AU3947300A - Method for causticisation of alkaline solutions - Google Patents

Description

WO 00/64812 PCT/AU00/00346 -1 METHOD FOR CAUSTICISATION OF ALKALINE SOLUTIONS FIELD OF THE INVENTION The present invention relates to a method for the causticisation of alkaline solutions. More particularly, the method of the present invention relates to a 5 method for the causticisation of alkaline solutions at elevated temperatures. BACKGROUND ART The application of the present invention will be discussed in the context of the causticisation of Bayer process liquors, although the scope of the invention encompasses the causticisation of any alkaline solution, including the black kraft 10 liquors of the pulp and paper industry. The Bayer process is widely used for the production of alumina from alumina containing ores, such as bauxite. The process involves contacting alumina containing ores with recycled caustic aluminate solutions, at elevated temperatures, in a process commonly referred to as digestion. Solids are 15 removed from the resulting slurry, and the solution cooled. Aluminium hydroxide is added to the solution as seed to induce the precipitation of further aluminium hydroxide therefrom. The precipitated aluminium hydroxide is separated from the caustic aluminate solution, with a portion of the aluminium hydroxide being recycled to be used as seed and the remainder recovered as 20 product. The remaining caustic aluminate solution is recycled for further digestion of alumina containing ore. As aluminium hydroxide is precipitated and bauxite dissolved, the concentrations of sodium hydroxide present in the process solution decrease, whilst concentrations of sodium carbonate increase, reducing the efficacy of the solution 25 for digestion of further aluminium-containing ore. Accordingly, processes aimed at improving the ratio of hydroxide to carbonate (causticising) Bayer liquors have been developed.

WO 00/64812 PCT/AU00/00346 -2 Conventional causticisation techniques are seemingly limited by an upper threshold. Attempts to causticise a Bayer liquor beyond a maximum TC/TA ratio (where TA represents total alkali concentration and TC represents total caustic concentration, both expressed as gL 1 sodium carbonate) result in the formation of 5 undesirable tricalcium aluminate hexahydrate, and poor efficiency in the utilisation of lime. For a typical Western Australian Bayer liquor with a TA of about 250 gL 1 under digest conditions, conventional techniques generate a maximum TC/TA of about 0.81, with tricalcium aluminate as the stable lime phase. The tricalcium aluminate by-product of conventional causticisation contributes to the alkalinity of 10 red mud waste from the Bayer process, and to the environmental difficulties associated therewith. Further, conventional causticisation requires reaction times of the order of one hour, and any reductions in such afford opportunities for greater plant efficiency. United States Patent 3210155, granted to Cagnolatti et al. discusses a process 15 that is essentially outside causticisation of the liquor used to wash red mud on the filter presses, prior to its disposal. The novelty of the Cagnolatti process resides in the fact that the causticised liquor is used to wash the red mud prior to it being pumped into the second washer. The process uses extremely dilute liquors, with TA about 15 to 20 gL 1 to achieve high resultant TC/TA ratios and high lime 20 efficiencies. The temperatures used by Cagnolatti are below 1050C, and no regard is had to the speed at which this reaction temperature is attained. Dry lime is used, and residence times approximate those of conventional outside causticisation, being 1 to 1 hours. In accordance with conventional theory, the efficacy of this process is substantially reduced as the TA of the feed liquor is 25 increased. United States Patent 4486393, granted to Baska et al. causticises red mud from washer underflow to recover sodium hydroxide bound therein. Similarly to Cagnolatti, the high resultant TC/TA ratios of the Baska process are dependent on the use of extremely low TA liquors, with the liquor entrained in the mud being 19 30 to 22 gL -1 . The temperatures used by Baska are similar to those of Cagnolotti, not exceeding 1000C, with no regard being had to the speed at which the reaction temperature is reached. Long residence times, up to 5 hours, are advocated. In WO 00/64812 PCT/AU00/00346 -3 accordance with conventional theory, the efficacy of the Baska process is also substantially reduced as the TA of the feed liquor is increased. United States Patent 2522605, issued to W. H. Cundiff, discusses the causticisation of diluted spent Bayer process liquors by the addition of lime. The 5 use of diluted liquors in this manner is undesirable as it results in an overall dilution of the plant liquor stream, requiring further evaporation, which is time and cost intensive. Cundiffs reactions take place in a preferred temperature range of 250 0 F to 290 0 F (1150C to 1430C) (column 6, line 33), although no regard is had to the speed at which this reaction temperature is attained. 10 According to Cundiff, spent liquors are used because 'the equilibrium.., is shifted by the presence of sodium aluminate so as to further lower the proportion of soda ash causticised' (column 2, lines 19 to 22). Additionally, Cundiff notes that 'the reaction time does not materially effect the causticising efficiencies in terms of lime utilisation and degree of conversion to soda and therefore is not critical' 15 (column 6, lines 43 to 46). In a recent review of the field, Whittington (Whittington, BI (1996), 'The Chemistry of CaO and Ca(OH) 2 relating to the Bayer Process', Hydrometallurgy, 43, 13-35) notes the relationship between elevated temperatures and lime efficiencies, stating that "predictions that...increasing reaction temperatures 20 thermodynamically favour CaCO 3 formation have been experimentally verified". As support for his conclusion, Whittington refers to several studies investigating the relationship between temperature and lime efficiency. Most notably, Solymar and Zoldi (Solymar, K and Zoldi, J (1993), Light Metals, 185-194) report causticisation experiments at temperatures ranging from 120 to 25 2400C, and demonstrate that TC/TA increases with reaction temperature. However, whilst the reaction temperatures studied exceed those used in conventional causticisation (about 1000C), the quoted TC/TA ratios at 1600C do not exceed those obtained under conventional conditions.

