According to a first aspect of the invention, there is provided a polymeric or oligomeric compound of formula I

wherein Ri is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, fused aromatic rings, substituted fused aromatic rings, and aralkyl groups, or is absent;

R2 is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, linked aromatic rings, fused aromatic rings, substituted fused aromatic rings and aralkyl groups; each X is independently H, alkyi or phenyl or X-N-R2-N-X form a ring in which R2 is alkyl or substituted alkyl and the two X groups together are alkyl or substituted alkyl; and m is an integer greater than or equal to 2.

According to a second aspect of the invention, there is provided a method of preparing a compound of formula I

wherein m is an integer greater than or equal to 2, the method including the steps of reacting, in an anhydrous solvent, a bi-functional organic acyl halide of formula 11

wherein Y is Cl, Br or l; with a thiocyanate salt to form a rection mixture; and adding, to the rection mixture, a diamine of formula III or formula IV or an amino or alkylamino substituted heterocyclic amine

wherein Ri is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, fused aromatic rings, substituted fused aromatic rings, and aralkyl groups, or is absent; R2 is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, linked aromatic rings, fused aromatic rings, substituted fused aromatic rings and aralkyl groups; and each X is independently H, alkyl or phenyl or X-N-R2-N-X form a ring in which R2 is alkyl or substituted alkyl and the two X groups together are alkyl or substituted alkyl.

In one embodiment of the invention, a diamine of formula IV is added to the rection mixture. The diamine may be a heterocyclic amine e. g. a bi-functional or multi-functional heterocyclic diamine, such as piperazine or a substituted piperazine.

Preferably, the compound of formula III or IV is added to the reaction mixture in the same anhydrous solvent as the acyl halide, e. g. anhydrous acetone.

Preferably, the rection of the acyl halide with the thiocyanate salt is effected at elevated temperature or with heating at reflux. Similarly, the rection mixture is preferably heated under reflux after the bi-functional amine or heterocyclic amine has been added.

The thiocyanate salt is typically a mental thiocyanate salt, e. g. KSCN, or ammonium thiocyanate.

Typically, the acyl halide is an acyl chloride.

According to a third aspect of the invention, broadly, there is provided a method of recovering a transition metal from an aqueous solution containing the transition metal, the method including contacting the solution with a sorbent comprising a polymeric or oligomeric compound having at least one thiourea moiety in its repeating unit thereby to absorb and retain the transition metal on the sorbent.

Preferably, the polymeric or oligomeric compound has two thiourea moities in its repeating unit.

More particularly according to the third aspect of the invention, there is provided a method of recovering a transition metal from an aqueous solution containing the transition metal, the method including contacting the solution with a sorbent comprising a polymeric or oligomeric compound of formula I

wherein Ri is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, fused aromatic rings, substituted fused aromatic rings, and aralkyl groups, or is absent; R2 is selected from linear and branche alkyl and alkenyl groups, C6 aromatic rings, substituted C6 aromatic rings, linked aromatic rings, fused aromatic rings, substituted fused aromatic rings and aralkyl groups; each X is independently H, alkyl or phenyl or X-N-R2-N-X form a ring in which R2 is alkyl or substituted alkyl and the two X groups together are alkyl or substituted alkyl; and m is an integer greater than or equal to 2, thereby to absorb and retain the transition metal on the sorbent.

The solution may be strongly acidic, having a pH of 2 or less. The transition metal may then be one or more of the Platinum Group Metals, i. e. Pt, Pd, Rh, Ir, Os, or Ru in any of their oxidation states, e. g. Pt (lV), Pt (lI), Pd (ll), Ir (lV), Rh (III), Rh (l), Ru(IV), Os (IV), or the like. The transition metal may also be Au (I/III).

The solution may be an effluent stream from a metal separation process for the production of Platinum Group Metals and may thus include anions such as chlorides, nitrates and sulphates, and transition metals, other than Platinum Group Metals, such as Cu, Ni, Fe and Zn.

Instead, the solution may have a pH of greater than 4. The transition metal may then be one or more of Ni (il), Cu (II), Fe(II), Fe (III), Zn(II), Co (il), Co(III), Hg(II), Ag (l) or the like.

The sorbent may absorb the transition metal, when it is a Platinum Group Metal, up to at least 20 mass % of the dry mass of the sorbet. Preferably, the sorbent absorbs the transition metal up to at least 25 mass % of the dry mass of the sorbet. Most preferably, the sorbent absorbs the transition metal up to at least 30 mass % of the dry mass of the sorbet.

