Method and system for operating a chemical plant

Abstract

The present invention provides a method (100a) and system for controlling the operation of a chemical plant arranged to carry out a chemical process which produces iron and / or manganese in elemental form from respective oxides thereof using liquid sodium as a reductant of the respective oxides and which can also produce an oxide or hydroxide of at least one of calcium, magnesium and iron from a carbonate mineral of at least one of calcium, magnesium and iron. The method (100a) comprises assigning (103a) a respective value to the no. of mol of each one of the chemical species involved in the chemical process as starting materials and products thereof, subject to constraints which are derived from the law of conservation of mass of the constituent elements of those chemical species. The method also assigns (103b) respective values to the nos. of mol of chemical species which are consumed during the chemical process by electrolytic or thermochemical decomposition into their constituent elements, subject to another set of constraints which are derived from the minimum requirements for these constituent elements by the chemical process itself. The former set of constrains allows the materials balance of the process to be controlled and the latter set of constraints allows the energy balance of the process to be controlled. The method then comprises operating (104) the chemical plant to carry out the process according to the values thus assigned. Respective values may be assigned (103a, 103b) to the nos. of mol of a preferred subset of the chemical species involved in the process as starting materials and products thereof and / or a preferred subset of the chemical species which are consumed during the process by electrolytic or thermochemical decomposition into their constituent elements, before values are then assigned to remaining ones of the nos. of mol of other chemical species not in the preferred subset. This allows the chemical plant to be operated in such a way as to optimise its performance relative to such variables as the respective nett amounts of a preferred one or more of the materials it consumes and / or produces, the energy efficiency of a part or the whole of the chemical process and / or its cost effectiveness, for example. Amongst other things, the constraints also ensure that the chemical process always produces less carbon dioxide than instead calcining a carbonate mineral of at least one of calcium, magnesium and iron to produce its corresponding oxide, and that in a large number of cases, it can produce no carbon dioxide at all. Also disclosed is a graphical technique which may be used to assign the respective values to the nos. of mol of the chemical species involved in the chemical process subject to the stated constraints, whereby the method and system of the invention may be implemented.

Description

[0001] Method and System for

[0002] Operating a Chemical Plant

[0003] Field of the Invention

[0004] The present invention concerns a method and system for operating a chemical plant. It therefore belongs to the field of control methods and systems for industrial plants and processes, and in particular to those for controlling plants and processes in the chemical industry.

[0005] Background of the Invention

[0006] The present applicant's co-pending UK patent application no. 2417075.5 ("Industrial Chemical Process and Apparatus”; applicant's ref: NE-P-GB 007) describes an industrial chemical process, which integrates a method for producing iron and / or manganese in elemental form from oxides thereof with a method for producing an oxide or hydroxide of at least one of calcium, magnesium and iron from an ore comprising a carbonate mineral of at least one of calcium, magnesium and iron. In the first of these methods, an amount of liguid sodium in excess of the stoichiometric amount thereof is used in a redox reaction between the liguid sodium and an oxide of iron or manganese, respectively, to precipitate out from the liguid sodium both the respective metal in elemental form and other insoluble products comprising sodium oxide. This first method is described in more detail in the present applicant's co-pending UK patent application nos. 2417059.9 ("Carbon-Free Method and Apparatus for Producing Iron and Steel”; applicant's ref: NE-P-GB 001) and 2417063.1 ("Carbon-Free Method and Apparatus for Producing Manganese”; applicant's ref: NE-P-GB 008).

[0007] In the other of the aforementioned methods, namely that for producing an oxide or hydroxide of at least one of calcium, magnesium and iron from an ore comprising a carbonate mineral of at least one of calcium, magnesium and iron, chlorine gas produced by electrolysis is used to produce hydrogen chloride, and at least some of this hydrogen chloride is dissolved in liguid water to produce hydrochloric acid. The ore comprising the carbonate mineral is then added to at least some of this hydrochloric acid to dissolve the carbonate mineral therein and produce carbon dioxide gas and an agueous solution respectively comprising at least one of calcium chloride, magnesium chloride and ferrous chloride. A first portion of the sodium oxide from the redox reaction of the first method, or sodium hydroxide derived from hydrating at least some of the first portion of sodium oxide, is reacted with at least some of this agueous solution to produce an agueous solution of sodium chloride and a precipitate respectively comprising at least one of calcium hydroxide, magnesium hydroxide and ferrous hydroxide. Any of these hydroxides may then be dehydroxylated to produce their respective oxides. Meanwhile, a second portion of the sodium oxide from the aforementioned redox reaction, or sodium hydroxide derived from hydrating at least some of the second portion of sodium oxide, is used in at least one of: (I) a reaction with at least some of the hydrochloric acid to produce an agueous solution of sodium chloride, and (ii) a reaction with at least some of the carbon dioxide gas to produce at least sodium carbonate. The agueous solution of sodium chloride produced from the first portion of sodium oxide and / or from the second portion of sodium oxide may then be recycled back for electrolysis. For example, it may be dried to produce solid sodium chloride, which may then be fused and electrolysed using the Downs process to produce liquid sodium and chlorine gas again.

[0008] The hydrogen chloride used to produce the hydrochloric acid may be made by combusting the chlorine gas produced by electrolysis with hydrogen gas, and / or by reacting it in a reverse Deacon reaction (RDR) with high- temperature water vapour derived from the aforementioned hydroxide(s) when they are dehydroxylated to produce their respective oxide(s). The latter technique is described in more detail in the present applicant's co-pending UK patent application no. 2417058.1 ("Method and Apparatus for Producing Hydrochloric Acid”; applicant's ref: NE-P- GB 006). If the former technique is used to make at least some of the hydrogen chloride, at least some of the hydrogen gas for combustion with the chlorine gas may be produced by electrolysis, for example of water and / or as a product of the chlor-alkali process for the electrolysis of brine and / or of the Castner process for the electrolysis of sodium hydroxide. However, the present applicant's co-pending UK patent application no. 2517327.9 ("Method and Apparatus for Producing Liquid Sodium”; applicant's ref: NE-P-GB 011) also describes a method whereby at least some of the hydrogen gas may be produced other than by electrolysis, by instead using a thermochemical process, which decomposes solid-phase NaOH into its constituent elements. The same process described therein also produces liquid sodium, which may then be used in the redox reaction with the oxide of iron or manganese. As also described therein, at least some of the solid-phase NaOH which is decomposed in this manner may be derived by hydrating at least some of the second portion of sodium oxide produced by the redox reaction itself, thus recycling this liquid sodium without the need to produce more of it by electrolysis.

[0009] Moreover, in the chemical process described in UK patent application no. 2417075.5, the molar ratio of input materials, namely of the respective oxides of iron and / or manganese on the one hand to the carbonate mineral on the other, can be varied as desired. Thus there are several choices available when operating this industrial chemical process. These include: the liquid sodium used in the redox reaction may be made by electrolysis and / or by decomposing solid-phase NaOH into its constituent elements, the hydrogen chloride used to make the hydrochloric acid may be made by combusting chlorine gas produced by electrolysis with hydrogen gas, in which case, the hydrogen gas may be made by electrolysis and / or by decomposing solid-phase NaOH into its constituent elements, and / or the hydrogen chloride may be made by reacting the chlorine gas in an RDR with high-temperature water vapour. Moreover, the relative sizes of the first and second portions of sodium oxide produced by the redox reaction, as well as whether neither, either or both of these portions are hydrated to make sodium hydroxide, are also adjustable, as are the proportions in which the second portion thereof is (I) reacted with at least some of the hydrochloric acid to produce an aqueous solution of sodium chloride, (ii) reacted with at least some of the carbon dioxide to produce at least sodium carbonate, and / or (ill) hydrated to produce solid-phase NaOH which is then decomposed into its constituent elements.

[0010] In the chemical process described in UK patent application no. 2417075.5, the oxides of iron and / or manganese on the one hand and the carbonate mineral on the other are both essential ingredients for respectively producing iron and / or manganese in elemental form, and an oxide or hydroxide derived from the carbonate mineral and oxygen as their corresponding products. In addition, however, common salt ( / '.e. , sodium chloride) and / or water are optional extra ingredients, which act as intermediaries in this integrated chemical process to produce its endproducts, but are only themselves consumed if supplied in excess. The various choices described above affect how much salt and / or water are consumed, as well as the corresponding amounts of other products produced, in addition to those already mentioned. For example, the total amount of sodium oxide, hydroxide, carbonate and / or silicate produced is determined by the amount of salt consumed, as is the corresponding amount of at least one of chlorine, hydrogen chloride and hydrochloric acid produced. Similarly, an amount of hydrogen produced, if any, is determined by the amount of water consumed. These choices also affect the total energy consumption of the process, as well as the relative amounts of its starting materials and products.

