Formaldehyde - the final straw?

Back in 1916, Fleming² wrote about formaldehyde: 'Within the last 15 years, Roberts patented a process by which skins were subjected to the action of antiseptic gas, that is formaldehyde, for a time in order to preserve them. The Conservadore Mercantile Company of the town, according to this patent, submitted skins after flaying and before salting, to the action of formaldehyde vapors (sic) in closed chambers.' While the formaldehyde may well have destroyed the putrefactive organisms, it probably also caused respiratory damage to the workers, being in a concentration of many parts per million. Even today, there are countries where casein is fixed with formaldehyde, or its 40% solution formalin³, without extraction facilities to reduce the likelihood of the chemical getting into the operators' lungs. And, while these practices should and must be rectified, the majority of the legislation coming via, usually, the automotive manufacturers, sets limits for formaldehyde that are completely unjustifiable in terms of chemistry. They do nothing more than pander to the worries of the consumer. Unfortunately, there is little the tanner can do other than try to meet the limits. In recent times, tanners have seen a dramatic increase in the levels of, as Wood called it, pseudo legislation, ie where there are limits set even though there is no official legislation⁴. Table 1 shows the current limits for formaldehyde in leather, none of which have any legal basis. The ridiculous situation is that while a person may wear leather shoes or garments or ride on leather seats safe in the knowledge that the goods are certified as having no more than 10ppm formaldehyde in them, they are allowed to slap on cosmetics containing formaldehyde at levels greater than 500ppm, or swill their mouths with oral hygiene products containing in excess of 1g/kg formaldehyde. So, where does the 10ppm level come from? As usual the automotive manufacturers are leading the way. In a paper given to the IULTCS Congress in Cape Town, South Africa, earlier this year, Gerhard Wolf, BASF, set out the conditions under which formaldehyde was formed and its avoidance⁵. He started by showing a table of the amount of formaldehyde in the environment (Table 2). The table suggests that eating apples is likely to be more harmful than sitting on a piece of car leather in terms of HCHO content. Measurement Wolf continued by defining the different states in which formaldehyde can be found in leather. 'In order to appreciate the analytical issues involved in measuring the formaldehyde content of leather, a distinction needs to be made between the different states in which the formaldehyde is bound to the leather.' The different forms closely resemble those catergorised for water. A distinction is made between free formaldehyde, reversibly bound formaldehyde and irreversibly bound formaldehyde. He said: 'Free formaldehyde, which is not bonded to the leather itself or to other substances applied to the leather, can be measured by the gas phase method.' This method is derived from a test devised by Audi/VW (PV 3925)⁶, and it has been adopted by many other automobile manufacturers⁷. It is based on the assumption that only the formaldehyde emitted from the leather into the interior of the vehicle needs to be considered. The conditions are fairly mild, three hours at 60ºC, and so it is assumed that only the free formaldehyde present in the leather will be detected and no reactions will take place that cause more formaldehyde to be liberated. The leather is suspended over water in a sealed vessel where the formaldehyde concentration in the three phases - air, water and leather - reaches an equilibrium. The formaldehyde in the aqueous phase is then derivatised, usually with 2,4-dinitrophenylhydrazine (DNPH), and then analysed by separation using HPLC, and UV detection at 350nm. Measurement of 'free formaldehyde emitted into a headspace from typical automotive leathers' has also been investigated by Hurlow et al. He presented a paper from Buckman Laboratories at the 2001 ALCA conference⁸, where he said that typical automotive leathers were prepared and representative samples of the leather were then tested for 'free' formaldehyde. The dynamics of formaldehyde release and a test method for its determination formed the basis of the paper. The analytical method utilised is based on the principle that is common to automotive test methods for released formaldehyde, as discussed above. Hurlow stated: 'This involves subjecting a leather sample, enclosed in a sample vessel, to varying conditions of temperature and time and collecting and analysing the formaldehyde that is subsequently emitted from the leather sample and into the atmosphere. The strategy for formaldehyde capture and detection involve elution of the headspace with nitrogen gas, capture on a DNPH column and detection using HPLC.' Non-gas methods The other types of formaldehyde determinable were discussed by Wolf in his IULTCS paper. He continues: 'The reversibly bound formaldehyde is also measured by the extraction method, and it is extracted from the leather with water. The results obtained encompass the sum of the reversibly bound formaldehyde and the free formaldehyde, and so they are usually not identical with the results obtained by the gas phase method.' Finally, the formaldehyde that is irreversibly bonded to the leather can be analysed by treating the leather with sulfuric acid for example. This should give a total formaldehyde figure and the bound aldehyde can be derived by difference. The difficulties lie in another extraction method, which is specified in DIN Standard 533159. This is mainly used for shoe and garment leather, but it is sometimes used for testing automotive leather. This colorimetric method is based on extracting the formaldehyde and treating the extract with acetyl acetone, which forms a yellow complex, 3,5-diacetyl-1,4-dihydrolutin. This is then measured at 412nm. However, there has been work done that suggests up to 80% of the results are false-positives. While the majority of these were below any levels set, it does not give confidence for an official method to have such wide variation in results.¹⁰ The philosophy of the standard 53315 is that the transfer of formaldehyde follows the sequence 'leather/water/contact with the human skin'. In his paper Wolf lists the limits for formaldehyde, depending on which eco-label is to be achieved. The Öke-Tex standard states that formaldehyde levels should be <75ppm or <300ppm dependent on whether the article comes into contact with the skin; Pirmasen states in their eco-label specifications that a level of <50ppm is acceptable for children, while <150ppm is acceptable for adults. However, on their website BASF state that the Pirmasens eco-label specifications have a level of <200ppm for children and <300ppm for adults.