WO 00/64812 PCT/AU00/00346 -4 Further, Xu (Xu B A (1991), "Lime Chemistry in the Bayer Process", PhD Thesis, Murdoch University, and also XU, B A, Giles D E and Ritchie I M (1998), Hydrometallurgy, 48, 205-224) conducted experiments ranging from 25 to 750C. TC/TA ratios calculated from these results do not exceed those obtainable by 5 conventional caustisation. International Patent Application WO 00/18684 (Worsley Alumina Pty Ltd) discloses a method for causticisation of Bayer liquors by first forming hydrocalumite via the reaction of slaked lime with pre-causticised liquor at relatively low temperatures. The hydrocalumite is then reacted with uncausticised 10 liquor at higher temperatures to form calcium carbonate. Notably, it is stated that "the direct reaction of calcium hydroxide with the carbonate ion as described in equation 1 does not occur to any appreciable extent" (page 10, lines 4-5), where equation 1 is as follows: Ca(OH) 2 + Na2CO 3 -- CaCO 3 + 2NaOH 15 Therefore, the higher temperatures are selected for the second stage simply to increase the rate of the reaction of the hydrocalumite to calcium carbonate (page 15, lines 1 to 5), and thus to reduce capital costs. Notably, significant improvements over the total caustic to total alkali ratio (CIS) attainable by existing methods are only achieved with the introduction of additives, 20 some of which would themselves produce undesirable effects. Further, the method disclosed by the Worsley patent must be performed in a two-step process, significantly increasing capital costs. It is one object of the present invention to at least partially overcome the abovementioned problems associated with the prior art, or provide an alternative 25 thereto. Throughout the specification, unless the context requires otherwise, the word "comprise" or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.

WO 00/64812 PCT/AU00/00346 -5 Throughout the specification, unless the context requires otherwise, the word "solution" or variations such as "solutions", will be understood to encompass slurries, suspensions and other mixtures containing undissolved solids. DISCLOSURE OF THE INVENTION 5 In accordance with the present invention there is provided a method for the causticisation of an alkaline solution, the method comprising the step of: adding a quantity of lime to the alkaline solution to produce a reaction mixture in such a manner that the reaction temperature of the reaction mixture so produced rapidly exceeds the atmospheric boiling point of the 10 alkaline solution. Preferably the ratio of total caustic concentration to total alkali concentration of the alkaline solution, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.82, after addition of the lime. In a further form of the invention, the ratio of total caustic concentration to total 15 alkali concentration, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.85. In a preferred form of the invention, the ratio of total caustic concentration to total alkali concentration, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.87. In a further preferred form of the invention, the ratio of total caustic concentration to 20 total alkali concentration, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.90. In one form of the invention, both the quantity of lime and the alkaline solution are maintained about the reaction temperature before the quantity of lime is added to the alkaline solution. 25 In an alternate form of the invention, the quantity of lime is maintained at a temperature below the reaction temperature before the quantity of lime is added to the alkaline solution, the reaction mixture so produced being rapidly heated to WO 00/64812 PCT/AU00/00346 -6 the reaction temperature. Preferably, the reaction mixture reaches the reaction temperature in less than 10 minutes. Preferably still, the reaction mixture reaches the reaction temperature in less than 5 minutes. The quantity of lime is preferably provided in the form of slaked lime. In a 5 preferred form of the invention, the slaked lime is provided in a highly reactive form, such as slaked lime putty. In a highly preferred form of the invention, the slaked lime putty has a low particle size and a high solids density. In one form of the invention, the slaked lime has particle size d 5 o = 8 - 10 pm, solids density > 400 gL -1 . 10 Preferably, the reaction temperature is above 1100C. Preferably still, the reaction temperature is between about 1100C and 3000C. In a highly preferred form of the invention, the reaction temperature is between about 1100C and 1700C. The preferred residence time of the quantity of lime in the alkaline solution is dependent on the reaction temperature, with higher temperatures corresponding 15 to shorter residence times. Preferably, the residence time of the quantity of lime in the alkaline solution is less than 15 minutes. Preferably still, the residence time is less than 10 minutes. In a highly preferred form of the invention, the residence time is less than 5 minutes. In one form of the invention, where the reaction temperature is approximately 20 1450C, the residence time is about 30 seconds. Preferably, the alkaline solution has an initial total alkali concentration, expressed as grams of sodium carbonate per litre of solution, of between about 60 and 350. Preferably still, the alkaline solution has an initial total alkali concentration, expressed as grams of sodium carbonate per litre of solution, of between about 25 100 and 180. In one form of the invention, the alkaline solution is provided in the form of a Bayer process solution. Where the alkaline solution is provided in the form of a Bayer process solution, the alkaline solution is preferably still provided in the form WO 00/64812 PCT/AU00/00346 -7 of a process side-stream as opposed to a digestion solution. In one specific form of the invention, the alkaline solution is provided in the form of a washer overflow liquor. In a further specific form of the invention, the alkaline solution is provided in the form of a washer feed liquor. 5 Where the alkaline solution is provided in the form of a Bayer process washer overflow liquor, the method of the present invention has the added advantage of dissolving remaining aluminium containing solids. The magnitude of the quantity of lime added to the alkaline solution is dependent upon the level of carbonate in solution and the total alkali concentration of the 10 alkali solution. Typically however, the magnitude of the quantity of lime added to the alkaline solution corresponds to between 5 and 12 grams of calcium oxide per litre of reaction mixture. In accordance with the present invention, there is further provided a method for causticisation of a Bayer process solution, the method comprising the steps of: 15 obtaining the Bayer process solution from a Bayer circuit as a side stream; heating the Bayer process solution to a predetermined temperature exceeding 1100C; adding a quantity of lime to the heated Bayer process solution to form a reaction mixture in such a manner that the temperature of the reaction 20 mixture rapidly exceeds the atmospheric boiling point of the Bayer process solution; passing the reaction mixture into a reaction vessel; retaining the reaction mixture in the reaction vessel for a period not exceeding 15 minutes to generate a causticised Bayer process solution; and 25 returning the causticised Bayer process solution to the Bayer circuit.