[0011] Accordingly, it would be desirable to provide a method and system for operating a chemical plant which is constructed to carry out such an integrated chemical process. These may therefore also be considered as a control method and system, respectively, for the chemical plant carrying out this process. It would also be desirable if such a method of operation and / or operating system could be used to adjust the operation of the chemical plant in such a way as to optimise its performance relative to such variables as the respective amounts of different ones of the materials it consumes and / or produces, its energy efficiency and / or its cost effectiveness, for example.

[0012] The entire contents of each of co-pending UK patent application nos. 2417075.5, 2417059.9, 2417063.1 , 2417058.1 and 2517327.9 mentioned above are incorporated herein by reference.

[0013] Object of the Invention

[0014] It is therefore an object of the invention to provide a method and system for operating a chemical plant that is arranged to carry out a chemical process which produces iron and / or manganese in elemental form from respective oxides thereof using liquid sodium as a reductant of the respective oxides and which can also produce an oxide or hydroxide of at least one of calcium, magnesium and iron from a carbonate mineral of at least one of calcium, magnesium and iron.

[0015] Description of the Invention

[0016] Accordingly, in one aspect, the present invention provides a method comprising operating a chemical plant arranged to carry out a chemical process which produces iron and / or manganese in elemental form from respective oxides thereof using liquid sodium as a reductant of the respective oxides and which can produce an oxide or hydroxide of at least one of calcium, magnesium and iron from a carbonate mineral of at least one of calcium, magnesium and iron, wherein the process involves a plurality of chemical species, X, as starting materials and products thereof, wherein X represents any one of M2O3, QCO3, NaCI, H2O, M, Q(OH)2, QO, Na2CC>3, CO2, Na2O, NaOH, Na4SiC>4, CI2, HCI, H2 and O2, and the method comprises the following. Assigning a respective value to each n(X), where n(X) is the no. of mol of chemical species X, subject to the following constraints: 0 < n(QC03) / n(M2O3) < 3 [Eqn. 1 a] n(M) = 2n(M2O3) [Eqn. 1 b] n(Q0) + n(Q(0H)2) = n(QC03) [Eqn. 1c] n(C02) + n(Na2CO3) = n(QC03) [Eqn. 1 d]

[0017] 2n(Na2O) + n(NaOH) + 2n(Na2CO3) + 4n(Na4SIO4) = n(NaCI) [Eqn. 1e]

[0018] 2n(CI2) + n(HCI) = n(NaCI) [Eqn. 1f] n(H2) + n(Q(0H)2) + [n(HCI) + n(NaOH)] 1 2 = n(H2O) [Eqn. 1g] n(Na20) + n(NaOH) + n(Na2CO3) + 2n(O2) = 3n(M2O3) + n(H2O) [Eqn. 1 h] wherein M represents at least one of Fe and Mn, Q represents at least one of Ca, Mn and Fe, the left-hand side of each of Eqns. 1 b to 1 h represents nett production by the chemical process and the right-hand side of each of Eqns. 1 b to 1 h represents nett consumption by the chemical process of each respective one of the chemical species, X, identified in Eqns. 1 a to 1 h. Assigning a respective value to each of n(NaCI)e, n(H20)eand n(NaOH)d, each of which is as defined hereinbelow, subject to the following constraints:

[0019] 0 < n(Na)d / n(M2O3) < 6 - 2r [Eqn. 2a] 2r < n(Na)e / n(M2O3) < 6 [Eqn. 2b] [n(Na)d+ n(Na)e] / n(M2O3) > 6 [Eqn. 2c]

[0020] 0 < n(H2)d / n(M2O3) < 3 - r [Eqn. 2d]

[0021] 0 < n(H2)e / n(M2O3) < r [Eqn. 2e]

[0022] [n(H2)d+ n(H2)e] / n(M2O3) > r [Eqn. 2f] wherein: n(Na)d= n(NaOH)d[Eqn. 3a] n(Na)e= n(NaCI)e[Eqn. 3b] n(H2)d = n(NaOH)d / 2 [Eqn. 3c] n(H2)e= n(H20)e[Eqn. 3d] and r represents the value assigned to the ratio n(QCC>3) I n(M2C>3). In Eqns. 3a to 3d, the right-hand side of each equation represents nett consumption by electrolytic or thermochemical decomposition of the stated chemical species into its constituent elements and the left-hand side of each equation represents nett production of the stated element by that decomposition. The method then comprises operating the chemical plant to carry out the process according to the values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned.

[0023] This method has at least the following advantages. Eqn. 1 a places a constraint on the ratio of the two starting materials QCO3 and M2O3, which ensures that a chemical plant controlled by this method always produces less carbon dioxide per n(QCC>3) it consumes than a plant instead conducting traditional calcination of QCO3 according to the equation:

[0024] QCO3 (S) — > QO (S) + C02<g) [Eqn. 4] Indeed, when r < 1.5, the process carried out by the chemical plant can produce no carbon dioxide at all. The constraints of Eqns. 1 b to 1 h derive from the law of conservation of mass of the constituent elements of the chemical species, X, which are involved in the chemical process as starting materials and products thereof, as follows:

[0025] M: [Eqn. 1 b]

[0026] Q: [Eqn. 1c] C: [Eqn. 1d] Na: [Eqn. 1e] Cl2: [Eqn. 1f] H2: [Eqn. 1g] O2: [Eqn. 1 h]

[0027] They therefore allow the respective value of each n(X) to be varied in a systematic way, so that consumption of a selected one or more of the starting materials and / or production of a selected one or more of the products of the chemical process may be optimised for the value assigned to the ratio of Eqn. 1 a. Assigning a respective value to each of n(NaCI)e, n(H2O)eand n(NaOH)d assigns a respective value to the no. of mol of each one of the chemical species which are consumed during the chemical process (although not necessarily as starting materials thereof) by being decomposed into their constituent elements either electrolytically or thermochemically. These values are assigned subject to constraints which are derived from the minimum requirements for these constituent elements by the chemical process itself. Thus Eqn. 2c defines the minimum requirement for liquid sodium used to reduce the iron and / or manganese oxides into iron or manganese, respectively, and Eqn. 2f defines the minimum requirement for hydrogen used to make hydrogen chloride gas by combustion with chlorine. Since apart from decomposing chemical species into their constituent elements either electrolytically or thermochemically, the chemical process carried out by the plant is exothermic overall, such electrolytic and / or thermochemical decomposition represents the chief component of the energy consumed by the chemical process. This therefore allows the energy consumption of the process to be optimised for the value assigned to the ratio of Eqn. 1 a as well.

[0028] The interrelationships between the respective no. of mol of the chemical species, X, involved in the chemical process as starting materials and products thereof, as well as between the respective no. of mol of the chemical species which are consumed during the chemical process by being decomposed into their constituent elements either electrolytically or thermochemically, and the constraints they are all subject to, may be represented using a graphical technique which is described hereinbelow.

[0029] In some embodiments, the method may comprise assigning respective values to a preferred one or ones ( / .e. , a preferred subset) of each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, subject to the same constraints, before using the same constraints to assign respective values to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d. In such a case, operation of the chemical plant is then optimised to the values assigned to the preferred subset of each n(X), n(NaCI)e, n(H20)eand n(NaOH)d. For example, the respective values assigned to a particular one or ones of the n(X) may be chosen to maximise or minimise production or consumption of one or more preferred chemical species X, and if so, the values subsequently assigned to the remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, subject to the same constraints, will ensure that production or consumption of the remaining species is consistent with maximising or minimising production or consumption, respectively, of the one or more preferred species. In another example, values may initially be assigned just to one or more of the n (N aCI)e, n(H2O)eand n(NaOH)d to target a particular value of energy consumption by the respective electrolytic and / or thermochemical decomposition(s), before the same constraints are then used to assign respective values to the other ones thereof and to each n(X), which are consistent with achieving that particular value of energy consumption.

[0030] In some embodiments, the method may comprise minimising the value of n(C02) for a given value of r, subject to the same constraints, before using the same constraints to assign respective values to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d. Thus the above constraints are used to assign respective values to the remaining n(X) when X CO2, which values minimise the value of n(C02) for a given value of r. In particular, the above constraints imply that for r <3 / 2, n(C02) may be set equal to 0; in other words, that the chemical process may be operated at such values of r without producing any carbon dioxide at all.

[0031] In some embodiments, the method may comprise applying the additional constraint that 1 < n(CI2)eI n(QCC>3) s 1 .5. It follows from Eqn. 2b and from the following equation for fusing and electrolysing sodium chloride:

[0032] NaCI (S) — > NaM+1 / 2Cl2<g) [Eqn. 5] that r < n (Cl2)eI n(M2O3) s 3, which is minimized when n (Cl2)eI n(M2O3) = r. But r = n (QCO3) I n(M2O3) by definition. Therefore, the no. of mol of gaseous chlorine produced by electrolysis of NaCI is minimized when n(CI2)e= n(QC03), i.e., when n(CI2)eI n(QC03) = 1 . Applying this equality as an additional constraint would mean that for a given value of r, only a sufficient amount of chlorine would be produced by electrolysis to make enough hydrochloric acid in which to dissolve the carbonate mineral, and that the chemical process would produce no nett amount of gaseous chlorine, hydrogen chloride or hydrochloric acid in excess of that. However, relaxing this equality slightly so that 1 < n(CI2)eI n(QC03) s 1 .5, more preferably < 1 .25, has the advantages that it gives greater flexibility in assigning values to other ones of the n(X) when X Cl2or HCI, whilst still allowing the chemical process to produce a reduced amount of gaseous chlorine, hydrogen chloride and / or hydrochloric acid as end product(s), any of which may be desired, as well as giving greater flexibility to assigning respective values to n (N aCI)e, n(H2O)eand n(NaOH)d. This more relaxed constraint is also able to accommodate real-world losses and inefficiencies in the process.

