This disclosure relates to the detection of cyanide and to methods for doing so. More particularly, it relates to the detection of impurities that could give false results.

Cyanide compounds are essential substances in a variety of industrial applications. A number of industrial processes produce substantial quantities of cyanides. There have been cases of accidental release into the environment, a major problem given the toxicity of cyanides. The problem is compounded by the fact that cyanide pollution is difficult to measure - according to the US EPA, only oil and grease pollution are more difficult. Quick and accurate methods of cyanide detection are therefore very important.

The US Environmental Protection Agency has a quantitative test method 1904 "Cyanide in waters and extracts using titrimetric and manual spectrophotometric procedures". However, a quick qualitative test is desirable, to ascertain whether there are cyanide ions present. There are a number of these. Some examples include Yang et al, Org. Lett. 2006, 25, 5721-5723) (hydrogen bonding), Zelder, Inorg. Chem. 2008, 47, 1264-1266 (metal coordination), Cho et al, J. Am . Chem. Soc. 2008; 130, 12163-12167 (bond-forming reactions between the nucleophilic cyanide and an electrophilic carbon ) and Badugu et al, J. Am. Chem. Soc. 2005, 127, 3635-3641 (bond-forming reactions between the nucleophilic cyanide and a boron centre). A general reference "An overview and comparison of methods for cyanide analysis" was presented at the 2009 Pittsburg Conference on analytical Chemistry and applied spectroscopy in March 2009.

More recently, there has been a method utilizing cobalt-containing corrinoid complexes, that is, compounds with this corrin skeleton

and with cobalt as the metal ion. These have proved effective as cyanide ion detectors The affinity of cyanide to corrinoids has been known for some time (Mannel-Croise et al Inorg. Chem. 2009; 48, 1272-1274 and in PCT publication WO 2012/136793). While this method has proved interesting, there have been problems arising with the use of buffer solutions to maintain a steady pH, ideally at around 9.0-10.0, necessary for correct implementation of the Mannel-Croise method, are mentioned below. Typical suitable buffers include N-cyclohexyl-2-aminoethanesulfonic acid (CHES) and 4-(2-hydroxyethyl)-l- piperazineethanesulfonic acid (Hepes). The use of these liquid buffers can introduce problems. For example, the resulting dilution can change the sample matrix and lead to false results.

A further problem is that certain metal ions, for example, Co ²⁺ , Fe ²⁺ Fe ³⁺ , Zn ²⁺ , Cu ²⁺ and Ni ²⁺ , can form thermodynamically stable complexes with cyanide. This removes free cyanide from a sample under test. The only way to avoid false results in such circumstances is to utilise other processes, which is inconvenient and invariably slows down the detection process, making rapid detection nearly impossible.

The concentrations of such metal ions are relatively high, as shown in the following table of typical concentrations in plating baths:

(Source: Chiu et al, Water Poll. Control, Vol. 86, No. 1, 12-17 (1987)). The presence of such ions is also a problem in the cyanide-based recovery of precious metals, such as gold, silver and platinum. It is well known that certain ions, particularly copper, that can be present in precious metal recovery processes, consume cyanide. It is described, for example, by La Brooy et al (Minerals Engineering, Vol.7, No.10, pp. 1213-1241 (1994) at p.1218). This reduces the efficiency and therefore the profitability of the recovery process.

It has now been found that these problems can be substantially overcome by means of a revised method. There is therefore provided, as part of a cyanide detection system, a metal

contamination detection system comprising a solid buffer, which pH has been adjusted to reach a value from 9 to 10.

There is additionally provided a method of detecting metal contamination in a cyanide detection system, comprising adding a test sample to a solid buffer prior to testing for cyanide. The use of a solid buffer brings two advantages:

(a) It causes no dilution, meaning that any result is not altered;