WO 00/64812 PCT/AU00/00346 -8 Preferably, the Bayer process solution is obtained from the Bayer circuit from a point upstream of precipitation. Preferably, the reaction temperature is between about 1100C and 3000C. Preferably still, the reaction temperature is between about 1100C and 170C. 5 Preferably, the residence time does not exceed 10 minutes. Preferably still, the residence time is less than 5 minutes. Preferably, the Bayer process solution has a total alkali concentration of between 60 and 350 gL 1 . The Bayer process solution may be heated in one or more heating stages. In one 10 form of the invention, the or each heating stage involves steam heating. The or each heating stage may be performed in a shell and tube type heat exchanger, a plate type heat exchanger, a spiral type heat exchanger. Where the heating stage involves steam, the heating stage may be performed by direct steam injection. 15 In one form of the invention, the reaction vessel is agitated. In a further form of the invention, the reaction vessel is provided in the form of a pipe-reactor. The method of the present invention may, before the reaction mixture is passed into the reaction vessel, comprise the additional step of: passing the reaction mixture through an in-line mixing device. 20 In one form of the invention, the causticised Bayer process solution is returned to the Bayer circuit prior to a solid separation stage to allow the removal of lime residue solids. In a more specific form of the invention, the causticised Bayer process solution is returned to the Bayer circuit prior to a settling stage. In an alternate more specific form of the invention, the causticised Bayer process 25 solution is returned to the Bayer circuit prior to a filtration stage.

WO 00/64812 PCT/AU00/00346 -9 In an alternate form of the invention, the method for causticisation of a Bayer process solution of the present invention may, before returning the causticised Bayer process solution to the Bayer liquor circuit, comprise the additional steps of: directing the causticised Bayer process solution to a dedicated solid 5 separation stage to remove lime residue solids. In a more specific form of the invention, the causticised Bayer process solution is directed to a dedicated settling stage. In an alternate more specific from of the present invention, the causticised Bayer process solution is directed to a dedicated filtration stage. 10 Where the method for causticisation of a Bayer process solution of the present invention includes the step of directing the causticised Bayer process solution to a dedicated solid separation stage to remove lime residue solids, the method may further comprise the step of: collecting the lime residue solids for use as an aid to the filtration of the main 15 liquor stream. The method for causticisation of a Bayer process solution of the present invention may, before returning the causticised Bayer process solution to the Bayer circuit, comprise the additional step of: cooling the causticised Bayer process solution in one or more cooling 20 stages. The or each cooling stage may be performed in a shell and tube type heat exchanger, a plate type heat exchanger, a spiral type heat exchanger. Where the method for causticisation of a Bayer process solution of the present invention comprises the step of cooling the causticised Bayer process solution in 25 one or more cooling stages, the method may further comprise the step of: WO 00/64812 PCT/AU00/00346 - 10 donating heat generated by the or each cooling stage to the or each heating stage. In one form of the invention, at least a portion of the or each cooling stage is performed by flash cooling. Where at least a portion of the or each cooling stage 5 is performed by flash cooling, vapour emitted during flash cooling may be used in the or each heating stage. Any of the above described methods may also concurrently remove phosphate from the alkaline solution. BRIEF DESCRIPTION OF THE DRAWINGS 10 The present invention will now be described, by way of example only, with reference to one embodiment thereof, and the accompanying drawings, in which: Figure 1 is a schematic flow sheet showing how a method in accordance with the present invention may be utilised in a Bayer Process circuit; Figure 2 is a plot of TC/TA ratio against reaction time for a reaction involving 15 the addition of a quantity of slaked lime to a sample of green Bayer process liquor to form a reaction mixture at 1450C, in accordance with a first example of the present invention; Figure 3 is an X-ray diffraction spectrum of a solid phase isolated from the reaction mixture of Figure 2; 20 Figure 4 is a thermogravimetric analysis trace of the solid phase of Figure 3; Figure 5 is a composite plot of liquor total caustic to total alkali (TC/TA) ratios against reaction time for reactions involving the addition of a quantity of slaked lime corresponding to 5, 7 and 12 gL 1 , to a sample of green Bayer process liquor with an initial TA of 220 gL 1 , at 1450C in accordance with a 25 second example of the present invention; WO 00/64812 PCT/AU00/00346 -11 Figure 6 is a composite plot of liquor TC/TA ratios against reaction time for reactions involving the addition of a 5 gL -1 of slaked lime to samples of green Bayer process liquor with an initial TA of 220, 240 and 260 gL -1 at 1450C in accordance with a second example of the present invention; 5 Figure 7 is a composite plot of liquor TC/TA ratios for experiments involving causticisation of green liquor with initial TA 150gL 1 with different samples of slaked lime in accordance with a third example of the present invention. Figure 8 compares peak TC/TA ratios at 1300C for experiments involving causticisation of a green liquor with initial TA 150gL 1 with different samples of 10 slaked lime in accordance with a fourth example of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION By way of example, the method of the present invention is described in the context of the causticisation and phosphate reduction of a Bayer process solution, although such should not be seen as limiting the generality of the foregoing 15 description. Figure 1 shows a liquor 10 from a digestion stage 12 being fed into a thickener 14. Underflow 16 from the thickener 14 is passed to a series of washers 18, the overflow 20 from which is obtained from a Bayer circuit as a side stream, and heated to a temperature of about 1450C by way of heaters 22. 20 A quantity of slaked lime putty 24 is pre-heated to about 1450C, then added to the heated Bayer process solution to form a reaction mixture in such a manner that the temperature of the reaction mixture rapidly exceeds the atmospheric boiling point of the Bayer process solution. The magnitude of the quantity of lime added to the alkaline solution is dependent 25 upon the level of carbonate in solution and the total alkali concentration of the alkali solution. Typically however, the magnitude of the quantity of lime added to WO 00/64812 PCT/AU00/00346 -12 the alkaline solution corresponds to between 5 and 12 grams of calcium oxide per litre of reaction mixture. The mixture so formed is then passed to a reaction vessel 26, for example a pipe reactor, where it is retained for a period not exceeding about five minutes, thereby 5 generating a causticised Bayer process solution. The causticised Bayer process solution is then cooled by way of coolers 28 before being directed to a dedicated solid separation stage 30 to remove lime residue solids and returned to the Bayer circuit. The lime residue solids may be collected use as an aid to the filtration of the main liquor stream. 10 The vapour 32 generated by the flash cooling of the causticised Bayer process solution is transferred to the heating stage for the pre-causticised Bayer process solution and/or the heating stage for the lime putty. The following examples, each comprising several experiments, are intended to assist in the understanding of the reaction parameters of the present invention. 15 It must be appreciated that the following description of the examples is not to limit the generality of the above description of the invention. Example 1 A green liquor solution was prepared and maintained at a temperature of 145 0 C in a 3-litre autoclave. Slaked lime slurry was blown into the autoclave using a large, 20 stainless steel sampling bomb, thereby approximating the operation of plant digesters, and the resulting mixture sampled after 0, 10 and 30 minutes. Each sampling involved filtering a liquor sample through a 0.45 pm ACRODISC filter, analysing the filtrate for total caustic and total alkali by conventional titration methods, then washing and drying the solids at 600C before analysing such by X 25 ray diffraction techniques. Figure 2 plots TC/TA ratio against reaction time and shows a marked increase in liquor TC/TA ratio, from 0.824 to 0.868, within 10 minutes of the start of the WO 00/64812 PCT/AU00/00346 -13 reaction. The X-ray diffraction spectrum of Figure 3 indicates that calcium carbonate and tricalcium aluminate hexahydrate are both present in the solid phase of the reaction mixture, and the thermogravimetric analysis shown in Figure 4 confirms the existence of calcium carbonate. 5 Whilst the experiments of Example 1 illustrate that the method of the present invention is effective in causticising a Bayer process solution beyond the levels achieved by the prior art, the incomplete transfer of lime slurry from the stainless steel bomb renders such unsuitable for quantitative experiments. Example 2 10 A quantity of slaked lime slurry was sealed in glass ampoules and reacted with 600 mL samples of blow-off liquor at 1450C in a 1L autoclave. The experiment was performed with slaked lime charges corresponding to 5, 7 and 12 gL 1 CaO, and with blow-off liquor with total alkali concentrations of 220, 240 and 260 gL 1 sodium carbonate. Samples were taken at 0, 0.5, 2, 5, 10, 20 and 30 minutes 15 after contact between the slaked lime charge and the blow-off liquor was initiated. Liquor samples were filtered twice through 0.45pm ACRODISC filters, with the filtrate being analysed for total caustic and total alkali by conventional titration methods. The solids washed, dried at 600C then subjected to X-ray diffraction and thermogravimetric analyses. 20 The results of the titration analyses of the filtrate, expressed as a TC/TA ratio, for alkaline solutions with initial TA values of 220, 240 and 260 gL 1 appear below in Tables 1,2 and 3 respectively. Several reactions were performed for most slaked lime charges.