[0033] In some embodiments, the method may comprise assigning the respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, subject to the same constraints, for which the total value of (E)in- (E)out for a given value of r lies within 20%, preferably 10%, of its minimum value, wherein (E)inrepresents energy consumed by the chemical process and (E)out represents energy recovered from the chemical process. As already noted above, the chief component of (E)inis the energy required to decompose chemical species participating in the chemical process into their constituent elements either electrolytically or thermochemically. The total value of (E)inis therefore a function of and calculable from the values assigned to n(NaCI)e, n(H2O)eand n(NaOH)d. (E)out, on the other hand, is not simply the total amount of heat generated by exothermic reactions in the chemical process, since some of this heat is inevitably lost to the environment by dissipation. It is therefore better represented by E(WHR), which is the total amount of heat which can be recovered from these exothermic reactions and used to reduce the value of (E)in. Despite this, the value of E(WHR) can still be determined as an empirical quantity, which can either be measured directly from the chemical plant itself or estimated based on the plant's past or expected performance. Thus the energy efficiency of the chemical process can be optimised as (E)in- (E)out.

[0034] In some embodiments, at least one of the values assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d may be varied over time subject to the same constraints. For example, the value of E(WHR) may change over time following start-up or turn-down of the chemical plant, in which case, at least one of the values assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d may also be varied over time in response thereto to maintain a desired optimisation of the plant's operation, such as a desired value of a particular one or more of n(X) or the minimisation of the total value of (E)in- (E)out. Since the constraints described above include not only equalities but also some inequalities, they define a parameter space over which the chemical process may be operated, thereby allowing the values assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d to be varied over time. This has the advantage that operation of the chemical plant may therefore be changed as desired without having to alter the construction or arrangement of the plant itself, provided that the plant is initially constructed to accommodate a desired part or all of the parameter space defined by these constraints.

[0035] In some embodiments, the method may comprise assigning the respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, subject to the same constraints, for which the total value of a parameter, p, given by: p = Z p(X)out m(X)out + Z p(E)out (E)out - {Z p(X)inm(X)in+ Z p(E)in(E)in} [Eqn. 6] lies within 20%, preferably 10%, of its maximum value, wherein (X)inand (E)inrespectively represent chemical species and energy consumed by the chemical process, and (X)out and (E)out respectively represent chemical species produced by and energy recovered from the chemical process. Thus the total value of p is calculable as: p = p(M) m(M) + p(Q(OH)2) m(Q(0H)2) + p(QO) m(Q0) + p(Na2CO3) m(Na2CO3) + p(CO2) m(C02) + p(Na2O) m(Na20) + p(NaOH) m(NaOH) + p(Na4SiO4) m(Na4SiO4) + p(CI2) m(CI2) + p(HCI) m(HCI) + p(H2) m(H2) + p(O2) m(O2) + p(E)TE(WHR) - {p(ore 1) m(ore 1) + p(ore 2) m(ore 2) + p(NaCI) m(NaCI) + p(H2O) m(H20) + p(E)eE(NaCI)en(NaCI)e+ p(E)eE(H2O)en(H2O)e+ p(E)TE(NaOH)dn(NaOH)d} [Eqn. 7]

[0036] In Eqn. 7, m(X) is, of course, related to n(X) by m(X) = Mr(X) n(X), where Mr(X) is just the relative molecular mass of chemical species X in units of mass mol’1. m(ore 1) and m(ore 2) are defined by the following two equations: m(ore 1) = m(M2O3) I (proportion of M2O3 in ore 1) [Eqn. 8a] m(ore 2) = m(QC03) I (proportion of QCO3 in ore 2) [Eqn. 8b]

[0037] In Eqn. 7, the respective values of E(NaCI)e, E(H2O)eand E(NaOH)d are constants which are measurable and are determined by the construction of the chemical plant. E(WHR) is a variable which is partially a function of n(X) for any particular value of r and partially determined by how the chemical plant is constructed, but is also measurable empirically, as noted above. The respective values of each p(X), p(E)eand P(E)T are extrinsic to the chemical plant and are therefore received as inputs to the method. The parameter p in each case may, for example, represent the availability and / or demand for each chemical species X, electrical energy and thermal energy in the chemical process. Whereas most p(X) usually have positive values, p(CC>2) may have a negative value. For example, if p(CO2) represents a tax per unit mass of CO2 produced, then p(CC>2) will have a negative value. In an alternative example, if p(CC>2) represents a carbon credit, then p(CC>2) will have a positive value, but in such a case, m(C02) must instead exceptionally be calculated as the positive-valued difference between the mass of CO2 produced by the chemical process itself and the mass of CO2 produced by a prior art process against which the chemical process is benchmarked.

[0038] In some embodiments in which the respective values assigned to each n(X), n(NaCI)e, n(H20)eand n(NaOH)d maximise the total value of the parameter p as just described, at least one of the values of each p(X), p(E)eand P(E)T received as inputs may be dynamically variable over time, in which case, the values assigned to each n(X), n(NaCI)e, n(H20)eand n(NaOH)d may be varied over time in response thereto, whereby maximisation of the total value of p over time may be maintained.

[0039] Whereas various different embodiments of this method have been described above, each of which respectively focuses on a particular one of its features, such embodiments may also be used in combination with each other. For example, minimising production of carbon dioxide may be prioritised first, after which the energy efficiency of the chemical process may be prioritised next, and so on. Such an order of prioritization may be reflected in the order in which the respective values are assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d according to the method.

[0040] In a second aspect, the present invention also provides a data processing system comprising means for carrying out any of the methods described herein. In a preferred embodiment of such a data processing system, the system may be arranged to carry out the method by receiving a subset of the respective values to assign to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d from a user of the system, applying the constraints described herein to the subset of values thus received to reject and / or modify any of the subset of values received from the user which conflict with these constraints and derive remaining ones of the respective values to assign to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, to provide a full set of respective values to assign which are consistent with the constraints, and then operating the chemical plant to carry out the process according to this full set of consistent values, as is described further hereinbelow. In a third aspect, the present invention also provides a computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out any of the methods described herein.

[0041] In a fourth aspect, the present invention additionally provides a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out any of the methods described herein.

[0042] Brief Description of the Drawings

[0043] Further features and advantages of the present invention will become apparent from the following detailed description, which is given by way of example and in association with the accompanying drawings, in which:

[0044] Fig. 1 is a graph of n(X) I n(M2O3) as a function of r, when X is a species comprising M or Q;

[0045] Fig. 2 is a graph of n(NaOH)d I n(M2O3) and n (N aCI)eI n(M2O3), each as a function of r;

[0046] Fig. 3 is a graph of n(NaOH)d I n(M2O3) and n(H2O)eI n(M2O3), each as a function of r;

[0047] Fig. 4 is a graph illustrating an example of n(HCI) I n(M2O3) as a function of r;

[0048] Fig. 5 is a graph of n(H2) I n(M2O3), n(NaOH)d I n(M2O3), n(NaCI)eI n(M2O3) and n(H2O)eI n(M2O3), each as a function of r, when n(Ch) is minimised;

[0049] Fig. 6 is a graph of n(CC>2) I n(M2O3) and n(Na2CC>3) I n(M2O3), each as a function of r, when n(CC>2) is minimised; Fig. 7 is a graph illustrating an example of other n(X) I n(M2O3), n(NaOH)d / n(M2O3), n(NaCI)e / n(M20s) and n(H2O)eI n(M2C>3), each as a function of r, when n(CC>2) is minimised;

[0050] Fig. 8 is a flow diagram of a first embodiment of a method of operating a chemical plant;

[0051] Fig. 9 is a flow diagram of a second embodiment of such a method;

[0052] Fig. 10 is a flow diagram of a third embodiment of such a method;

[0053] Fig. 11 is a flow diagram of a fourth embodiment of such a method;

[0054] Fig. 12 is a flow diagram of a fifth embodiment of such a method;

[0055] Fig. 13 is a flow diagram of a sixth embodiment of such a method;

[0056] Fig. 14A is a flow diagram of a seventh embodiment of such a method;

[0057] Fig. 14B is a flow diagram of an eighth embodiment of such a method; and

[0058] Fig. 15 is a schematic diagram of an embodiment of a data processing system for carrying out any of the methods described herein.