The solution may be at elevated temperature when being contacte with the sorbent, e. g. at a temperature of about 60°C to 80°C.

The method according to the third aspect of the invention may include recovering the transition metal from the sorbent, e. g. by treating the sorbent with a strong acid, or by ashing the sorbet.

R1 may be- (CH2) n-where n is an integer greater than or equal to one,- (CH3) 2C-CH2-,- (CH3) CH- (CH3) CH-,-CH =CH-,-C6H4-, C6H3 (COOH)- or - (CH2) pC6H4 (CH2) q- where p and q are integers greater than or equal to one.

When Ri is- (CH2) n-, n is typically between 2 and 8.

Preferably Ri is the aromatic ring-C6H4-.

R2 may be- (CH2) k- and k may be an integer greater than or equal to 1, e. g. 6.

Instead, R2 may be the aromatic ring-C6H4-or the linked aromatic rings-C6H4-C6H4-.

When X-N-R2-N-X form a ring, which is preferred, the ring may be a piperazine or substituted piperazine, e. g. 2,5 dimethyl piperazine.

Preferably, m is between 2 and 1000. More preferably, m is between 10 and 20.

Typically, the compound of formula I comprises 50 mass % to 55 mass % C, e. g. 53.3 mass%, 4 mass °/0 to 5 mass % H, e. g. 4.7 mass %, 15 mass % to 17 mass % N, e. g. 15.4 mass %, and 14 mass % to 18 mass % S, e. g. 14.9 mass %.

The compound of formula I may be selected from the group consisting of poly (N-terephthaloylthiourea)-N', N'-piperazineand poly- (N- isophthaloylthiourea)-N', N'-piperazine.

The invention will now be described, by way of example, with reference to the following examples and the accompanying drawings, in which Figure I shows a graph of the absorbtion of Pt (ll) over time by poly (N-terephthaloylthiourea)-N', N'-piperazine; Figure 11 shows a graph of the absorbtion of Pt (ll) over time by poly (N-terephthaloylthiourea)-N', N'-piperazine and poly- (N- isophthaloylthiourea)-N', N'-piperazine;

Figure III shows a graph of the absorbtion of Pd (ll) over time by poly (N-terephthaloylthiourea)-N', N'-piperazine and poly- (N- isophthaloylthiourea)-N', N'-piperazine; Figure IV shows a graph of the absorbtion of Pt (ll) in the presence of Ni (ll) and Cu (ll) over time by poly (N-terephthaloylthiourea)-N', N'- piperazine; Figure V shows a graph of the absorbtion of Pt (ll) in the presence of Pd (ll) over time by poly (N-terephthaloylthiourea)-N', N'-piperazine; Figure VI shows a graph of percentage precious metal left in solution, as a function of time, when a precious metal containing solution is contacte with poly- (N-terephthaloylthiourea)-N', N'-piperazine; and Figure VII shows a graph of percentage platinum and selenium left in solution, as a function of time, when a platinum and selenium containing solution is contacte with poly-(N-terephthaloylthiourea)-N', N'-piperazine.

EXAMPLE 1 Synthesis of poly (N-terephthaloylthiourea)-N', N'-piperazine To a stirred solution of KSCN (7.7973 g, 0.0802 mol) in dry acetone (100 ml) was added a solution of terephthaloyl chloride (8.1278 g, 0.0400 mol) in acetone (210 ml) under an inert atmosphere. The mixture was heated under reflux for 40 min, after which a solution of piperazine (3.4452 g, 0.0400 mol) in acetone (130 mi) was added dropwise. The rection mixture was heated under reflux for a further 40 min. The mixture was then poured into ice water (400 mi) to crystallise the product, hereinafter referred to as Bila. Once the acetone had evaporated off, the product was collecte by centrifugation and was washed three times with water and then twice with acetone. The crude

product was then dried in vacuo. The yield of B1 a was 10.1907 g (76.18 mass %).

1.0022 g of B1 a was redissolved in and recrystallised from a mixture of dimethyl sulphoxide and water to yield the product hereinafter referred to as B1 b. The yield of B1 b was 0.3324 g (33.23 mass %).

A further portion of B1 a was again redissolved in and recrystallised from a mixture of dimethyl sulphoxide and water to yield the product hereinafter referred to as B1 c.

The procedure for the synthesis of poly (N- terephthaloylthiourea)-N', N'-piperazine (Product B1 a) as set out above was repeated to give the product hereinafter referred to as B7a.

Elemental analysis of the products B1 a, B1 b, B1 c and B7a provided the results as set out in Table 1.