(b) Any cyanide-complexed metal ions present will show up as a precipitate, indicating that action need be taken to remove them prior to the actual cyanide test. Any buffer that can be prepared in solid form may be used in the detection system. Particular examples are carbonate (carbonate-bicarbonate) buffer , glycine buffer, AMP (2-amino-2-methyl- 1-Propanol) buffer, AMPD (2-Amino-2-methyl-l,3-propanediol) buffer and CHES, more particularly CHES. The buffer is prepared by lyophilisation of a liquid buffer. The preparation of such a buffer may be performed by any means known to the skilled person. Details of a particular preparation may be found in the following examples, but these are purely exemplary and are not intended to be in any way limiting. The pH that the buffer provides and stabilises is from 9-10. Particularly about 9.5 In use, the sample to be tested is dissolved in water and the solution added to the solid buffer. The formation of a precipitate indicates the presence of metal contamination, and this means that other art-recognised methods must be used to remove the metals prior to testing. The method is particularly effective when used in conjunction with the Mannel-Croise corrinoid method hereinabove defined. The Mannel-Croise method is performed using a buffer in the pH range between 9 to 10, meaning that, if no metals are detected, the buffer solution used for metal detection can be immediately tested, something that is not possible with many other cyanide-testing methods. Together the two provide a simple, easily-performed test method for cyanide and metal contamination that could lead to false results in cyanide detection. There is therefore also provided a cyanide detection kit comprising

(i) a metal contamination detection component comprising a solid buffer and

(ii) a cyanide detection component , comprising: at least one first colorimetric solid phase extraction device, said first device comprising a non-polar solid phase loaded with a first corrinoid compound selected from the group consisting of:

a cobalamine according to formula (I),

a cobyrinic acid hepta CI -4 alkyl ester derivative (II), a cobyrinic acid (III) and a cobinamide (IV),

wherein X is CN, OH2, OH, halogen (CI, Br, I) or CH3,

means for passing a gaseous or liquid sample over or across said device, and means for detecting a colour change of said first corrinoid compound.

Particular example of buffers are carbonate, glycine, AMP, AMPD and CHES, more particularly CHES.

In operation, a sample to be tested for cyanide is added to the dried buffer. If interfering metal ions are present, they complex with the buffer and a precipitate forms. The nature and form of this precipitate differs, depending on the nature of the metal ion. However any precipitate formed clearly indicates the presence of interfering ions and that the sample should be treated to liberate cyanide or select different testing procedures, as all metals shown to precipitate with the buffer also complex cyanide.

The method hereinabove described provides a quick and reliable test for metals that can complex cyanide and give a false result. This is an indication either that removal of the offending ions is necessary prior to retesting or choosing an alternative testing and sample preparation method. In the case of precious metals recovery, it can indicate an incomplete result, that is, that there has been an incomplete recovery of the metals from the ore, indicating the possibility of recovering more metal from the ore prior to disposal, and therefore less waste and a more economical result.

The disclosure is further described with reference to the following non-limiting examples. EXAMPLE 1

Preparation & Employment of solid buffer

The buffer solutions [Carbonate, Glycine, Boric Acid, AMP, AMPD, CAPSO, NaOH(as reference), CHES] were prepared in a concentration of 0.5 M and 0.25M in water and its pH is set to a value between 9.0 and 10.0 by addition of hydrochloric acid or sodium hydroxide solution. The liquid buffer was filled into plastic vials and dipped into liquid nitrogen, before the vials were closed with perforated lids. The vials were put into a lyophilizer at -83° C and a pressure of 0.08 mbar. After 12-24 hours, the buffers have dried and were ready for use.

Prior to use, 0.3 ml of the water sample to be tested for cyanide and any potential interferences is added to lyophilized buffer. EXAMPLE 2

Testing of interfering metal ions

The presence of selected metal ions was simulated by the addition to water of 30, 50, 80, 100, 200, 250, 300, 500 and 1000 ppm of the metal ions shown in the following table. This provided the limits of detection shown in the table. limit of detection (ppm) for each buffer system with 0.25 M

Similar results were achieved with a higher buffer concentration of 0.5M

To a sample of each of these metal ions in water was added potassium cyanide to give a CN ^" concentration of 1.0 ppm. These samples were then added to the dried buffer prepared in Example 1. In each case, there was a coloured or white precipitate, indicating the presence of a metal that had complexed with the CN ^" .

Had a cyanide-containing sample, also containing any one of these metal ions, been tested by corrinoid-based cyanide detection assay, the cyanide would not have shown up and the tester would have falsely believed that no cyanide was present.

The test indicates that a metal contaminant is present, and that it is necessary to remove it prior to testing for cyanide.