[0059] Detailed Description

[0060] A graphical technique will now be described, which can be used to assign respective values to each of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints described above. This graphical technique plots these respective values as a function of r. Accordingly, Fig. 1 is a graph which plots the ratio r = n (QCO3) I n(M2O3) on the r-axis or abscissa thereof and n(X) I n(M2O3) on the y-axis or ordinate, where X represents a chemical species participating in the chemical process as a starting material or product thereof. In this graph, positive values on the y-axis represent nett production of a species X by the chemical process and negative values on the y-axis represent nett consumption of a species X by the chemical process. Thus its starting materials have negative y-values, whereas its products have positive y-values. Such a graph may be used to represent the constraints on the respective values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d described above, as follows. In Fig. 1 , the dashed line at r = 3 represents the upper bound of the constraint of Eqn. 1 a and the y-axis itself represents the lower bound thereof. The dashed line at r = 3 and the y-axis therefore define the domain of r. The constraint of Eqn. 1 b is represented as follows. Nett consumption of M2O3 is represented by a horizontal line at y = -1 and nett production of M is represented by a horizontal line at y = +2 because Eqn. 1 b requires that n(M) = 2n(M2C>3). The constraint of Eqn. 1c is represented as follows. Since n(M2O3) is represented by the line at y = -1 and r = n(QCC>3) I n(M2O3) by definition, nett consumption of QCO3 is represented by a line at y = -r. Combined nett production of QO + Q(OH)2 is therefore represented by a line at y = +r because Eqn. 1c requires that n(Q0) + n(Q(OH)2) = n(QC03). Nett production of QO and nett production of Q(OH)2 may be separated out from each other and represented individually, provided that n(Q0) + n(Q(OH)2) = n(QC03) for all values of r over the domain of r, in order to meet the constraint of Eqn. 1c. For example, nett production of QO could be represented by a line at y = +2r / 3, in which case, nett production of Q(0H)2 would have to be represented by a line at y = +r / 3, in order to meet the constraint of Eqn. 1 c. This type of graph may therefore be used to assign values to respective ones of n(X) for any given value of r, subject to the constraints. For example, if nett production of QO were represented by a line at y = +2r / 3, and nett production of Q(0H)2 by a line at y = +r / 3 as just suggested, then for a given value of r within the domain of r, say for r = 0.75, it can be seen that n(M2O3) = — 1 j, n(M) = +2j, n(Q0) = +0.5j and Q(OH)2 = +0.25j, where in each case, j = | n(M2O3) |. Before describing how other starting materials and products of the chemical process may be represented on such a graph to comply with the constraints of Eqns. 1d to 1 h, how the constraints of Eqns. 2a to 2f may also be represented graphically will be described next.

[0061] Fig. 2 is a similar type of graph to Fig. 1 , in that the ratio r = n(QC03) I n(M2O3) is again plotted on the r-axis or abscissa thereof. In this case, however, the y-axis or ordinate plots n(Y) I nCIVhOs), where Y represents a chemical species which is either consumed or produced by electrolytic or thermochemical decomposition during the chemical process. Thus chemical species which are decomposed into their constituent elements have negative y-values, whereas elements produced by such electrolytic or thermochemical decomposition have positive y-values. Note also that the scale on the y-axis in the graph of Fig. 2 is different from that on the y-axis in the graph of Fig. 1 . The constraints of Eqns. 2a to 2c are represented in Fig. 2 as follows. The constraint of Eqn. 2c defines the total requirement by the chemical process for liquid sodium, which may be produced by electrolysis of NaCI, by thermochemical decomposition of NaOH in accordance with UK patent application no. 2517327.9, or by a combination of both. This constraint is represented in the graph of Fig. 2 by the shaded region bounded from below by the horizontal line at y = +6. The constraint of Eqn. 2a, which represents how much of this liquid sodium may be produced by thermochemical decomposition of NaOH, is represented by the shaded triangle bounded from above by the r-axis and from below by the line y = 2r - 6. The constraint of Eqn. 2b, which represents how much of this liquid sodium may be produced by electrolysis of NaCI, is represented by the shaded triangle bounded from below by the line y = -6 and from above by the line y = — 2r. These two triangles therefore overlap in the region bounded from above by the line y = -2r and from below by the line y = 2r - 6, as indicated in Fig. 2. Thus it may be seen that at r = 3, for example, the total requirement for liquid sodium can only be met by electrolysis of NaCI and not by thermochemical decomposition of NaOH at all, whereas at lower values of r, a combination of both may be used. The amount of each, in other words, the respective values of n(NaOH)d I n(M2O3) and n(NaCI)eI n(M2O3) as a function of r, may therefore be represented in each case by a line, the left-hand end of which intercepts the y-axis somewhere between the values of y = 0 and -6 and the right-hand end of which lies at the apex of the respective shaded triangle at r = 3, like the dashed lines labelled di and d2 in the case of n(NaOH)d I n(M2O3) and the dashed lines labelled ei and in the case of n(NaCI)eI n(M2O3), which are only included in Fig. 2 by way of example.

[0062] However, because of the additional constraint imposed by Eqn. 2c, the respective values of n(NaOH)d / n(M20s) and n(NaCI)eI n(M2O3) are not independent of each other. In particular, as one of them is reduced, so the other must be increased to compensate and ensure that the constraint of Eqn. 2c is met for all values of r over the domain of r. This means, for example, that the different combinations of lines di and ei, di and d2 and all meet the additional constraint of Eqn. 2c, but that the combination of lines d2 and ei does not. Note also that whereas in Fig. 2, by way of example and for ease of explanation, all of the dashed lines are straight, they do not have to be, and may instead be curved or even non-differentiable at one or more values of r, provided that each of these lines lies within its respective shaded triangle in order to meet the constraints of Eqns. 2a and 2b, respectively, and that the constraint of Eqn. 2c is also met for all values of r over the domain of r, as just described.

[0063] Suppose, for example, that n(NaOH)d I n(M2O3) as a function of r is represented by the line di and that n (NaCI)eI n(M2C>3) as a function of r is represented by the line e^. Respective values may now therefore be assigned to n(NaOH)d I n(M2O3) and n (N aCI)eI n(M2O3) for any given value of r. For example, if r = 1 .5, say, it can be seen that n(NaOH)d = -1 .5j and n (N aCI)e= — 5j, where j = | n(M2O3) | in both cases. Now that respective values have been assigned to both n(NaOH)d and n(NaCI)e, the energy consumption of these two decompositions can also be calculated as +1 ,5j E(NaOH)d and +5j E(NaCI)e, respectively, where E(NaOH)d and E(NaCI)eare the constants described previously. (Note that calculating these values of energy consumption requires a change of sign because energy consumption is conventionally represented by positive values, whereas negative energy values conventionally represent its production.) Moreover, now that values have been assigned to each of n(NaOH)d I n(M20s) and n(NaCI)eI n(M2O3), these values may also be used to derive the corresponding values of n(Na)d I n(M20s) and n(Na)eI n(M20s) as functions of r from Eqns. 3a and 3b, respectively. The value of n(Cl2)eI n(M20s) as a function of r then also just follows from Eqn. 5 as n(Cl2)e= n(Na)e12. For the sake of example and using the same illustrative choice as was just supposed of using line to represent n(NaCI)eI n(M20s) as a function of r, the dashed line labelled (C je in the graph of Fig. 2 represents the value of n(Cl2)eI n(M20s) as a function of r which this choice of line implies. Having described how the constraints of Eqns. 2a to 2c may be represented graphically by Fig. 2, how the constraints of Eqns. 2d to 2f may be represented graphically will be described next. Fig. 3 is therefore the same type of graph as Fig. 2, in which r = n(QCC>3) I n(M2O3) is plotted on the abscissa and n(Y) I n(M2C>3) is plotted on the y-axis or ordinate, where Y again represents a chemical species which is either consumed or produced by electrolytic or thermochemical decomposition during the chemical process. The constraints of Eqns. 2d to 2f are represented in Fig. 3 as follows. The constraint of Eqn. 2f defines the total requirement by the chemical process for gaseous hydrogen to make hydrogen chloride by combustion with gaseous chlorine, so that the hydrogen chloride thus produced may be used to make hydrochloric acid in which to dissolve the carbonate mineral. This hydrogen may be produced by thermochemical decomposition of NaOH in accordance with UK patent application no. 2517327.9, by electrolysis of H2O, or by a combination of both. This constraint is represented in the graph of Fig. 3 by the shaded region bounded from below by the line y = +r. The constraint of Eqn. 2d, which represents how much of this hydrogen may be produced by thermochemical decomposition of NaOH, is again represented by the shaded triangle bounded from above by the r-axis and from below by the line y = 2r - 6, which follows from Eqn. 3c. The constraint of Eqn. 2e, which represents how much of the hydrogen may be produced by electrolysis of H2O, is represented by the shaded triangle bounded from above by the r-axis and from below by the line y = -r. These two triangles therefore overlap in the region bounded from above by the r-axis, from below-left by the line y = -r and from below-right by the line y = 2r - 6, as is indicated in Fig. 3. Thus it may be seen that at r = 3, for example, the total requirement for gaseous hydrogen can only be met by electrolysis of H2O and not by thermochemical decomposition of NaOH at all, whereas at lower values of r, a combination of both may be used.