WO 00/64812 PCT/AU00/00346 -14 Time Lime Charge (mins) 5 gL' 1 7 gL " 1 12 gL

"

' 0 0.810 0.824 0.825 0.815 0.814 0.820 0.827 0.813 0.803 0.817 0.829 0.5 0.844 0.863 0.867 0.854 0.858 0.847 0.881 0.860 0.867 0.868 0.868 2 0.825 0.865 0.863 0.856 0.858 0.829 0.881 0.860 0.858 0.856 0.853 5 0.832 0.866 0.864 0.857 0.855 0.832 0.877 0.860 0.847 0.848 0.842 10 0.826 0.864 0.865 0.855 0.844 0.827 0.869 0.861 0.847 0.842 0.838 20 0.818 0.857 0.859 0.848 0.834 0.818 0.858 0.854 0.831 0.828 0.839 30 0.814 0.849 0.850 0.840 0.828 0.820 0.848 0.847 0.827 0.820 0.832 Table 1 (TA = 220 qL 1 ) Time Lime Charge (mins) 5 gL- " 7 gL - 12 gL 0 0.806 0.804 0.818 0.807 0.821 0.802 0.823 0.5 0.825 0.813 0.850 0.827 0.852 0.835 0.874 2 0.824 0.814 0.854 0.825 0.862 0.823 0.851 5 0.820 0.807 0.855 0.833 0.861 0.824 0.853 10 0.817 0.810 0.850 0.811 0.846 0.823 0.845 20 0.809 0.811 0.845 0.813 0.832 0.813 0.833 30 0.806 0.810 0.842 0.815 0.827 0.814 0.829 Table 2 (TA = 240 qL

"

