reactors.

Processes using more than one tank or compartment include the Prayon Mark IV dihydrate process, which is now one of the most widely used processes. It led to the development of the Prayon pH range of hemihydrate processes and the Norsk Hydro dihydrate processes, from which the Norsk Hydro HH and HDH processes were developed. In the Prayon Mark IV dihydrate process, phosphoric acid is produced at a concentration. of about 28% P₂O₅. The reactants are fed to the multi compartment attack tank (Figure 1), which is constructed of concrete and lined with rubber and carbon brick. Specially designed openings are provided in the attack tank inner walls so that the reaction slurry can flow from one compartment to the next. Each compartment is provided with a single agitator of special design to fulfil the functions of mixing, solids suspension, and foam breaking.

The attack tank is cooled by circulating the slurry through a low-level flash cooler, in which the reaction slurry is cooled by evaporation of water under vacuum. Circulating power requirements are kept to a minimum by using an axial-flow circulation pump and by the low elevation of the flash cooler above the attack tank. The temperature in the attack tank is controlled by varying the vacuum applied to the flash cooler.

Slurry from the tank overflows to the digestion system, where the calcium sulfate crystallises. The digestion section can be a single tank, depending on plant capacity. For larger plants it will comprise two or three digestion tanks of rubber-lined carbon steel; each is equipped with a single agitator.

The reaction tank (Figure 2) is constructed of rubber-coated steel or concrete and is lined with carbon bricks. Baffles are fitted onto the walls to prevent the slurry from rotating bodily as a single mass inside the tank; The cover of the reactor is constructed of polyester or ebonite-coated panels. The phosphate rock is fed by a special duct within a cylindrical shroud at one or two points, according to the size of the tank, and into the turbulence zone of the central agitator on the opposite side to the gas extraction hood. The sulfuric acid is introduced into one, or several, independent discs fixed to the drive shafts of some of the surface coolers. The proprietary equipment distributes the acid so evenly over the entire surface of the tank that 98% acid can be introduced directly without prior dilution. There is no risk of local sulfuric acid concentration excess or temperature peaks, which can adversely affect the crystallisation.

Temperature control is provided by a flow of air over the surface of the slurry in the reactor. The tank cover, positioned between 1 and 1.2 m above the slurry level, is perforated to allow the entry of atmospheric air across about one-half the diameter of the tank and is fitted with a gas outlet hood on the opposite side of the tank. A fan provides the circulation of air. Rhone-Poulenc’s DIPLO process [16] is a variation of the single-tank process (Figure 3). It is based on the same principles as the single-tank process but differs in that two attack tanks are employed in series with no recirculation between them. The process has been developed to handle low-grade, less-reactive, or unground phosphates. The two processes (single-tank and DlPLO) developed by Rhone-Poulenc were acquired by Speichim, which is now owned by Technip.

Similar in some respects to the above-described Rhone-Poulenc system is the process developed by Societe Industrielle d’Acide Phosphorique et d’Engrais (SIAPE, Tunisia), which also features a single tank in the reaction section although its construction is somewhat different. The reaction system comprises a cylindrical tank divided into a central compartment, (into which the phosphate rock, sulfuric acid, and returned acid are fed), and an outer ring-shaped compartment. The tank is usually constructed of concrete and is coated on the bottom and sides with rubber, which is then covered with carbon bricks.

The process was specially developed to use the low grade Gafsa rock, which has a relatively high carbonate content, and is primarily used in Tunisia. The carbon dioxide released in the attack section promotes slurry circulation. Because of the high gas concentration, a slurry low-density zone is created. By having one opening at the bottom, a natural draft is set up, which moves the slurry out in a continuous flow. This circulation is further enhanced by an axial-radial double-impeller agitator working as a pump.

Since the quality of indigenous rock has continued to decrease, SIAPE has modified its process by adding a digestion tank with two agitators (Figure 4). Slurry at 78°-80°C is fed from the outer section of the reactor to the digestion tank, where it is cooled to 72°-73°C before being returned to the central compartment of the reactor. This arrangement allows a temperature gradient to be maintained between the recycle slurry and the slurry in the attack tank, which is necessary for good crystal growth [17, 18].