[0064] However, the amount of hydrogen produced by thermochemical decomposition of NaOH is already determined by the choice of line in the graph of Fig. 2 because this hydrogen is produced as a co-product of the liquid sodium produced in the same way. Suppose, for example, that the line di has already been chosen in the graph of Fig. 2, as was supposed above. The same line di may therefore be reproduced in Fig. 3 as well, as shown therein. In order to meet the minimum value of the total requirement for hydrogen imposed by the constraint of Eqn. 2f, therefore, how much of this hydrogen may be produced by electrolysis of H2O must be at least the amount represented by the dashed line labelled 63 in Fig. 3. Accordingly, a value may now be assigned to n(H2O)eI n(M2C>3) for any given value of r, in addition to that already assigned to n(NaOH)d / n(M20s) previously, as described above in relation to Fig. 2. For example, if r = 0.5, say, it can be seen that n(H2O)eI n(M2O3) = 0, whereas for 1 < r < 3, n(NaOH)d / n(M2O3) = 3 (1 - r) / 2. The corresponding energy consumption when 1 < r < 3 can also be calculated as the positive value product of n(H2O)eE(H2O)e, where E(H2O)eis the constant described previously. Moreover, the total amount of hydrogen produced by the combination of these two techniques, n(H2)totai I n(M2Os) = [n(H2O)e+ n(NaOH)d] I n(M2O3), can now also be represented in Fig. 3 by the dashed line which intercepts the y-axis at y = +1 .5 and continues from there to y = +1 at r = 1 , after which it follows the line y = +r from there to r = 3.

[0065] Thus Fig. 2 plots n(Cl2)eI n(M2Os) as a function of r and Fig. 3 plots n(H2)totai I n(M2Os) as a function of r, based on values of n(NaOH)d I n(M2Os) and n(NaCI)eI n(M2Os) which were chosen initially, purely by way of example, to be represented by the dashed lines di and 62, respectively. n(Cl2)e / n(M2Os) and n(H2)totai I n(M2Os) may both therefore be transferred to a graph of the same type as Fig. 1 , instead representing the production of HCI by combusting gaseous hydrogen and chlorine with each other, as in Fig. 4. This is done just by reflecting the line representing n(Cl2)eI n(M2C>3) from Fig. 2 and the line representing n(H2)totai I n^Os) from Fig. 3 in the r-axis of Fig. 4, as is shown therein, to represent that these two elements are now being consumed rather than produced. From this, the total amount of hydrogen chloride gas produced, n(HCI) I n (M2O3), as well as the amount of excess chlorine remaining thereafter, n(Cl2)excess I n(M2C>3), both as a function of r, can each be determined from the equation for this reaction:

[0066] CI2 (g) + H2 (g) — > 2 HCI(g) [Eqn. 9]

[0067] As may be seen from Fig. 4, n(H2)totai I n(M2O3) is the limiting factor in determining n(HCI) I n(M2O3) according to Eqn. 9. However, from this, the value of n(HCI) I n(M2O3) as a function of r may be determined to be that represented in Fig. 4 by the dashed line which intercepts the y-axis at y = +3 and continues from there to y = +2 at r = 1, after which it follows the line y = +2r from there to r = 3. The value of n(Cl2)excess I n(M2O3) as a function of r is then just given by [n(H2)totai - n(Cl2)e] I nCIVhOa), as is also shown in Fig. 4. This excess CI2 production may be eliminated by imposing the additional constraint that n(Cl2)e= n(QC03), as described previously. In such a case, the values of n(NaOH)d I n (M2O3), n (N aCI)eI n(M2O3) and n(H2O)eI n(M2O3) chosen initially would be different. In particular, the additional constraint that n(Cl2)e= n(QC03) is met if n (N aCI)eis minimised, so that n (N aCI)e / n (M2O3) = — 2r, n(NaOH)d is maximised, so that n(NaOH)d I n(M2O3) = 2r - 6, and n(H2O)eI n(M2O3) = 0 when 0 < r < 1 .5 and n(H2O)eI n(M2O3) = 3 - 2r when 1 .5 < r < 3, as are shown in Fig. 5. If so, thermochemical decomposition of such an amount of NaOH as a function of r would also produce an excess amount of hydrogen over that required for producing sufficient HCI to make hydrochloric acid in which to dissolve the carbonate mineral as shown in Fig. 5 as well. However, as mentioned previously, it is preferable that the more relaxed additional constraint that 1 < n(Cl2)eI n(QC03) s 1.5 is applied instead.

[0068] In Fig. 4, the total requirement for HCI to make hydrochloric acid in which to dissolve the carbonate mineral is represented by the shaded region bounded from below by the line at y = +2r. Thus, in contrast to a situation in which the production of excess CI2 is eliminated by imposing the additional constraint that n(Cl2)eI n(QC03) = 1 , in the example situation which is represented in Fig. 4, not only is an excess amount of CI2 produced as shown in Fig. 4, but an excess amount of HCI is produced as well when 0 < r < 1. This excess amount of HCI is represented in Fig. 4 by the vertical height as a function of r of the triangle bounded from above by the dashed line which intercepts the y-axis at y = +3 and continues from there to y = +2 at r = 1 and bounded from below by the line y = +2r. Thus it may be seen that a graph like Fig. 4 can be used to determine the respective values not only of n(HCI) I n(M2O3) and n(Cl2)excess I n(M2O3) as a function of r, but also of n(HCI)eXcess I n(M2O3) as a function of r, for a given set of values previously assigned to n(NaOH)d I n (M2O3) , n (NaC l)eI n(M2O3) and n(H2O)eI n(M2O3).

[0069] Use of the graphical technique described above in relation to Figs. 1 to 5 may therefore be repeated to assign respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d as a function of r, subject to the constraints thereon described herein. The order in which these values are assigned may be chosen so that values are first assigned to a preferred one or more of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, in order to optimise the values of these preferred one(s) as desired, before values are then assigned to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d which are consistent with the constraints. Moreover, as shown above, when values are assigned to n(NaCI)e, n(H2O)eand n(NaOH)d, the respective energy consumption of each of these different types of electrolytic and thermochemical decomposition as a function of r can also be derived, which allows the total energy consumption of the chemical process to be optimised as well.

[0070] In Fig. 1 , the respective values of n(M2C>3), n(M) and n(QCC>3) and the aggregate value of n(Q0) + Q(OH)2, each as a function of r, are fully determined by the constraints of Eqns. 1 b and 1c and the definition of r. This means that the lines respectively representing each of these values as a function of r have fixed locations on the graph of Fig. 1. It was shown above in relation to Figs. 2 and 3, however, that lines respectively representing the values of n(NaCI)e, n(H2O)eand n(NaOH)d as a function of r can be put in different locations on such a graph as desired, within the constraints of Eqns. 2a to 2f, because these constraints all contain inequalities. Fig. 4 shows how the respective values of other ones of the n(X), apart from those shown in Fig. 1 , such as n(HCI) and n(CI2), then depend in part on the values assigned to n(NaCI)e, n(H2O)eand n(NaOH)d within the constraints of Eqns. 2a to 2f and in part on the remaining constraints of Eqns. 1d to 1 h. This means that lines respectively representing the values of these other n(X) as a function of r may also appear in different locations on a graph like Fig. 1 .

[0071] For example, the total amount of liquid sodium produced as shown in Fig. 2 is also consumed by the chemical process when it is used to produce iron and / or manganese in elemental form as a reductant of their respective oxides, in other words to produce the line labelled M in Fig. 1 . However, the same redox reaction also produces an amount of sodium oxide which is given by 2n(Na2O) + 4n(Na4SiC>4) = n(Na<4)). As described in the "Background of the Invention” section above, this sodium oxide may then be consumed by the chemical process in several different ways. As described therein, a second portion of it may be used to capture at least some of the C02which is released by the carbonate mineral when it dissolves in the hydrochloric acid, to produce sodium carbonate as a nett product of the chemical process. Suppose that it is desired to minimise the amount of C02which is produced by the chemical process as a nett product, in other words to minimise the value of n(C02) for a given value of r, subject to the same constraints. The amount of C02produced when the carbonate mineral dissolves in the hydrochloric acid is just given by n(C02) produced = n (QCO3) consumed. However, since the value of n (QCO3) as a function of r is fixed, this means that the value of n(C02) produced when the carbonate mineral dissolves in the hydrochloric acid as a function of r is fixed as well. Thus a line representing the amount of this C02which can be consumed to produce sodium carbonate may be plotted on a graph like Fig. 1 at y = -r, in the same position as the corresponding line for QCO3, as shown in Fig. 6.

