1. General
Avoiding skin exposure to chemicals is the best way to control dermal exposure and to prevent occupational skin disease. Protective clothing and gloves can reduce the residual risk. The safe protection time is determined by three processes:

While there are adequate performance standards and testing standards for penetration of chemicals through skin protective material, and for degradation of elastomeric membranes, it has to be stated that permeation is not always handled and tested in an appropriate manner. Based on a research project for the German Work Insurance, and on several projects for industry and for authorities, the following points have been worked out.

2. How permeation works
After the chemical landed on the skin it may be solved in the outer surface of the membrane, then diffuse through the layers that form the barrier, and finally be set free from the inner side of the membrane. There it can meet the skin - or the permeation test sampling medium.

2.1 How permeation is determined
A cut-out piece is placed into a test apparatus. The chemical is placed onto that piece and a collection medium is passed by or pressed against the inner surface of the test medium. If any breakthrough occurred then the amount of the permeated chemical is determined by chemical detection. Adequate detection is easy for volatile solvents (FID or PID may be used) and for inorganic acids or alkaline solutions (a pH electrode is appropriate). Detection is tricky for chemicals that are not volatile and not soluble in water (e.g. PAH), or that are reactive (e.g. isocyanates, or some aldehydes). In these cases a solid medium may be pressed against the inner surface and replaced and analyzed in regular intervals. This may be a solid medium that was impregnated for giving stable derivatives of the chemical in test. Wetting the sampling medium will improve the adhesion and increase the transfer of any contaminant from the inner side of the membrane to the sampling disc. There is the hypothesis that solutions of these non-volatile and non-water-soluble, or reactive, chemicals are easier to monitor just by measuring the permeation of the solvent. This was confirmed by a number of comparative studies, especially on active ingredients of pesticides. There is another hypothesis that large molecules will not pass through protective barriers made of elastomers. But when applying the solid sampling technique for the permeation test, this assumption showed to be wrong for Polycyclic Aromatic Hydrocarbons (PAH). Permeation can be described as the permeation rate PR (the flow through the barrier) and as the normalized breakthrough time BTT. In European standards, BTT is the only criterion that is accepted for assigning protection factors to PPE and gloves. Normalized BTT is the time that goes between the start of the test and the point of time when permeation exceeds the threshold of 0.1 or 1 mg/(cm² x min) (see ASTM 739 resp. EN 374-3)

2.2 Conditions in test and conditions in service
Several parameters of the standard permeation tests as described in the testing standards ASTM 739, ISO 6529 and EN 374-3, differ from the conditions that are likely to be found in service.

2.2.1 Duration of exposure
Testing is done as continuous contact over 8 hours. In reality, most dermal exposures are intermittent or occasional. On the other hand, many gloves are re-used over several days and weeks. If no permeation occurs after 8 hours, nobody knows what will happen after (e.g.) 3 days. On the other hand, in the case of intermittent exposures to volatile chemicals a barrier may serve much longer than a standard eight hour test will predict - especially if evaporation is faster than permeation.

2.2.2 Temperature
Testing is done at 23 °C. In reality, the PPE or the glove will reach a higher temperature at the inner side due to body heat. This may shorten the breakthrough time and increase the permeation rate dramatically, giving worse protection and a shorter safe protection time than the standard test will predict.

2.2.3 Stretching
Testing is done without any mechanical challenge. In reality, the PPE or the glove will be stretched by movements. For gloves, closing the hand may lead to 20 % stretching or even more, resulting in a thinner membrane, at the knuckles. This was shown to lead to a shorter breakthrough time, but the impact of stretching is less dramatic than that of the elevated temperature.

2.3 A new approach
In Germany, a test procedure was designed that uses the standard test cell but with a simulation of in-use conditions (35 °C inside glove temperature and 20 % length stretching), as well as of short-term exposure and re-use if relevant. The new testing protocol was applied to 5 chemical products containing volatile organic solvents. A total of 19 protective gloves were tested and under these conditions the breakthrough time was ½ or even 1/3 when compared to the respective standard test results. This proved that the official testing standards are reflecting insufficiently the in-use conditions. Elevated temperature inside the PPE or glove due to body heat was the most important factor. Short-term or occasional exposure was also important for some (but nor for all) of the solvents that were monitored. Mechanical stretching (e.g. due to hand moving) showed to be of minor importance. In a number of projects the solid sampling permeation test method proved to give reliable results for PAH, isocyanates, acrylates, and for compounds of epoxy resins. Had these been tested with the standard testing technique only, then permeation would not have been detected just and only for analytical reasons.

3. How significant are test results?
Not only the testing protocol may lead to a limited significance of the testing results, also the barrier material itself shows variations of the barrier effect.

3.1 Membrane properties and barrier effect
Elastomers may show a barrier effect towards chemicals. The barrier is weak if the challenge chemical is soluble in the membrane or if it may move through the membrane by diffusion - and vice versa. A number of physico-chemical parameters influence these processes such as molecular size, hydrophobic/hydrophilic nature, bonding polarities, hydrogen bondings, Van der Waals forces and more. Many barrier membranes are built from different layers - this complicates a quantitative description and prediction of the permeation process enormously. Thickness of the membrane is another but well known important determinant of the barrier effectiveness.

3.1.1 Nitrile does not equal Nitrile, Latex does not equal Latex
Different brands and qualities of elastomers may have different ingredients and different degrees of cross-bonding. Therefore two membranes made of two elastomer brands, although being of the same type of elastomer, may show very different barrier properties.

3.1.2 Batch to batch variation
Even different batches of the same membrane may show different barrier properties because of variations of the manufacturing process and of the ingredients.

3.2 Analogies between chemical mixtures
Two mixtures containing the same or very similar chemicals may behave in a similar manner in many cases but there are some reports on unexpected observations. Small amounts of chemicals may change the physico-chemical properties of the total product dramatically in some cases. It will need chemical expertise to decide whether a selection of a barrier material by analogy should be verified by a new test or not. In case of doubt it is essential to carry out a permeation test not with the main ingredients only, but with the chemical product as it is.

3.3 A new approach
PPE and glove selection based on the polymer type (e.g. "Nitrile" rubber) can be misleading if the selected membrane shows a different barrier effect than the tested one. This is still more true if a "Nitrile" material is made from both Nitrile and Latex layers. The same showed to be true for Chloropren and for Latex materials. Butyl rubber qualities were reported to be better comparable between each other. Therefore test data should only be used for the PPE that was tested, and for PPE that was made from the same membrane brand and with a similar thickness.

4. Conclusions
Today, selection of PPE or gloves based on published data alone is a guess rather than a sound procedure. Earlier test data are only relevant for the selection of PPE or gloves if the test results refer to the barrier material in question, and if the tests were carried out with a chemical or a mixture that is similar to the challenge chemical. The standard permeation test method needs an update for inclusion of the influence of temperature, stretching, and exposure time patterns, and for testing chemicals that are neither volatile nor water soluble, or that are unstable under the testing conditions.