) Time Lime Charge (mins) 5 g/L 7 g/L 12 g/L 0 0.826 0.823 0.827 0.809 0.816 0.832 0.809 0.830 0.826 0.5 0.828 0.824 0.831 0.820 0.827 0.831 0.819 0.847 0.842 2 0.816 0.831 0.827 0.820 0.814 0.818 0.816 0.825 0.823 5 0.822 0.824 0.830 0.814 0.815 0.811 0.815 0.825 0.831 10 0.823 0.826 0.827 0.814 0.829 0.814 0.816 0.831 0.830 20 0.810 0.817 0.828 0.804 0.828 0.827 0.816 0.826 0.828 30 0.818 0.817 0.822 0.811 0.828 0.834 0.817 0.829 0.833 Table 3 (TA = 260 gL 1 ) Figure 5 is a composite plot of liquor total caustic to total alkali (TC/TA) ratios 5 against reaction time for selected reactions involving the addition of a quantity of slaked lime corresponding to 5, 7 and 12 gL 1 to liquor with a TA of 220 gL

-

'. The results were selected on the basis of comparable starting TC/TA ratios. Once WO 00/64812 PCT/AU00/00346 -15 allowances for different starting TC/TA ratios are made, the TC/TA against time curves for repetition reactions agree within standard error. From Figure 5, it is apparent that the liquor TC/TA ratio rises rapidly upon breaking the glass ampoule and releasing the lime slurry into the liquor, the 5 maximum TC/TA ratio being achieved within two minutes of the start of the reaction, following which the TC/TA ratio decreases with increasing residence time. Table 4 summarises lime efficiency data for the experiments represented by Figure 5, the lime efficiency data being calculated from the results of liquor 10 titration analyses for samples taken after 30 seconds, where the TC/TA ratio is about its maximum, and after 30 minutes. Slaked Lime Charge Lime Efficiency at Lime Efficiency at (gL 1 -' CaO) T = 0.5 minutes T = 30 minutes. 5 gL 1 99% 66% 7 gL 1 75% 24% 12 gL 1 ' 53% 9% Table 4 From Figure 5 and Table 4, it is apparent that the magnitude of the rise in TC/TA ratio increases with increasing lime charge. However, the gains in TC/TA ratio 15 are not proportional to the increase in lime charge, indicating that the gains approach an upper limit. Lime efficiency falls from 99% where a lime charge of 5 gL 1 is employed, to 63% for the 12 gL -1 charge, suggesting that an optimum charge exists at which both lime efficiency and TC/TA gain is maximised. Additionally, the rate of decrease in TC/TA ratio from the point of maximum gain 20 increases with increasing lime charge. Without being bound by theory, the presence of calcium carbonate suggests that at temperatures above that at which conventional causticisation is conducted, the rate of formation of calcium carbonate increases such that it overcomes the formation of the thermodynamically favoured product, tricalcium aluminate. Accordingly, it is WO 00/64812 PCT/AU00/00346 -16 apparent that the greatest net increases in TC/TA ratios are attained at short residence times, particularly where high lime charges are used, whilst increased residence times result in the formation of tricalcium aluminate. Figure 6 is a composite plot of liquor TC/TA ratios against reaction time for 5 selected reactions involving the addition of a 5 gL 1 of slaked lime to samples of green Bayer process liquor with initial TA values of 220, 240 and 260 gL -1 . Again, the reactions were selected on the basis of comparability of initial TC/TA ratios. From Figure 6 it is apparent that the gain in TC/TA ratio decreases with increasing liquor TA, the greatest increase in liquor TC/TA being recorded for liquor with a 10 starting TA of 220 gL 1 , and a recorded increase in TC/TA of only 0.01 gL 1 being recorded for liquor with a starting TA of 260 gL- 1 . Given that the standard error for the concentration measurements taken is ±0.01 gL- 1 , it is apparent that little or no causticisation occurred. X-ray diffraction analysis of solids isolated from the reaction revealed the presence of appreciable quantities of calcium carbonate in 15 the 220 and 240 gL-1 reactions, whilst the solid isolated from the 260 gL 1 reaction exhibited a diffraction spectrum with calcium carbonate peaks barely exceeding the background level. Again, without wishing to be bound by theory, it is postulated that the rate of formation of calcium carbonate is retarded in liquors with high initial TA, such that the increase in reaction rate associated with the 20 elevated temperatures of the present invention are offset. Table 5 summarises lime efficiency data for the experiments represented by Figure 6, the lime efficiency data being calculated from the results of liquor titration analyses for samples taken after 30 seconds, where the TC/TA ratio is about its maximum, and after 30 minutes. TA Lime Efficiency at Lime Efficiency at (gL

-

') T = 0.5 minutes T = 30 minutes. 220 99% 66% 240 92% 71% 260 11% 0% 25 Table 5 WO 00/64812 PCT/AU00/00346 -17 Table 5 shows calculated lime efficiencies decrease from 99% at 220 gL -1 TA to 11% at 260 gL 1 TA, dropping to zero in the latter case after 30 minutes, so alkaline solutions with lower total alkali concentrations afford opportunities for greater increases in TC/TA ratio, and higher lime efficiencies. 5 Example 3 Dilute green liquor solutions were prepared by digesting spent liquor with hydrate. Samples of slaked lime slurry from the Kwinana refinery of Alcoa of Australia Ltd were sealed in glass ampoules, which were then placed in a Parr 1L autoclave with 600 mL of the filtered prepared dilute green liquor. The autoclave and its 10 contents were then heated to the desired reaction temperature. Upon attainment of the reaction temperature the glass ampoule was broken, releasing its contents into the hot liquor, and the experiment commenced. The range of reaction conditions under which the experiments were undertaken is as follows: Slaked lime charge: 8.1 - 9.0 gL -1 15 Liquor TA: 150 gL - 1 Initial liquor TC/TA ratio: 0.825 - 0.842 Initial liquor A/TC ratio: 0.669 - 0.677 Temperature: 1000C to 145 OC Sample Times: 0, 0.5, 2, 5, 10, 20 and 30 minutes 20 The liquor samples were filtered twice through 0.45 pm acrodiscs and analysed by ALIAN titration analysis and for calcia in liquor by ICP. The solid reaction products were washed, dried at 60 OC and analysed by XRD.