[0072] Also suppose, for simplicity, that n(Na4SiC>4) is assigned the value 0 over the domain of r, so that n(Na20) = n(Na<4)) / 2 as a function of r. A line representing this total amount of sodium oxide available for consumption may therefore also be plotted on the graph of Fig. 6 at y = -3, as shown. However, as described in the "Background of the Invention” section, a first portion of this sodium oxide is used to produce an aqueous solution of sodium chloride. Since n(Na20) is constant, this reduces the amount of sodium oxide left over for the second portion of sodium oxide. The size of the aqueous solution of sodium chloride produced by the first portion of sodium oxide as a function of r is given by n(NaCI) produced = 2n (QCO3) consumed. Since this aqueous solution of sodium chloride may be recycled by being dried and then fused and electrolysed, it is not a nett product of the chemical process. The amount of this sodium chloride produced, in other words n(NaCI) as a function of r, is therefore represented in Fig. 6 by the dashed line at y = +2r. The size of the second portion of sodium oxide remaining as a function of r is then just given by n(Na20) - n(NaCI) 1 2, which is represented in Fig. 6 by the dashed line at y = r - 3.

[0073] Since it is desired to minimise the amount of CO2 produced by the chemical process as a nett product, the amount of sodium carbonate produced by the second portion of sodium oxide mirrors the amount of CO2 produced when the carbonate mineral dissolves in the hydrochloric acid when 0 < r < 1 .5. In other words, n (Na2CC>3) produced = n(CC>2) consumed, both as a function of r over these values of r, as shown in Fig. 6. Over the same values of r, the amount remaining from the second portion of sodium oxide which can be recycled by being hydrated and then subjected to thermochemical decomposition diminishes as both the size of the second portion of sodium oxide reduces and the amount of sodium carbonate produced increases. In other words, n(NaOH) available for thermochemical decomposition = 2 [- n(2nd portion of Na20) - n(Na2CC>3)] as a function of r when 0 < r < 1.5. This is represented in Fig. 6 by the dashed line at y = 6 - 4r. However, when r = 1.5, n(CC>2) = n(2nd portion of Na20) at y = -1 .5, so that all of the second portion of sodium oxide is consumed in producing sodium carbonate and n(NaOH) available for thermochemical decomposition = 0. Moreover, when r > 1.5, | n(2nd portion of Na2O) | continues to decrease as n(NaCI) continues to increase. This means not only that n(Na2CC>3) starts to decrease as | n(2nd portion of Na2O) | gets smaller, but that CO2 also starts to be produced as a nett product of the chemical process, as the amount of CO2 produced when the carbonate mineral dissolves in the hydrochloric acid continues to grow. This nett production of CO2 is represented in Fig. 6 by the line at y = 2r - 3 when 1 .5 < r < 3, which also follows from the constraint of Eqn. 1 d. Nonetheless, even at a high value of r, like r = 2.5 for example, the chemical process still produces a third less CO2 than traditional calcination of QCO3 according to Eqn. 4.

[0074] Having now minimised the value of n(CC>2) as a function of r as shown in Fig. 6, values can be assigned to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, as will now be described in relation to Fig. 7. This can be done by supposing, for example, that it is next desired to minimise the consumption of salt by the chemical process, which can be done by maximising the values of n(NaCI) recycled for electrolysis and n(NaOH) recycled for thermochemical decomposition. Accordingly, the dashed lines from Fig. 6 respectively representing n(NaCI) and n(NaOH) available for recycling, each as a function of r, can be converted into lines respectively representing n(NaCI)recycied and n(NaOH)d in Fig. 7 by reflecting them in the r-axis, as shown in Fig. 7. (Note that the scale on the y-axis in the graph of Fig. 6 is different from that on the y-axis in the graph of Fig. 6.) However, the constraint of Eqn. 2c requires that n (N aCI)e+ n(NaOH)d > 6 over the domain of r. Thus a nett amount of NaCI may also need to be consumed by the chemical process as a starting material, depending on the value of r. This is represented in Fig. 7 by the line at y = -2r when 0 < r < 1 .5 and y = 2r - 6 when 1 .5 < r < 3. Since n (Na2CC>3) as a function of r was already determined above in relation to Fig. 6, the constraint of Eqn. 1e shows that with these assumptions, no nett amount of Na2<3 or NaOH is produced by the chemical process as well. In other words, nett consumption of NaCI is balanced by nett production of Na2CC>3 over the domain of r. n(NaCI)eas a function of r is then just given by n(NaCI)reCycied + n(NaCI), as shown in Fig. 7.

[0075] As was seen above in relation to Fig. 4, the minimum amount of HCI required by the chemical process to make hydrochloric acid in which to dissolve the carbonate mineral is given by a line at y = +2r. Since the nett consumption of NaCI is now also known, the constraint of Eqn. 1f allows a value to be assigned to n(Cl2) as a function of r as well, as shown in Fig. 7. It can also be seen from the line representing n(NaCI)ein Fig. 7 that the total amount of chlorine gas produced by electrolysis always exceeds the amount of chlorine needed to make the HCI, over the domain of r. However, since | n(NaOH)d | diminishes from 6 to 0 as r goes from 0 to 1 .5, the amount of hydrogen gas produced by thermochemical decomposition of NaOH reduces in proportion thereto, even as the requirement for hydrogen to make the HCI increases, until when r = 1 , not enough hydrogen is produced by thermochemical decomposition of NaOH to make the HCI by combustion with chlorine. At r > 1 , therefore, hydrogen to make the HCI may also be produced by electrolysis of water. The minimum requirement for hydrogen at y = +r imposed by the constraint of Eqn. 2f can therefore now be used to assign a value to n(H2O)eas a function of r as well, which is represented in Fig. 7 by the dashed line at y = 3 - 3r when 1 < r < 1 .5 and at y = -r when 1 .5 < r < 3. (n(H2O)e= n(NaOH)d in this example when r = 9 / 7.) The respective energy consumption of each of n(NaOH)d, n(NaCI)eand n(H2O)eas a function of r can therefore now also be calculated, as described previously.

[0076] It only remains to assign values to n(H2), n(H2O) and n(O2) as a function of r, using the remaining constraints of Eqns. 1g and 1 h. When r < 1 , thermochemical decomposition of NaOH produces an amount of hydrogen in excess of the minimum amount required to make HCI by combustion with chlorine. Since the minimum amount of hydrogen required to make HCI as a function of r is y = +r, it follows from Eqn. 3c that the nett amount of hydrogen remaining thereafter, n(H2) = [-n(NaOH)d 1 2] - r = - (4r - 6) 12 - r, which is represented in Fig. 7 by the line at y = 3 - 3r. As may be seen, this line just continues the dashed line representing n(H2O)ewhen r < 1. When r > 1 , all of the hydrogen produced either by thermochemical decomposition of NaOH or electrolysis of water is used to make HCI, so n(H2) = 0 when r > 1 . The constraint of Eqn. 1g may now therefore be used to determine n(H2O). Suppose, for simplicity, that n(Q(OH)2) is assigned the value 0 over the domain of r. It is already know that n(HCI) = 0 over the domain of r because no nett excess of HCI is produced, and that n(NaOH) = 0 as well over the domain of r. Eqn. 1g therefore implies that n(H2) produced = n(H2O) consumed. This allows a line representing nett consumption of H2O to be plotted on Fig. 7 as the mirror image of n(H2) in the r-axis, at y = 3r - 3 when r < 1 and at y = 0 when r > 1. In other words, when r > 1 , all of the water consumed by electrolysis is recovered by drying Q(OH)2 to make QO and is then reused to make HCI.

[0077] The constraint of Eqn. 1 h also reduces to n(Na2CO3) + 2n(O2) = 3n(M2O3) + n(H2O). However, values have now been assigned to the respective no. of mols of all the species in this equation, apart from n(O2). This therefore allows n(O2) to be calculated as a function of r as well. In particular, it is known from Fig. 1 that n(M2O3) = -1 , n(Na2CO3) as a function of r is as shown in Fig. 6, and n(H2O) as a function of r is as shown in Fig. 7. It follows that n(02) is assigned the value y = 3 - 2r when r < 1 and y = +r / 2 when r > 1 , as is also shown in Fig. 7. In a possible variant of the illustrated embodiment, if it were also desired to minimise the consumption of water by the chemical process, then equimolar amounts of hydrogen and oxygen both produced by thermochemical decomposition of NaOH when r < 1 could be combusted together to reduce the nett consumption of water to zero over the domain of r, in which case, n(C>2) would just be assigned the value y = +r / 2 over the whole domain of r as well. However, this would be a trade-off between the material cost of the water consumed, the value of the hydrogen otherwise produced and the nett cost of heat for the thermochemical decomposition less the waste heat recovered from that combustion, which are respectively represented by p(H20), p(H2) and p(E)r. Thus it may be seen that extrinsic parameters like these may be used to help determine what values to assign to respective ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, including the value of r.

