6 SUMMARY AND CONCLUSIONS

THE TENDENCY of microclimates in the vicinity of wood and wood products to corrode lead can be measured by following the weight gain of standard test coupons with exposure time at a temperature of 50�C and a RH of 80%. It can reasonably be assumed that for all tested materials the amount of metal oxidized per unit surface area is directly proportional to the square root of the exposure time.

Corrosiveness of atmospheres in the vicinity of wood products varies significantly with the type of wood product but is not correlated with the pH of the wood product. In the worst case, corrosion was accelerated by a factor of 46, compared to the average corrosiveness of the laboratory air under test conditions. The least offensive material accelerated corrosion by a factor of 11. Corrosiveness seems to increase slightly with the time between sample preparation (milling) and the corrosion test. Treatment and storage effects on corrosiveness need further systematic investigation.

The results of this study cannot be generalized with regard to wood species, product, length of storage, or other independent factors. Testing of individual batches of materials is preferable to any “rule of thumb.” The procedure used in this study seems reliable, simple, and fast enough to allow routine tests of potential construction materials.

Corrosion rate measurement by the method described here is less intrusive than an estimation of corrosiveness by measuring atmospheric composition, since no appreciable amounts of trace gases or vapors are adsorbed or consumed due to reaction. According to previous reports (Clarke and Longhurst 1961), lead corrosion is significantly accelerated by acetic acid concentrations as low as 0.5 ppm. To detect such low concentrations using gas analysis, sample volumes of 20 liters and more are usually required (Devorkin et al. 1972; Budde and Eichelberger 1979). Corrosion rate measurement with lead coupons can be carried out in air spaces with a volume of less than 1 liter.

All organic materials can be expected to decompose at some rate, while releasing organic acids. This has been shown not only for wood products but also for various types of coatings (Donovan and Moynehan 1965). The atmospheric corrosion test proposed here is a reasonably fast and reliable way of testing storage and display systems. The tests would be particularly convenient if weight gain measurements were carried out continuously, using an automatic balance in the test environment.