Table 1: Elemental analysais of products B1a, B1b, B1c and B7a: Calculated * B1a B1b B1c B7a Mass % Mass % ll/lass % MasS % Mass % C 53.2 52.53 49.39 53.59 52.03 H 4.5 4.45 3.95 4.60 4.66 N 16.1 15.69 17. 87 15.56 1. S 15.8 14.85 18.20 14.50 14.64 * Calculated from C27H27O4S3N7

As can be seen from Table 1, B1 b can be deemed to be an unsuccessful attempt at recrystallisation of Bila, assuming that the calculated mass composition accurately reflets the composition of poly(N-terephthaloylthiourea)-N', N'-piperazine.

EXAMPLE2 Synthesis of poly- (N-isophthaloylthiourea)-N', N'-piperazine To a stirred solution of KSCN (3.9827 g, 0.0410 mol) in dry acetone (55 ml) was added a solution of isophthaloyl chloride (4.0612 g, 0.0200 mol) in acetone (100 mi) under an inert atmosphere. The mixture was heated under reflux for 60 min, after which a solution of piperazine (1.7227 g, 0.0200 mol) in acetone (75 mi) was added dropwise. The rection mixture was heated under reflux for a further 60 min. The mixture was then poured over ice water (200 ml) to crystallise the product hereinafter referred to as B4a. Once the acetone had evaporated off, the product was collecte by centrifugation and was washed three times with water and then twice with acetone. The product was dried in vacuo. The yield of B4a was 5.3180 g (79.51 mass %).

0.5024 g of B4a was redissolved in and recrystallised from a mixture of dimethylformamide and water to yield the product hereinafter referred to as B4b. The yield of B4b was 0.2563 g (51. 02 mass %).

Elemental analysis of the products B4a and B4b provided the results as set out in Table 2.

Table 2: Elemental Analysis of the products B4a and B4b:

Ca (cutated *'B4a B4b Mass % Mass % Mass % C 53.2 52.87 53.89 H 4.5 4.64 4.57 N 16.1 15.89 15.61 S 15.8 15.12 1 15.38 11 Calculated from C27 H27 °4 S3 N7 EXAMPLE3 A number of absorption rate studies were carried out for the products B1 a and B4a. In each of the absorbtion rate studies, a tenfold molar excess of the product (B1 a or B4a) being investigated was added to an acid solution (pH < 2) containing at least one Platinum Group Metal, and the solution was stirred vigorously. Samples (approximately 5 ml) were withdrawn with a pastuer pipette at predetermined times and were filtered through filter paper (Whatman No 1) into pill vials for analysis. Each absorption rate study was conducted in quadruplet, with the starting time of each solution being staggered by exactly 1 minute, so that it was possible to take samples from each solution at exactly the right time. In each case, samples were taken within 20 seconds of the time required. Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was used as the detection method for the Platinum Group Metal or metals in the amples.

The following absorption rate studies were carried out:

Absorption Rate Study A: 0.170 g of B1 a was added to a 0,05M HCI solution containing 100ppm Pt (II). The results for two separate studies (A1 and A2) are set out in Table 3 and is shown in Figure I.

Table 3: time % Pt absd (1996) % Pt absd (1997) (minutes) during Study Au dring Study A2 15 56.75 74.0 30 74.12 88.45 60 93.6 98.68 120 99.87 99.76 240 99.87 100 1440 100 100 Absorption Rate Studv B: 0,314 g of B1 a was added to a 0,05M HCI solution containing 100ppm Pd (II). The results for two separate studies (B1 and B2) are set out in Table 4.

Table 4: time % Pd absd % Pd absd (minutes) during Study Bu dring Study B2 15 99.99 99.46 30 100 99.99 60 100 100

Absorption Rate Studv C: 0,17 g of B4a was added to a 0,05M HCI solution containing 100ppm Pt (li). The results are set out in Table 5 and in Figure 2, compare to the same study for B1 a (Study A2).

Table 5: time % Pt absd (B4a) % Pt absd (B1a) (minutes) l 15 50.26 74.0 30 54.71 88.45 60 62.51 98.68 120 66.71 99.76 240 80.49 100 1440 99.76 100 Absorption Rate Studv D: 0,314 g of B4a was added to a 0,05M HCI solution containing 100ppm Pd(II). The results are set out in Table 6 and in Figure 3, compare to the same study for B1 a (Study B2).