[0078] Fig. 8 shows a first embodiment of a method 100a of operating a chemical plant. The method 100a comprises assigning 103a respective values to each n(X) subject to the constraints on the n(X) described herein and assigning 103b respective values to n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints thereon also described herein. In general, assigning 103a the respective values to each n(X) and assigning 103b the respective values to n(NaCI)e, n(H2O)eand n(NaOH)d may be conducted in any sequence or temporal order or may overlap in time, provided that a full set of values which is consistent with all the constraints thereon is produced as a result. The method 100a then comprises operating 104 the chemical plant according to the full set of values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103a, 103b.

[0079] Fig. 9 shows a second embodiment of a method 100b of operating a chemical plant. The method 100b comprises all the same features as the method 100a just described, except that in the present embodiment, the temporal order in which the respective values are assigned subject to the constraints thereon is now defined as follows. Firstly, values are assigned 103s to a preferred subset of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints thereon, then the same constraints are used to assign 103r values to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d. The values assigned 103s to the preferred subset may be chosen as desired within the constraints thereon, for example to maximise or minimise the preferred subset of values, after which the values assigned 103r to remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d are at least partially dependent on the values already assigned 103s to the preferred subset as a result of the constraints on them. Thus the values assigned 103s to the preferred subset are given priority when assigning the full set of values. This in turn allows optimisation of the chemical process according to the values assigned 103s to the preferred subset. This process may be repeated more than once, so that values are firstly assigned to a preferred subset, then values are assigned to a next preferred subset, and so on, until all the values have been assigned, subject to the constraints. The method 100b then comprises operating 104 the chemical plant as in as the method 100a according to the full set of values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103s, 103r.

[0080] Fig. 10 shows a third embodiment of a method 100c of operating a chemical plant. The method 100c is a special case of the method 100b because the subset of values which are first assigned 103s is the value of n(X) when X = CO2. Moreover, the value which is assigned 103c to n(CC>2) is the minimum value of n(CC>2) as a function of r which is allowed by the constraints. The same constraints are then used to assign 103r values to the remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, as before, and the chemical plant is operated 104 according to the full set of values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103c, 103r. This therefore ensures that production of CO2 by the chemical plant is minimised, whatever the values are which are assigned to the remaining ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d.

[0081] Fig. 11 shows a fourth embodiment of a method 100d of operating a chemical plant. The method 100d comprises all the same features as the method 100a described above in relation to Fig. 8, except that before assigning 103 values to respective ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints thereon, an additional constraint that 1 < n(Cl2)eI n(QCC>3) < 1.5 is applied 105. This additional constraint ensures that the values assigned 103 to respective ones of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d cause the chemical process to produce sufficient chlorine by electrolysis to make enough hydrochloric acid in which to dissolve the carbonate mineral, and only to produce a small or no nett amount of gaseous chlorine, hydrogen chloride or hydrochloric acid in excess of that, whereby overproduction of these chlorinated species is avoided. The chemical plant is then operated 104 according to the full set of values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d assigned 103 according to this enlarged set of constraints.

[0082] Fig. 12 shows a fifth embodiment of a method 100e of operating a chemical plant. The method 100e comprises assigning 103e respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints on them described herein, for which the total value of (E)in- (E)out as a function of r lies within 20% of its minimum value. The total value of (E)in- (E)out may be calculated as described previously above. The method 100e then comprises operating 104 the chemical plant according to the values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103e, in other words, in a way which is close to the most energy efficient.

[0083] Fig. 13 shows a sixth embodiment of a method 100f of operating a chemical plant. The method 10Of comprises assigning 103t respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints on them described herein, except that in this embodiment, one or more of the values assigned 103t to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d changes over time. How each of the values changes over time may be preprogrammed, for example, in order to meet planned delivery schedules of starting materials to and / or products from the chemical plant, or may be in response to changes in input data, for example data concerning the performance of the chemical plant and / or data extrinsic to the chemical plant. The method 10Of then comprises operating 104t the chemical plant according to the values of n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103t as one or more of them changes over time.

[0084] Fig. 14A shows a seventh embodiment of a method 100g of operating a chemical plant. The method 100g comprises receiving 101 respective values of p(X), p(E)eand P(E)T, receiving 102 a value of E(WHR) and assigning 103p respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d, subject to the constraints thereon, for which the total value of p lies within 20% of its maximum value, as described previously above. The method 100g then comprises operating 104 the chemical plant according to the values of each of the n(X), n(NaCI)e, n(H2O)eand n(NaOH)d thus assigned 103p, as before.

[0085] Fig. 14B shows an eighth embodiment of a method 10Oh of operating a chemical plant. The method 10Oh is a special case of the method 100g because the method 10Oh comprises all the features of the method 100g. In the present embodiment, however, at least one of the respective values of each p(X), p(E)eand P(E)T which are received 101 a are dynamically variable over time. In alternative possible variants of the method 10Oh, the value of E(WHR) received 102 may or may not be dynamically variable over time as well. The method 100h then comprises assigning 103q respective values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints on them described herein, wherein one or more of the values thus assigned 103q changes over time. In this respect, therefore, the method 10Oh is also a special case of the method 10Of described above in relation to Fig. 13. In the present embodiment, however, the values of n(X), n(NaCI)e, n(H2O)eand / or n(NaOH)d which change over time change in response to how the respective values of p(X), p(E)eand / or p(E)r change over time as well, in order to maintain maximisation of the total value of p.

[0086] Fig. 15 schematically shows an embodiment of a data processing system 10 for carrying out any of the methods described herein. The data processing system 10 comprises a microprocessor 12, a memory 14 and a graphical user interface (GUI) 16. The microprocessor 12 is in two-way communication with both the memory 14 and the GUI 16. The microprocessor 12 has a first input channel h for receiving data from the chemical plant, such as readings from temperature and pressure gauges, flow meters and so on, whereby operation of the chemical plant can be monitored by the microprocessor 12 and the values of empirical quantities which depend thereon, such as E(WHR), can be determined. The microprocessor 12 also has a second input channel I2 for receiving data extrinsic to the chemical plant, such as the respective values of at least one of one or more p(X), p(E)eand P(E)T. The second input channel I2 may typically be implemented via an internet connection, for example. The microprocessor 12 also has an output channel O for transmitting control signals to actuators of the chemical plant, such as pumps, switches, valves and so on, whereby operation of the chemical plant can be controlled by the microprocessor 12, according to the respective values assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d.

[0087] The memory 14 is for storing such data as the layout and scale of the chemical plant, the respective values of such constants as E(NaCI)e, E(H2O)eand E(NaOH)d, as well as the sets of constraints described herein. The memory 14 may also be able to log expected and / or past performance of the chemical plant under different conditions. The GU1 16 is for two-way communication with a user of the system. It may therefore display to the user such things as the values of data received via the first and second input channels 11, 12 and one or more items of data stored in the memory 14, such as the constraints on the values of each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d. In particular, the GUI 16 may display these constraints to the user using the graphical technique described above. The GUI 16 may also receive from the user such things as user preferences, which may include a subset of the respective values to assign to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d and / or a preferred measure on which to optimise performance of the chemical plant, such as to minimise its energy consumption, for example. However, the GU1 16 may also be able to receive other types of data from the user as well. These may be empirical data, like the proportion of M2O3 in ore 1 and / or the proportion of QCO3 in ore 2. Alternatively or additionally, the GUI 16 may be able to receive manual entry of data by the user instead of via the second input channel I2, like the respective values of one or more of p(X), p(E)eand P(E)T.

[0088] During operation, therefore, the microprocessor 12 receives data via the first and second input channels h, I2 and via the GUI 16, and retrieves data from the memory 14. The microprocessor 12 checks whether any value(s) entered by the user via the GUI 16 conflict with the constraints retrieved from the memory 14, and if so, either rejects the conflicting value(s) by notifying the user or modifies the conflicting value(s) and informs the user accordingly, in each case via the GU1 16. The microprocessor 12 then assigns a full set of consistent values to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints described herein, and uses the values thus assigned to control operation of the chemical plant via the output channel 0. Optionally, the microprocessor 12 may also record the values thus assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d in memory 14 as the respective values of a set of independent variables, against which the performance of the chemical plant can then also be logged in memory 14 as a set of data received via the first input channel h which depend thereon. Moreover, the microprocessor 12 may operate dynamically, so that as it receives new data via the first and / or second input channels h, 12 and / or via the GU1 16, it updates the values assigned to each n(X), n(NaCI)e, n(H2O)eand n(NaOH)d subject to the constraints thereon.

[0089] Whereas the present invention has been described above by reference to particular examples and embodiments, the scope of the invention should not be taken to be limited thereby and is instead defined by the appended claims.

[0090] The following page of description gives definitions of each of the algebraic expressions used herein.