WO 00/64812 PCT/AU00/00346 -18 Tests on the slaked lime used in Examples 1 and 2 indicated that it did not react to completion. Hence the results of Examples 1 and 2 do not necessarily reflect the maximum TC/TA ratios that could be attained using the method of the present invention. Accordingly the experiments of Example 2 were then repeated using 5 standard and de-sanded slaked lime slurries from the Pinjarra refinery of Alcoa of Australia Ltd, under the conditions listed in Table 6. Pinjarra Desanded Slaked Lime Pinjarra Standard Slaked Lime Slaked lime charge 9.0 - 9.2 g/L 8.5 - 8.6 g/L Liquor TA: 150 g/L 150 g/L Initial liquor TC/TA ratio 0.832 - 0.840 0.830 - 0.838 Initial liquor A/TC ratio 0.670 - 0.685 0.669 - 0.671 Maximum TC/TA 0.885-0.943 0.913-0.918 Temperature 100C to 145 OC 130 0 C Table 6 The data from the experiments using the original Kwinana lime slurry closely accord with those of Example 2. After addition of the lime, the TC/TA ratio rises to 10 a maximum within 5 minutes, then gradually begins to decrease. The TC/TA ratio decreases more rapidly with increasing temperature, and maximum TC/TA ratios increase with increasing temperature. The TC/TA ratios achieved at temperatures above 1000C were again consistently above the calculated equilibrium TC/TA ratio of 0.870. 15 The results obtained using Pinjarra desanded slaked lime display similar behaviour to that observed with Kwinana slaked lime. Maximum TC/TA ratio increases with increasing reaction temperature and optimum residence times become progressively shorter. The subsequent drop in liquor TC/TA ratio becomes more rapid at higher temperatures. From the shapes of the reaction 20 curves indicate that the transition between thermodynamic and kinetic control still appears to occur between these two temperatures. Hence changing the type of slaked lime does not alter the behaviour of the reaction.

WO 00/64812 PCT/AU00/00346 - 19 However, the difference between these two sets of data lies in the maximum TC/TA ratio obtained for similar lime charges. As can be seen in Figure 7, those obtained using Pinjarra slaked lime are consistently higher than those produced with Kwinana slaked lime. This difference is not due to the slightly lower lime 5 charges used in the experiments with Kwinana slaked lime. Experiments at 1300C, using Kwinana lime charges of 8.8 gL 1 - 9.0 gL -1 CaO resulted in maximum TC/TA ratios of 0.880 - 0.885. These results were not significantly different from those obtained using lime charges of 8.1 g/L - 8.2 g/L CaO. The experiments using Kwinana slaked lime had achieved their maximum attainable 10 TC/TA ratios, notwithstanding the lower lime charge and, hence, the size of the lime charge did not substantially contribute to the difference between the Kwinana and the Pinjarra desanded slaked lime results. XRD analyses of the products of the causticisation reactions shows the presence of unreacted Ca(OH) 2 in the residue when Kwinana slaked lime is used. In 15 contrast, when Pinjarra slaked lime is used, no Ca(OH) 2 is apparent in the product. The lower TC/TA ratios obtained with Kwinana slaked lime are attributed to the fact that not all of the slaked lime reacts completely. In contrast, the slaked lime produced at Pinjarra reacts completely, thereby attaining very high TC/TA ratios. Thus, the slaked lime produced at the various refineries is not necessarily 20 of the same quality and "reactivity". Such differences can have a significant impact on the degree of causticisation achieved. In this case, at 1450C, a 42 point difference was observed between the Pinjarra desanded (0.942 TC/TA) and the Kwinana (0.900 TC/TA) results. The reactivity of the slaked lime also affects lime efficiencies. Despite a slightly 25 higher lime charge, the lime efficiencies obtained using Pinjarra desanded slaked lime are consistently higher than those attained with Kwinana slaked lime. This is due to both the higher TC/TA ratios reached in the Pinjarra desanded experiments as well as the presence of unreacted Ca(OH) 2 in the residue of the experiments with Kwinana slaked lime.

