While there are many arguments about chromium in leather, there are certain issues that can be accepted to operate in the international supply chain:

* Chromium III is different from chromium VI

* Chromium III is not toxic but chromium VI is

* There needs to be an accepted test method and agreed limit to enable international trade to occur

At the present time, there is much discussion in different organisations within the European Commission and these are at different stages of development.

Previous results have been reported which involved spiking trials of crust leather with potassium chromate, to give various concentrations of Cr VI within the leather. The research indicated that leather has a reductive capacity and that between 4-40g of Cr VI can be reduced per gram of leather (when spiking at 50g/g).

There are also many BLC members who are frustrated by chromium VI contents varying:

* In reports from test house to test house and

* From the same sample at different times

Recent analytical investigations at BLC confirm that these frustrations are based upon variability inherent in the test method. It can be seen that the initial results of 5.3mg/kg decay to 0.6 after four weeks. In some cases, the opposite has been reported.

This data does not confirm that chromium VI is either being created or destroyed but that the repeatability of the method is unreliable at the levels of 3mg/kg that are currently operating.

The test method

There are several methods for determining chromium VI in leather but they are all based upon aqueous extraction and subsequent development of colour with diphenylcarbazide. The colour is read by spectrophotometric means. The extraction can be influenced by colour leaching from the leather; the use of Dionex cartridges can assist in clearing colour but nevertheless colour is an influencing factor.

This test method is referred to in Vogel ‘Text book of quantitative chemical analysis – fifth edition, 1989’ as a primary method, so we have no reason to consider the method invalid. It is understood that the method is based upon the oxidation of the diphenylcarbazide to the carbazone due to chromium VI (which converts to chromium III); the chromium III reacts with the diphenylcarbazone to form a pink colour.

In the matrix (aqueous extract), there are many chemicals that could conceivably influence the oxidation of diphenylcarbazide, and there is an excess of chromium III in leather extracts that could create the pink colour. It is necessary to ensure that a limit of detection is set such that this possibility of a false positive is eliminated.

The limits of detection for different methods

Vogel implies a range of detection for the diphenylcarbazide method of 0.2 to 0.5mg/l in an analytical solution. It should be noted that this limit applies to clean solutions in perfect analytical matrices in the absence of the many potential contaminants from leather. Hence, this is the implied limit of detection of chromium VI in a test solution. However, when carrying out analysis of leather, extraction of the chromium from the leather matrix is required.

The method IUC 18 requires 2g of leather to be extracted in 50ml of test solution (phosphate buffer) and the resulting solution is then analysed. Due to this extraction and dilution, the range of detection of 0.2-0.5mg/l chromium VI in solution actually implies a range of detection from 5-12.5mg/kg based on the leather.

IUC 18 and DIN 53314 are both methods for determining chromium VI in leather based upon the diphenylcarbazide method. Both indicate that 3mg/kg is the limit of detection. The CEN standard for chromium VI in leather (CEN TS 14495) was published in April 2003. This method states ‘In the case of levels of chromium VI being calculated at <10mg/kg the following shall be reported: According to the analysis carried out, chromium VI was not detected in this sample.’

It is important when considering the requirement for setting a limit of detection that the method is ‘capable’, ie able to accurately reproduce results, and that commercial laboratories can achieve the accuracy required to ensure that pass/fail criteria are equitable for both parties.

Repeatability and reproducibility

Based upon inter-laboratory trials carried out with EU test houses, during the preparation of pr CEN TS 14495, the repeatability and reproducibility are shown in Table 1.

The standard deviations for the repeatability and reproducibility are given in Table 1. It is clear from this data that repeatability within each laboratory is of the order 1.7mg/kg and between laboratories of the order 2.5mg/kg.

BLC’s own laboratory repeatability has been recently investigated. Table 2 identifies the variance.

While Vogel claims a range of detection of around 0.5mg/l in simple solution, for the more complex matrix of leather aqueous extracts in the real world the accuracy is much less. The reasons for this cannot be easily determined but the fact remains that repeatability (ie accuracy) is far greater than the current limits of 0.1mg/kg which can be required from some of the brands.

Factors affecting the limit set

The criteria for setting limits in materials should be based upon the risk to the manufacturer and consumer. The inclusion of chromium VI in the criteria is not challenged.

The methods that are well known and commercially available (in DIN, IU systems, EN and Japan) are all based on the same chemistry so there is no point in challenging the test method.

There are many studies in the public domain which consider the effect of chromium VI on the human body. CEN BT 132 is considering this issue at the present time. The concentration at which skin is sensitised to chromium VI is very unclear. Data sources indicate that levels from 1-1000mg/kg might be the ‘no effect limit’. The percentage of the population sensitive to chromium is 0.5%.


BLC supports the broadly held view that the limit of 0.1mg/kg, is technically not achievable on leather extracts.

The current requirements set out in DIN 53314 and IUC 18 (3mg/kg) are currently in use but give rise to inter-laboratory disagreement, because the limit is set too close to the inherent error levels (repeatability and reproducibility) in the method.

A limit of 3mg/kg might be justified but BLC believes the limits set out in the CEN TS 14495 Determination of chromium VI content in leather of 10mg/kg is both technically correct on analytical considerations and also on a risk assessment basis.