Another type of single-reactor system is Jacob’s dihydrate phosphoric acid process comprising an annular reactor with separate cooler seal and filter feed compartments [18]. The reactor design allows phosphate rock and sulfuric acid to be added at several points. The reactor is equipped with multiple agitators, and cooling is generally provided by low-level flash coolers (Figure 5). The combined action of the cooler circulation pumps and back-mixing from agitator to agitator provides the necessary degree of recirculation in the reactor. The system allows easy sulfate control, good crystal growth, and low nucleation rates. This process is one of the more widely used ones.

Among processes using a single-tank reactor, some processes are the conventional type and others the so-called ‘isothermal’ type. In isothermal processes the reactor is maintained at a constant value – hence the name ‘Isothermal’ – by keeping the contents in rapid circulation. However this type of process is used by only a few plants.

Retention times in industrial plants are typically around 3-4 hours. One of the reasons relates to the formation of good gypsum crystals, as discussed previously. Another reason is the difficulty of dose control of free sulfuric acid (SO₄^-2) content of the liquid phase when the reaction time is short. Close control of this value is extremely important. Although the optimum level of control may depend on the character of the rock, a level of about 1.5% is typical. Serious upsets can occur when the SO₄^-2 level varies appreciably from the optimum. Obviously, the shorter the reaction time, the faster (and more often) problems can arise.

Filtration system

The function of the filtration step is to separate the gypsum (and any insoluble materials derived from phosphate rock or formed in the reaction) from the phosphoric acid product as completely, efficiently, and economically as possible. All modern plants use only continuous horizontal vacuum filters.

The most popular types of filters are the tilting-pan rotary filters, rotary table filters, and belt filters. In each of these filters, the cycle proceeds through the following steps:

• deposition of the phosphoric acid-gypsum slurry on the filter,
• collection of product acid by application of vacuum,
• two or three countercurrent washes to complete the removal of phosphoric acid from the gypsum,
• discharge of the washed gypsum, and
• washing of the filter cloth to prevent accumulations of scale-forming materials.

The sequence of operations is illustrated in Figure 6. In the washing sections, successively weaker solutions of phosphoric acid are collected. The last wash is with fresh water or sometimes with water recycled from a gypsum pond or from sumps that collect cloth-washing water and spillage or drippings.

The very weak acid collected in the last section is returned to the preceding section with the filtrate from the first wash being recycled to the reaction vessels. Some of the product acid also may be recycled to the digestion step to control the percentage of solids in the slurry at a manageable level, usually 25%-40%.

Filters are usually rated according to their active surface area, which may range up to about 320 m². The rate of rotation (in a rotary filter) or the rate of travel (in a belt filter) is variable to permit adjustments as required by the filtration characteristics of the slurry and other factors.

The production rate may vary widely, depending mainly on the filter type and the phosphate rock quality, but a common design factor is 3- 6 tonnes of P₂O₅/m²/day, sometimes higher. This rate has been falling as rock quality declines. The filtration rate is affected primarily by the size and shape of gypsum crystals which, in turn, are affected by conditions in the reaction section including the type of phosphate rock, use of crystal habit modifiers (CHM), control of reaction conditions, etc. Insoluble impurities in the rock, such as clay and silica, may affect filtration rates adversely [19]. The filtration rate is also affected by the temperature, concentration viscosity of the acid and the desired recovery. While many plants strive for maximum recovery, in specific plants there is often an economic optimum operating rate at which increased production is attained at some sacrifice of recovery.