WO 00/64812 PCT/AU00/00346 - 20 Further, different results were generated by the sanded and de-sanded Pinjarra slaked lime samples, the de-sanded samples producing better results. The results generated by both the sanded and de-sanded Pinjarra samples being superior to the Kwinana samples. 5 Example 4 Slaked lime slurries from the refineries of Alcoa of Australia Ltd at Wagerup, Point Comfort and the bright hydrate plant at Kwinana were tested under the conditions listed in Table 7. Wagerup Slaked Point Comfort Slaked Commercial Slaked Lime Lime Lime Slaked lime charge 9.4 - 9.5 g/L 9.0 g/L 9.0 - 9.1 g/L Liquor TA 150 g/L 150 g/L 150 g/L Initial liquor TC/TA ratio 0.833 0.826 - 0.832 0.825 - 0.842 Initial liquor A/TC ratio 0.674 - 0.675 0.677 - 0.679 0.670 - 0.671 Maximum TC/TA 0.883-0.890 0.906-0.912 0.919-0.922 Temperature 130 0 C 1300C 130 0 C Table 7 Further, a commercially available lime putty was tested. The putty is produced by 10 slaking commercial dry lime in steam. This process produces a very fine (dso 0 = 8 - 10 pm) slaked lime slurry with a high solids density (> 400 gL 1 ). The data from these experiments are in accordance with those of Examples 2 and 3. In all cases the TC/TA ratio rises to a maximum within 30 seconds of the addition of slaked lime. Lime efficiencies are below 100 % as excess slaked lime 15 was added so as to maximise TC/TA. The lowest lime efficiencies were obtained with Wagerup slaked lime. Figure 8 compares the maximum TC/TA ratios obtained with each of the slaked lime slurries tested. Pinjarra desanded slaked lime is the most reactive of all of the slaked limes tested, producing maximum TC/TA ratios of 0.926 to 0.930. The 20 lime putty is as reactive as Pinjarra standard slaked lime, but less reactive than WO 00/64812 PCT/AU00/00346 -21 the desanded lime. Point Comfort slaked lime is as reactive as Pinjarra standard slaked lime, but not as reactive as Pinjarra desanded slaked lime. The Kwinana and Wagerup refineries produced the least reactive slaked lime slurries. Of the slaked lime samples studied, the de-sanded Pinjarra slaked lime slurry was 5 the most reactive, producing maximum TC/TA ratios of up to 49 points higher than the least reactive slaked limes at 130 OC. Point Comfort slaked lime is as reactive as Pinjarra slaked lime before it is de-sanded, but not as reactive as Pinjarra slaked lime after sand removal. In terms of reactivity, Point Comfort slaked lime ranks among the better performers in the suite of slaked limes tested. 10 Thus, the reactivity of the slaked lime slurry will have a significant effect on the maximum TC/TA ratio at any reaction temperature. Slaked limes that do react completely, leaving unreacted Ca(OH) 2 in the causticiser residue, will produce lower TC/TA ratios than those obtained from highly reactive slaked lime. This difference in reactivity may be due to variations in the control of the slaking 15 process. From these results, it can generally be seen that the method of the present invention is highly effective at causticising alkaline solutions, in the form of Bayer process solutions, beyond the levels achievable by the prior art. This is achieved with excellent lime efficiency, by producing calcium carbonate as the dominant 20 product. In addition to the increased lime efficiency and the higher TC/TA ratios achievable over conventional causticisation, the diminished production of tricalcium aluminate hexahydrate arising from the method of the present invention will result in decreased residual red mud alkalinity, reducing the environmental difficulties 25 associated therewith. Further, it is anticipated that the calcium carbonate by product of the method of the present invention will be of a particle size suitable for use as a filter aid.

WO 00/64812 PCT/AU00/00346 - 22 The highly desirable short residence times of the present invention offer a considerable advantage over the residence times associated with conventional causticisation, being over an hour. Given the benefits for maximum TC/TA gain and lime efficiency associated with 5 liquors with relatively low TA levels, it is envisaged that the method of the present invention would be applied to such liquors, such as washer overflow liquors. It is envisaged that the fine CaCO 3 produced by the method of the present invention may be utilised as a low-cost filter aid, replacing the tricalcium aluminate hexahydrate filter aid presently used and thereby further reducing the lime 10 requirement. It is envisaged that the method of the present invention may also concurrently remove phosphate from the alkaline solution. It is further envisaged that the lime residue solids produced by the method of the present invention may be re-calcined and utilised as a source of lime in the 15 method of the present invention. It is still further envisaged that the lime residue solids produced by the present invention may be utilised in one ore more subsequent causticisation operations downstream from the point of application of the method of the present invention. Modifications and variations such as would be apparent to the skilled addressee 20 are considered to fall within the scope of the present invention.

Claims (54)

1. A method for the causticisation of an alkaline solution, the method characterised by the step of: adding a quantity of lime to the alkaline solution to produce a reaction 5 mixture in such a manner that the reaction temperature of the reaction mixture so produced rapidly exceeds the atmospheric boiling point of the alkaline solution.

2. A method according to claim 1 characterised in that the ratio of total caustic concentration to total alkali concentration of the alkaline solution, when 10 expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.82, after the addition of the quantity of lime.

3. A method according to claim 2 characterised in that the ratio of total caustic concentration to total alkali concentration, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.85. 15

4. A method according to claim 3 characterised in that the ratio of total caustic concentration to total alkali concentration, when expressed in grams of sodium carbonate per litre of solution, exceeds approximately 0.87.

5. A method according to claim 4 characterised in that the ratio of total caustic concentration to total alkali concentration, when expressed in grams of sodium 20 carbonate per litre of solution, exceeds approximately 0.90.

6. A method according to any one of the preceding claims characterised both the quantity of lime and the alkaline solution are maintained about the reaction temperature before the quantity of lime is added to the alkaline solution.

7. A method according to any one of claims 1 to 5 characterised in that the 25 quantity of lime is maintained at a temperature below the reaction temperature WO 00/64812 PCT/AU00/00346 -24 before the quantity of lime is added to the alkaline solution, the reaction mixture so produced being rapidly heated to the reaction temperature.

8. A method according to claim 7 characterised in that the reaction mixture reaches the reaction temperature in less than 10 minutes. 5

9. A method according to claim 8 characterised in that the reaction mixture reaches the reaction temperature in less than 5 minutes.

10. A method according to any one of the preceding claims characterised in that the quantity of lime is provided in the form of slaked lime.

11. A method according to claim 10 characterised in that the slaked lime is 10 provided in the form of slaked lime putty.

12. A method according to claim 11 characterised in that the slaked lime putty has a low particle size and a high solids density.