Table 6:

time % Pd absd (B4a) % Pd absd (B1a) (minutes) 15 79.01 99.46 30.86.325 99.99 60 92.35 100 120 97.26 100 240 99.79 100 1440 100 100 Absorption Rate Studv E: 0,0943 g of B1 a was added to a 0,05M HCI solution containing 60ppm Pt(II) 500ppm Cu(II) and 500ppm Ni (II). The results are set out in Table 7 and in Figure 4, compare to the previous study A2 for B1 a in the absence of Ni (ll) and Cu(II) set out above.

Table 7: time % Pt absd % Pt absd (minutes) in the presence of (without Ni & Cu) Cuill) and Ni(II) 5 34.72 Sample not taken 10 40.21 Sample not taken 15 46.01 74.0 30 57.82 88.45 60 73.70 98.68 120 91.36 99.76 240 99.26 100

Absorption Rate Studv F: 0,157 g of B1 a was added to a 0,05M HCI solution containing 50ppm Pd(II), 500ppm Cu (ll) and 500ppm Ni (II). The results are set out in Table 8, compare to the previous study B2 for B1 a in the abence of Ni(II) and Cu(II) set out above.

Table 8: time % Pd absd % Pd absd (minutes) in the presence of (without Ni & Cu) Cu(II) and Ni(II) 5 77.4 Sample not taken 10 87.33 Sample not taken 15 94.35 99.46 30 99.98 99.99 60 100 100 Absorption Rate Studv G: 0,17 g of B1 a was added to a 0,05M HCI solution containing 50ppm Pt(II) and 27ppm Pd(II). The results are set out in Table 9 and in Figure 5. In Figure 5, the Pt(II) absorption is compare to Study A1 set out above.

Table 9:

time (minutes) % Pt absd % Pd absd 8 49.06 99.74 15 63.41 99.84 30 84.78 100 60 98.33 100 120 99.5 100 EXAMPLE4 A number of capacity studies were conducted by adding differing amounts (in duplicate or triplicate) of the product under investigation (B1 a, B1 c, B4a) to an acidic solution containing at least one Platinum Group Metal such that the metal was in excess. The solutions were stirred vigorously for 3 days for palladium solutions and 7 days for platinum solutions, at which time a 5 mi sample was withdrawn and filtered into a pill vial for ICP-AES analysis.

The result for the capacity study for B1 a in a Pt(II) solution is shown in Table 10.

Table 10: Results for capacity of B1a for Pt (ll) I Bla mg Pt absd g Pt/g B1a RUN A 0.0314 9.971 0.3175 RUN B 0.0311 9.944 0.3197 RUN C. 03 9.967 0.3215 RUN D 0.0307 9.938 0.3237 An average of 0.3206 g Pt was absorbe per gram of B1 a.

The results for the capacity study for B1 a in a Pt(II) solution is shown in Table 11.

Table 11: Results for capacitv of B1a for Pt(II) g B1a mg Pd absd g Pd/g Bla I RUN A 0.0200 4.51 0.2255 | RUN B 0.0201 4.67 0.2323 RUN C 0.0202 4.56 0.2257 RUN D 0.0249 5.52 0.2217 RUN E 0.0250 5.54 0.2216 RUN F 0.0297 5.73 0.1929 RUN G 0.0298 6.91 0.2319 The results obtained for the palladium capacity of B1 a seem more consistent than the platinum capacity of Bl a, with an average capacity of 0.2216 g Pd adsorbe per gram of B1 a.

The results for the capacity study for B1 c in a Pt(II) solution is shown in Table 12.

Table 12: Results for capacity studies of B1c for Pt(II) g B1c mg Pt absd g Pt/g B1c RUN A 0.0100 5.17 0.5170 RUN B 0.0103 5.48 0.5320 RUN C 0.0149 6.32 0.4242 RUN D 0.0151 6.42 0.4252 The results for the capacity study for B4a in a Pt(II) solution is shown in Table 13.

Table 13: Results for capacitv of B4a for Pt (ll) g B4a mg Pt absd g Pt/g B4a RUN A 0.0151 4.52 0.2993 RUN B 0.0153 8.31 0.5431 RUN C 0.0200 5.29 0.265 RUN D 0.0201 5.50 0.274 RUN E 0.0204 9.26 0.454 Clearly, the capacity of B4a for Pt (II), despite being inconsistent, is substantially less than the capacity of B1 a for Pt (II), as shown in Table 11 above.

The results for the capacity study for B4a in a Pd(II) solution is shown in Table 14.