[0091] Definitions:

[0092] M = Fe and / or Mn

[0093] Q = at least one of Ca, Mn and Fe ore 1 = a substance comprising an oxide of M (typically an ore, but may also be a waste product) ore 2 = an ore comprising a carbonate mineral of Q

[0094] X = a chemical species being any one of M2O3, QCO3, NaCI, H2O, M, Q(OH)2, QO, Na2CC>3, CO2, Na2O, NaOH, Na4SiC>4, CI2, HCI, H2 and O2 m(X) = mass of chemical species X n(X) = no. of mol of chemical species X r = n(QC03) I n(M2O3) n(Na)d = no. of mol of liquid sodium produced by decomposing solid-phase NaOH into its constituent elements according to UK patent application no. 2517327.9 n(Na)e= no. of mol of liquid sodium produced by electrolysis of NaCI n(Cl2)e= no. of mol of gaseous chlorine produced by electrolysis of NaCI n(H2)d = no. of mol of gaseous hydrogen produced by decomposing solid-phase NaOH into its constituent elements according to UK patent application no. 2517327.9 n(H2)e= no. of mol of gaseous hydrogen produced by electrolysis of H2O n(02)d = no. of mol of gaseous oxygen produced by decomposing solid-phase NaOH into its constituent elements according to UK patent application no. 2517327.9 n(02)e= no. of mol of gaseous oxygen produced by electrolysis of H2O n(NaOH)d = no. of mol of solid-phase NaOH consumed by being decomposed into its constituent elements according to UK patent application no. 2517327.9 n(NaCI)e= no. of mol of NaCI consumed by being electrolysed n(H2O)e= no. of mol of H2O consumed by being electrolysed

[0095] E(NaOH)d = no. of kJ mol1(NaOH) of thermal energy consumed by decomposing solid-phase NaOH into its constituent elements according to UK patent application no. 2517327.9

[0096] E(NaCI)e= no. of kJ mol1(NaCI) of electrical energy consumed by electrolysis of NaCI E(H20)e= no. of kJ mol1(H2O) of electrical energy consumed by electrolysis of H2O

[0097] E(WHR) = total no. of kJ of usable heat recovered from exothermic reactions in the chemical process p(X) = a parameter of chemical species X per unit mass thereof p(E)e= the same parameter of electrical energy per kJ thereof

[0098] P(E)T = the same parameter of thermal energy per kJ thereof

[0099] (p(E)e and p(E)i are therefore interconvertible by a dimensionless number)

Claims

AMENDED CLAIMSreceived by the International Bureau on 8 June 2026 (08.06.2026)1. A method (100a - 100h) comprising:operating a chemical plant arranged to carry out a chemical process which produces iron or manganese in elemental form from respective oxides thereof using liquid sodium as a reductant of the respective oxides and which can produce an oxide or hydroxide of at least one of calcium, magnesium and iron from a carbonate mineral of at least one of calcium, magnesium and iron, wherein the process involves a plurality of chemical species (X) as starting materials and products thereof, and the method comprises assigning (103a) a respective value to each n(X), where n(X) is the no. of mol of chemical species X, subject to the following constraints:0 < n(QCO3) I n(M2O3) < 3n(M) = 2n(M2O3)n(Q(OH)2) + n(QO) = n(QCO3)n(CO2) + n(Na2CO3) = n(QCO3)2n(Na2O) + n(NaOH) +2n(Na2CO3) + 4n(Na4SIO4) = n(NaCI)2n(CI2) + n(HCI) = n(NaCI)n(H2) + n(Q(OH)2) + [n(HCI) + n(NaOH)] 12 = n(H2O)n(Na2O) + n(NaOH) + n(Na2CO3) + 2n(O2) = 3n(M2O3) + n(H2O)wherein M represents at least one of Fe and Mn, Q represents at least one of Ca, Mg and Fe, the lefthand side of each equation represents nett production by the chemical process and the right-hand side of each equation represents nett consumption by the chemical process of each respective one of the chemical species (X) identified therein;assigning (103b) a respective value to each of n(NaCI)0, n(H2O)eand n(NaOH)d subject to the following constraints:0 < n(Na)d / n(M2O3) < 6 - 2r2r < n(Na)e / n(M2O3) < 6[n(Na)d+ n(Na)e] I n(M2O3) > 60 < n(H2)d / n(M2O3) < 3 - r0 < n(H2)e / n(M2O3) < r[n(H2)d+ n(H2)e] / n(M2O3) > rwherein:n(Na)d= n(NaOH)dn(Na)e= n(NaCI)en(H2)d= n(NaOH)d / 2n(H2)e= n(H2O)eand r represents the value assigned to the ratio n(QCO3) I n(M2O3); andoperating (104) the chemical plant to carry out the process according to the values of n(X), n(NaCI)0, n(H2O)eand n(NaOH)d thus assigned.

272. A method (100b, 100c) according to claim 1, comprising assigning (103c, 103s) respective values to a preferred subset of each n(X), n(NaCI)0, n(H2O)eand n(NaOH)d, subject to the same constraints, before using the same constraints to assign (103r) respective values to remaining ones of the n(X), n(NaCI)0, n(H2O)eand n(NaOH)d.

3. A method (100c) according to claim 2, comprising minimising (103c) the value of n(CC>2) for a given value of r, subject to the same constraints, before using the same constraints to assign (103r) respective values to remaining ones of the n(X), n(NaCI)0, n(H2O)eand n(NaOH)d.

4. A method (1 OOd) according to any one of the preceding claims, comprising applying (105) the additional constraint that 1 < n(Cl2)e / n(QCO3) 1.5.

5. A method (100e) according to any one of the preceding claims, comprising assigning (103e) the respective values to each n(X), n(NaCI)0, n(H2O)eand n(NaOH)d, subject to the same constraints, for which the total value of (E)in- (E)out for a given value of r lies within 20% of its minimum value, wherein (E)inrepresents energy consumed by the chemical process and (E)out represents energy recovered from the chemical process.

6. A method (1 OOf, 10Oh) according to any one of the preceding claims, wherein at least one of the values assigned (103a, 103b, 103c, 103e, 103r, 103s) to each n(X), n(NaCI)0, n(H2O)eand n(NaOH)d are varied (103t, 103q) over time subject to the same constraints.

7. A method (100g, 100h) according to any one of the preceding claims, comprising assigning (103p, 103q) the respective values to each n(X), n(NaCI)0, n(H2O)eand n(NaOH)d, subject to the same constraints, for which the total value of a parameter (p) given by:Z p(X)0Utm(X)0Ut + Z p(E)out (E)out - {Z p(X)inm(X)in+ Z p(E)in(E)in)lies within 20% of its maximum value, wherein (X)inand (E)inrespectively represent chemical species and energy consumed by the chemical process, and (X)out and (E)out respectively represent chemical species produced by and energy recovered from the chemical process, whereby the total value of p =p(M) m(M) + p(Q(OH)2) m(Q(0H)2) + p(QO) m(Q0) + p(Na2CO3) m(Na2CO3) +p(CO2) m(C02) + p(Na2O) m(Na20) + p(NaOH) m(NaOH) + p(Na4SIO4) m(Na4SIO4) +p(CI2) m(CI2) + p(HCI) m(HCI) + p(H2) m(H2) + p(O2) m(O2) + p(E)TE(WHR) - {p(ore 1) m(ore 1) + p(ore 2) m(ore 2) + p(NaCI) m(NaCI) + p(H3O) m(H20) +p(E)0E(NaCI)en(NaCI)e+ p(E)0E(H2O)en(H2O)e+ p(E)TE(NaOH)dn(NaOH)d}wherein:m(ore 1) = m(M2O3) I (proportion of M2O3in ore 1),m(ore 2) = m(QC03) I (proportion of QCO3in ore 2), andthe values of each p(X), p(E)0and P(E)T are received (101, 101a) as inputs.

8. A method (1 OOh) according to claim 7, wherein at least one of the values of each p(X), p(E)0and P(E)T received (101a) as inputs are dynamically variable over time, and at least one of the values assigned (103q) to each n(X), n(NaCI)0, n(H2O)eand n(NaOH)d are varied over time in response thereto, to maintain maximisation of the total value of p over time.

9. A data processing system (10) comprising means (12, 14, 16) for carrying out the method of any one of claims 1 to 8.

10. A computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the method of any one of claims 1 to 8.

11. A computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any one of claims 1 to 8.STATEMENT UNDER ARTICLE 19 (1)Claim 1 on file has been amended to correct a typographical error at page 22, line 18, by replacing “Q represents at least one of Ca, Mn and Fe” with “Q represents at least one of Ca, Mg and Fe”.This amendment has the following impact on the description:At page 4, line 9: “Q represents at least one of Ca, Mn and Fe” should be replaced by “Q represents at least one of Ca, Mg and Fe”.At page 21, line 3: “Q = at least one of Ca, Mn and Fe” should be replaced by “Q = at least one of Ca, Mg and Fe”.

Images