13. A method according to claim 12 characterised in that the slaked lime putty has particle size d 50 o = 8 - 10 pm, solids density > 400 gL 1 . 15

14. A method according to any one of the preceding claims characterised in that the reaction temperature is above 110°C.

15. A method according to claim 14 characterised in that the reaction temperature is between about 1100C and 3000C.

16. A method according to claim 15 characterised in that the reaction temperature 20 is between about 110OC and 1700C.

17. A method according to any one of the preceding claims characterised in that the residence time of the quantity of lime in the alkaline solution is less than 15 minutes. WO 00/64812 PCT/AU00/00346 - 25

18. A method according to claim 17 characterised in that the residence time is less than 5 minutes.

19. A method according to any one of the preceding claims where the reaction temperature is 1450C and the residence time is about 30 seconds. 5

20. A method according to any one of the preceding claims characterised in that the alkaline solution has an initial total alkali concentration, expressed as grams of sodium carbonate per litre of solution, of between about 60 and 350.

21. A method according to claim 20 characterised in that the alkaline solution has an initial total alkali concentration, expressed as grams of sodium carbonate 10 per litre of solution, of between about 100 and 180.

22. A method according to any one of the preceding claims characterised in that the alkaline solution is provided in the form of a Bayer process solution.

23. A method according to claim 22 characterised in that the alkaline solution is preferably still provided in the form of a process side-stream as opposed to a 15 digestion solution.

24. A method according to claim 23 characterised in that the process side-stream is obtained from the Bayer circuit from a point upstream of precipitation.

25. A method according to any one of claim 22 to claim 24 characterised in that the alkaline solution is provided in the form of a washer overflow liquor. 20

26. A method according to any one of claim 22 to claim 24 characterised in that the alkaline solution is provided in the form of a washer feed liquor.

27. A method according to any one of the preceding claims characterised in that the magnitude of the quantity of lime added to the alkaline solution corresponds to between 5 and 12 grams of calcium oxide per litre of reaction 25 mixture. WO 00/64812 PCT/AU00/00346 - 26

28. A method for causticisation of a Bayer process solution, the method characterised by the steps of: obtaining the Bayer process solution from a Bayer circuit as a side stream; heating the Bayer process solution to a predetermined temperature 5 exceeding 1100C; adding a quantity of lime to the heated Bayer process solution to form a reaction mixture in such a manner that the temperature of the reaction mixture rapidly exceeds the atmospheric boiling point of the Bayer process solution; 10 passing the reaction mixture into a reaction vessel; retaining the reaction mixture in the reaction vessel for a period not exceeding 15 minutes to generate a causticised Bayer process solution; and returning the causticised Bayer process solution to the Bayer circuit.

29. A method according to claim 28 characterised in that the reaction temperature 15 is between about 110OC and 3000C.

30. A method according to claim 29 characterised in that the reaction temperature is between about 1100C and 1700C.

31. A method according to any one of claims 28 to 30 characterised in that the residence time does not exceed 10 minutes. 20

32. A method according to claim 31 characterised in that the residence time is less than 5 minutes.

33. A method according to any one of claims 28 to 32 characterised in that the Bayer process solution has a total alkali concentration of between 60 and 350 gL -1 . WO 00/64812 PCT/AU00/00346 - 27

34. A method according to any one of claims 28 to 33 characterised in that the Bayer process solution is heated in one or more heating stages.

35. A method according to claim 35 characterised in that the or each heating stage involves steam heating. 5

36. A method according to claim 25 characterised in that the or each heating stage is performed by direct steam injection.

37. A method according to any one of claims 34 to 36 characterised in that the or each heating stage may be performed in a shell and tube type heat exchanger, a plate type heat exchanger, a spiral type heat exchanger. 10

38. A method according to any one of claims 28 to 37 characterised in that the reaction vessel is agitated.

39. A method according to any one of claims 28 to 37 characterised in that the reaction vessel is provided in the form of a pipe-reactor.

40. A method according to any one of claims 28 to 39 characterised in that before 15 the reaction mixture is passed into the reaction vessel, the method comprises the additional step of: passing the reaction mixture through an in-line mixing device.

41. A method according to any one of claims 28 to 40 characterised in that the causticised Bayer process solution is returned to the Bayer circuit prior to a 20 solid separation stage to allow the removal of lime residue solids.

42. A method according to claim 41 characterised in that the causticised Bayer process solution is returned to the Bayer circuit prior to a settling stage.

43. A method according to claim 41 characterised in that the causticised Bayer process solution is returned to the Bayer circuit prior to a filtration stage. WO 00/64812 PCT/AU00/00346 - 28

44. A method according to claim 41, the method being characterised by, before returning the causticised Bayer process solution to the Bayer liquor circuit the additional steps of: directing the causticised Bayer process solution to a dedicated solid 5 separation stage to remove lime residue solids.

45. A method according to claim 44 characterised in that the causticised Bayer process solution is directed to a dedicated settling stage.

46. A method according to claim 44 characterised in that the causticised Bayer process solution is directed to a dedicated filtration stage. 10

47. A method according to any one of claims 44 to 46 characterised by the additional step of: collecting the lime residue solids for use as an aid to the filtration of the main liquor stream.

48. A method according to any one of claims 28 to 47 characterised by the 15 additional step of: cooling the causticised Bayer process solution in one or more cooling stages.

49. A method according to claim 48 characterised in that the or each cooling stage is performed in a shell and tube type heat exchanger, a plate type heat 20 exchanger, a spiral type heat exchanger.

50. A method according to either of claims 48 and 49 characterised by the additional step of: donating heat generated by the or each cooling stage to the or each heating stage. WO 00/64812 PCT/AU00/00346 - 29

51. A method according to any one of claims 48 to 50 characterised in that at least a portion of the or each cooling stage is performed by flash cooling.

52. A method according to claim 51 characterised in that vapour emitted during flash cooling may be used in the or each heating stage. 5

53. A method according to any one of the preceding claims characterised in that phosphate ions are concurrently removed from the alkaline solution.

54. A method for causticising an alkaline solution substantially as hereinbefore described with reference to Example 1, 2, 3, 4 or 5.