Table 14: Results for canacitv of B4a for Pd (ll)

g B4a mg Pd absd g Pd/g B4a RUN A 0.0197 1.92 0.0975 RUN B 0.0204 1.98 0.0971 RUN C 0.0248 3.98 0.1605 1 RUN D 0.0250 2.27 0.0908 1 Clearly, the capacity of B4a for Pd (II), despite being inconsistent, is substantially less than the capacity of B1 a for Pd (il), as shown in Table 12 above.

The Applicants believe that, in order for more precise and consistent results for these capacity studies, either these solutions must be allowed to stir for longer period of time, or the study should be performed at elevated temperature in order to ensure that any kinetic problems are avoided.

EXAMPLE 5 The kinetics of Platinum Group Mental and Au absorption by poly-(N-terephthaloylthiourea)-N,(N-terephthaloylthiourea)-N , N-piperazine were tested using a typical primary feed solution from a PGM refinery.

Inductively coupled plasma atomic emission spectroscopy (ICP-AES) and inductively coupled plasma mass spectrometry (ICP-MS) were used to determine the individual PGM, Au, base metal (Cu, Ni, Fe, Pb) and amphoteric metal (Sb, Sn, As, Se, Te, Bi, Zn) concentrations of the feed solution. The individual PGM concentrations and the base metal concentrations of the solution were all above 1.9 g/litre. The amphoteric metal concentrations were all above 50 ppm.

Two 25 ml aliquots of the feed solution were then diluted to 500 ml volumes and 1.5 M HCI concentration. Each solution was heated to 70°C. Then 15 g of the polymer was added to each solution and the mixture continuousty stirred. At regular time intervals, a 10 mi sample of each solution was filtered to remove any polymer. The filtered sample solution was then analyse by ICP-AES and ICP-MS for its PGM, base mental and amphoteric metal content. After correcting for the dilution factor, these concentration values were recalculated as a percentage of the metal concentrations in the original feed.

The results are shown in Table 15 and in Figure 6 and indicate that all the PGMs were absorbe by the polymer within 200 minutes-this despite the initial high concentrations of base and amphoteric metals in solution.

Table 15: Results of Polvmer absorption of precious étals from real process solutions in 1.5M hvdrochloric acid: 15 q poly-(N-terephthaloylthiourea)-N',N'- piperazine polymer per 500 mi solution at 60-70° Celcius Time/ Pt/% Pd/% Rh/% Ir/% left Ru/% Au/% Os/% min left left left solution left left : Ceft solution solution solution solution solution solution 0 100 100 100 100 100 100 100 30 11 1 39 53 69 1 43 45 0 0 7 26 33 1 25 60 0 0 2 11 7 1 31 90 0 0 1 3 0 1 24 120 0 0 0 2 0 1 12 150 0 0 0 2 0 1 6 180 0 0 0 2 0 1 2 210 0 0 0 2 0 1 2 225 1 0 0 1 0 1 0

EXAMPLE 6 The kinetics of platinum (Pt) absorption by poly- (N- terephthaloylthiourea)-N', N'- piperazine was also tested using the effluent generated by a Pt recovery stage in a PGM refinery. The mother liquor of this effluent is a relatively pure Pt solution. For recovery, the Pt is precipitated as ammonium hexachloroplatinate. The solution remaining is regarde as an effluent. However, it still contains a significant amount of Pt, and is thus usually recycle in the refining process. Excepting for Se, few other metals are present in the solution.

The specific batch of effluent used for the absorption tests with the polymer was found to contain 40 ppm Pt and 120 ppm Se. For the tests, 500 ml aliquots of effluent with 2.0 g of polymer were stirred at room temperature. At regular intervals, a 10 ml sample of each solution was filtered to remove any polymer. The filtered sample solution was then analyse by ICP-MS for its Pt and Se content. These concentration values were recalculated as a percentage of the metal concentrations in the original effluent.

The results are shown Figure 7. It is clear that Pt is selectively and quantitatively absorbe by the polymer, while the Se remains dissolve in the effluent.

It is a further avantage of some of the polymeric or oligomeric compound of the invention, as exemplified, when used as sorbets, that they can absorb the Platinum Group Metal ions to up to 30% of the dry mass of the sorbent, and that it is relatively easy to recover the valable Platinum Group Metal ions from these sorbets.

It is yet a further avantage of the polymeric or oligomeric compound of the invention as exemplified that they can be prepared easily from inexpensive starting materials.

It is another avantage of the polymeric or oligomeric compound of the invention, as exemplified, that under appropriate conditions and in higher pH solutions, they can also serve as sorbets for transition metals other than Platinum Group étals.