Concentration and clarification

Phosphoric acid produced by most dihydrate processes contains 25%-30% P₂O₅ (filter acid). Acid of this concentration can be used in some fertiliser processes, but for most purposes it is economically preferable to concentrate it by evaporation of part of the water content. The desired concentration depends on the use; the requirements are given in Table 5. The above concentrations are merely guides to standard practice; it is quite possible to use other concentrations in most cases. For instance, 30% P₂O₅ acid has been used for TSP production by a process requiring extensive drying of the product. However, energy is usually more efficiently used by concentrating the acid than by drying the product with high rates of recycle. This is especially true when energy is available in the form of steam from an adjacent sulfuric acid plant. Precipitates form in phosphoric acid before, during, and after concentration. Compounds precipitating before concentration are likely to be mainly calcium sulfate and fluosilicates. A wide variety of compounds may form during and after concentration, depending on acid concentration. These compounds are collectively known as ‘sludge’ and cause many difficulties in handling and use of the acid. They also form scale in evaporators. Therefore, many manufacturers clarify the acid and either recycle the sludge or use it in fertiliser products where it causes the least trouble. Filtration on a press filter is a third option to recover the acid and dispose of the solids. Acid for shipment, in particular, should be well clarified. Phosphoric acid produced by hemihydrate processes (40%-50% P₂O₅) is relatively free from sludge. However as rock quality continues to decline, even HH processes can now suffer from post-precipitation and may require a settling step with sludge handling.

The amount of fluorine removed during concentration from 30% to 54% P₂O₅ may be 70%-80% of that originally present in the acid, most of which is volatilised and recovered as fluosilicic acid to prevent pollution of the cooling water in the condenser. In some cases saleable fluorine byproducts are produced such as fluosilicic acid, fluosilicates, cryolite, aluminum fluoride, hydrofluoric acid, and even liquid hydrogen fluoride.

Phosphoric acid concentrators may be classified as direct-fired or indirectly heated. In direct-fired evaporators, combustion gases come into direct contact with the acid, as in a spray tower or submerged combustion evaporators. Use of this type of evaporator has been abandoned because of the difficulty of cleaning the exhaust gases to recover acid mist and fluorine compounds.

Most phosphoric acid concentration processes heat the acid with steam in a heat exchanger under vacuum (Figure 7). Commonly used are tubular heat exchangers with forced circulation; the tubes can be of graphite or stainless steel. The impregnated graphite tubes are cheaper but are fragile and crack during operation. There was an opinion of users that over a period of a few years the capital and replacement costs for both systems were about the same. Now the situation has changed somewhat in favour of graphite tube exchangers because better materials are being used. The other type of exchanger is carbon tubes. Concentration from 26% to 54% P₂O₅ using steam heating may be carried out in one, two, or three stages, sometimes with intermediate clarifying steps to decrease scale formation in the heat exchangers.

Concentration from 28% to 54% P₂O₅ requires about 2 tonnes of steam per tonne of P₂O₅ in the acid concentrated. This amount is usually available from the sulfuric acid manufacture if the acid is produced by burning sulfur. Electric power requirements may range from 11 to 16 kWh/tonne of P₂O₅, depending on the scale of operation. About 6 tonnes of cooling water per tonne of P₂O₅ is required for condensing the water evaporated from the acid.

Utilisation of sludge

As mentioned previously, sludge impurities precipitate in phosphoric acid before, during, and after concentration. If the acid is used onsite for fertiliser production, it may be possible to use the acid without separating the sludge. However, in some cases the amount of sludge may be so great as to lower the grade of fertiliser products below that desired.

Sludge solids that form in filter acid (28% P₂O₅) are mainly gypsum and fluosilicates and may, in some cases, be returned to the phosphoric acid production unit without serious interference with its operation. Sludge forming after concentration is likely to contain a high proportion of iron and aluminum phosphate compounds. One example is (Al,Fe)₃KH₁₄(PO₄)₈•4H₂O, also known as ‘X-compounds’. Lehr has identified 38 distinct crystalline compounds occurring in sludge from wet-process acid [21]. The return of iron and aluminum compounds to the acid-production unit is likely to cause some difficulties, such as an increase of acid viscosity, phosphate precipitation and a reduction in the filtration rate. When this sludge must be separated, as is usually the case for shipment. it is often valorised or used for the production of Triple Super Phosphate (TSP). Most of the P₂O₅ in the sludge solids is citrate soluble but not water soluble; therefore, this solution is not advantageous when the TSP is sold on the basis of water solubility.