¹¹ The extent to which formaldehyde is liberated depends very much on the reaction conditions. It can be seen from Table 3 that a leather emits much more formaldehyde at higher temperatures and over longer periods. It is, therefore, very important that the prescribed conditions are adhered to exactly in order to ensure that the results can be compared with each other. Sources As was highlighted at the beginning of the article, formaldehyde was also used as a preservative, but one of the major sources of formaldehyde is retanning. As Harper¹² points out: 'With all the other [automotive] specifications achieved, leather in the automotive sector faces new challenges that relate not so much to performance but to health and safety. Concerns, particularly in Germany, have prompted a whole range of chemicals to be banned. Certain diamines that were present in dyestuffs required the development of a whole new range of black dyestuffs with good light and heat fastness. Chlorophenols, used as fungicides, went relatively painlessly, but two other areas have proved to be very troublesome. 'The first was formaldehyde, present in high concentrations in the mild tanning agents used in retanning and in dyestuff fixatives. The normal levels of free formaldehyde were anything between 50 and 200ppm, whereas the specified level was 10ppm or below the so-called baby sucking level. This required complete reformulation of the retanning process to select products that do not use formaldehyde in their manufacture, and to choose dyestuffs that have good enough fastness not to require cationic fixing agents. This does leave some formaldehyde present, in small quantities, as preservatives in the fatliquors; these can be neutralised by polymerisation with reactive proteins.' With the introduction of non chrome tanning agents, again for ill-conceived environmental reasons, it is ironic that the problem of formaldehyde may well have increased. The tanner seems to be caught between a using non chrome alternatives, the best of which can contain blends of aldehydes, and not using the best alternatives because they might produce formaldehyde under conditions unlikely to be achieved in the normal wearing of the item. In another paper at the IULTCS Congress, Tim Allsop13 from the New Zealand Leather and Shoe Research Association (LASRA) discussed how the industry is trying to combat the possible effects of formaldehyde being released. Increasingly stringent limits are being placed on the amounts of analysable formaldehyde permitted in wool-on skins, particularly those that are used for clothing leather or for infant bedding. He said: 'Some effective white tanning agents, notably those based on tetrakishydroxymethyl phosphonium (THPX) compounds, produce some formaldehyde as a byproduct of the tanning reaction. The objective of this work was to identify compounds that can be applied post tanning to remove most residual formaldehyde from the final product tanned with phosphonium compounds.' He points out that the method of extraction and determination of formaldehyde, which he bases on a Japanese method, but which uses a similar procedure of complexation with acetyl acetone as discribed above, may 'hydrolyse remaining tanning material and generate additional free formaldehyde.' Allsop continues: 'Method B of JIS L 1041: 1983 is a colorimetric procedure using acetyl acetone¹⁴ developed for the analysis of resin-finished textiles, but it has been adapted for use on leather and a modification of this method is going through the process of adoption as an International Standard (Draft IUC 19, Determination of formaldehyde content in leather¹⁵). This draft standard incorporates two analytical procedures - a modified version of the colorimetric method of JIS L 1041 and an HPLC procedure suitable for coloured extracts. Both methods use the same process to extract free formaldehyde from the leather.' In his paper, Allsop showed that the treatment of woolskins at the end of wet processing could be used to lower the residual levels of analysable formaldehyde to within the tolerance limits. However, there were some detrimental effects on the wool. He also said that holding THPX tanned woolskins in a wet condition lowered the residual formaldehyde concentrations but, as he pointed out: 'There would clearly be logistical problems in holding woolskin production for any length of time.' Other ways of reducing formaldehyde levels could include the obvious ones such as replacing products that contain, or could potentially contain, high levels of the chemical; modify the processing conditions to reduce the levels, such as that suggested by Allsop; incorporate formaldehyde scavengers; or modify the products that release formaldehyde. My personal suggestion would be to follow the example I saw on a bottle of mineral water: the constituent analysis of the water said the product contained 240mg/l total dissolved solids, 7mg/l potassium, and continued to list them. At the bottom it then added 'subject to seasonal change'. This is an excellent idea. On automotive leathers, there could be a contents list with the rider: 'Formaldehyde content <10ppm for no reason other than we [the tanner] can meet this demand of our buyer - but subject to seasonal variation.' Conclusion The industry seems to be stuck with having to meet unreasonable demands for low formaldehyde concentrations in its products. The fact that these levels can be met is not a reason to have to meet them. The problems will come when tanners are told to use non chrome alternatives as the tannage of choice. These may under certain conditions, which bear no relation to the actual likely usage of the article, produce formaldehyde and hence lead to more problems for the tanner. Thus, is there an 'alternative' alternative? Well, vegetable tannins could make more of a comeback, but no doubt some environmentalist would complain about the suspended solids or the amounts of COD being produced, and have them restricted as well. What is really needed is a radical solution. A bit of common sense in the formulation and introduction of eco-labels would be a good start. But what is really needed is for the leather industry to stand up and fight its corner instead of lying down and acting as a doormat.