This disclosure relates to the detection of cyanide and to methods for doing so.

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. A number of spectroscopic methods have been described for the qualitative analysis of cyanide. Some recent examples for the detection of cyanide in blood are described in papers by Blackledge et al (Anal. Chem. 2010, 82, 4216-4221), Zelder et al (Anal. Methods 2012, 4.9, 2632-2634) and Swezey et al (J. Analytical Toxicology, 2013, 37, 382-385). However, these do not give quantitative measurements, which must be done by the known laborious methods.

It has now been found that it is possible to make a rapid quantitative determination of cyanide by a quick and reliable method. There therefore provided a method of determining the proportion of cyanide present in a sample, comprising the following steps:

(i) Add the sample to be analysed to a sensor molecule that is selected from cobryrinic acid hepta Cl-4 alkyl esters of the Formula I

in which X is CN, ³ is H, and R ¹ and R ² are OCH ₃ ;

(ii) Subject the sample to UV-vis spectroscopic analysis in the range 450-700nm;

(iii) Determine the free cyanide concentration from the equation (i) C=(A-0.058)/0.104 (i) in which C is the free cyanide concentration and A is the ratio the ratio of the absorbances at 581 and 527 nm, the equation (i) having been derived from a calibration curve that is a plot of A ratios calculated at 0-lmg/L CN ^" with a constant 46 nmol of sensor compound.

The spectroscopic analysis may be carried out using any of the standard types of spectrometer commercially available, for example, the Cary™ 50 spectrometer ex Agilent Technologies. The analyses are carried out in quartz cells with a path length of 1 cm and the UV-vis spectra in the wavelength spectrum between 450-700 nm were measured at T= 22 ±1°C.

The calibration curve was prepared by adding portions of an aqueous 50 mg/L solution of CN ^" to a solution of the sensor compound (I) in water, buffered to pH 9.5 with N-cyc!ohexyS-2- aminoethanesuifonic acid (CHES), starting at 0 mg/L CN ^" and moving in 0.1 increments to l.Omg L CN ^" . The spectrum was taken at each wavelength, the ratio A computed and this ratio plotted against the cyanide concentration (Figure 1). This curve can then be used to provide the cyanide concentration of an unknown sample.

The sensor molecule of formula I exhibits two absorbance maxima in the 450-700 range at 497 nm and 527 nm. These maxima undergo a bathochromic shift to 538 nm and 581 nm upon binding of a cyanide moiety. The cyanide concentration was calculated using a ratio of absorbance: Abs max of the cyano derivative (581) over Abs max of the sensor compound (527 nm). The absorbance at 581 nm is not sensitive to interferences, while the absorbance at 527 leads to a linear correlation. The use of this ratio allows for better accuracy by removing the concentration dependence (Lambert-Beer law). The ratio of the absorbance values at the wavelengths of about 581 and about 527 nm is then introduced in equation (i) to obtain the free cyanide concentration.

The use of the ratio A, as opposed to the simple absorbance value, is a particular feature of this disclosure. The absorbance is very sensitive to small changes, for example, in the sensor concentration, giving an inaccurate result. The use of the ratio, in conjunction with the particular sensor molecules, overcomes this by removing concentration as a problem. This provides a method that allows a particularly accurate measure of cyanide concentration.