The sludge may be used in the production of nongranular monoammonium phosphate (MAP) which, in turn, is used as an intermediate in the production of compound fertilisers. There is no standard grade for MAP to be used as an intermediate; the user can formulate compound fertilisers on the basis of actual analysis. In this case also, the iron, aluminum, and magnesium compounds are not water soluble. In fact, there is no economical method for using sludge solids in countries where phosphate fertiliser is sold on the basis of water solubility.

Precipitation after concentration to 54% P₂O₅ is slow and never so complete but that more precipitate will form on further storage. However, clarification methods are available that reduce the sludge problem in merchant-grade acids to manageable levels.

Dihydrate process requirements

Raw material and utilities requirements vary among dihydrate processes, but the values shown in Table 6 may be considered mid-range values.

References

8. ‘WPA Production Using Phalaborwa Phosphate Rocks.’ 1993. Phosphorus and Potassium, 184:23-33.

9. ‘Phosphate Rock Grade and Quality.’ 1992. Phosphorus and Potassium, 178:28-36.

10. ‘The Acid Test.’ 1993. Phosphorus and Potassium, 188:21-32.

11 L Somerville, R. L. 1973. ‘Fundamentals of Wet Process Phosphoric Acid Plant Design,’ Paper presented before the American Chemical Society, Division of Fertilizer and Soil Chemistry, Chicago, IL, U.S.A.

13. Shearon, G. B. 1975. ”Wet Grinding and Feeding of Phosphate Rock — Plant City Phosphate Complex,’ IN Proceedings of the 25th Annual Meeting of the Fertilizer Industry Round Table. pp. 164-169, Washington, D.C. (U.S.A.).

18. ‘Reactors, Agitators and Filters for Phosphoric Acid Plants.’ 1991. Phosphorus and Potassium; 174: 23-37.

19. ‘Effect of Ionic Impurities on Gypsum Filterability.’ 1990. Phosphorus and Potassium, 168:20-21.

20. Moraillon, P., J. E Gielly, and B. Bigot. 1968. ‘Principles of Filter Design and Operation,’ IN Phosphoric Acid, A. V. Slack (Ed.), pp. 407-442, Marcel Dekker, Inc., New York, NY, U.S.A.

21. Lehr, J. R. 1968. ‘Nature of Impurities,’ IN Phosphoric Acid, A.V. Slack (Ed.), pp. 637-686, Marcel Dekker, Inc., New York, NY, U.S.A.

Links to Related IFS Proceedings

70, (1962), Origin of Processing of Phosphate Rock with Particular Reference to Benefication, Vincent Sauchelli

81, (1964), Insoluble Phosphate Losses in Phosphoric Acid Manufacture by the Wet Process: Theory and Experimental Techniques, S M Janikowski, N Robinson, W F Sheldrick

112, (1969), Single Stage Process for the Production of 50% Phosphoric Acid, L E Bostwick, W Turner

112, (1969), Kellogg-Lopker Phosphoric Acid Process, W C Weber, E J Roberts, I S Mangat, E Uusitalo

112, (1969), Dorr-Oliver HYS Phosphoric Acid Process, A C van Es, J Th Boontje

151, (1975), Newer Developments in Cleaning Wet Process Phosphoric Acid, R Blumberg

201, (1981), From Wet Crude Phosphoric Acid to High Purity Products, A Davister, G Martin

249, (1986), A Clean Technology Phosphoric Acid Process, S van der Sluis, Y Meszaros, J A Wesselingh, G M van Rosmalen

269, (1988), Phosphoric Acid – Wet Process: What Process? A Guide to Process Selection for Phosphoric Acid Manufacture by Sulphuric Acid Dissolution, P A Smith

806, (2017), Capturing phosphoric acid know-how in a training simulator, A Durand and S Joao

821, (2018), Approaches to improving the quality of phosphoric acid, T Henry

Links to external sources

Becker, P. (1989) Phosphates and Phosphoric Acid: Raw Materials: Technology, and Economics of the Wet Process. Marcel Dekker, Inc., New York, NY, U.S.A.

Havelange, S. et al. (2022). Phosphoric Acid and Phosphates in Ullmann’s Encyclopaedia of Industrial Chemistry.

Slack, A.V. (1968). Phosphoric Acid (Part I and II). Marcel Dekker, Inc., New York, NY, U.S.A.