When wood is used for construction, furniture or other objects, the material is exposed to water in the form of liquid water (rain) or water vapour (air humidity). Some of this water will be taken up by the wood material. This water markedly changes the wood properties. Therefore, wood scientists study how fast wood absorbs water and how much water wood absorbs (and how much wood a woodpecker would peck) so that they can predict the performance of wood in different environments.
In recent years, wood scientists have used automated sorption balances (also known as dynamic vapour sorption or DVS instruments) to investigate the uptake of water vapour in wood. Basically, these instruments are very fine scales enclosed in a chamber that controls the temperature and air humidity around a small sample mass (typically 5-20 milligrams). You can use a DVS to measure the uptake of water vapour as you change the relative humidity by observing the mass increase with time. Initially, the mass change is relatively fast, but over time it becomes slower and slower; if you waited long enough at constant temperature and humidity, the mass would eventually reach a constant level. We call this situation where the mass change stops “equilibrium”.
If you performed these measurements, you would find that it is almost impossible to achieve actual moisture equilibrium, because the resolution of the balance is very fine, fluctuations always occur, and the rate of absorption decreases with time.. In most cases, scientists stop measuring when the rate of mass change falls below a certain threshold level. Moisture equilibrium is not achieved this way, but previous researchers have assumed the final mass to be ”close enough” to equilibrium. We recently showed that the error is too high if we use the commonly applied threshold for the rate of mass change. We suggested researchers needed to wait longer at constant humidity to reduce the error and get closer to equilibrium. However, our suggestion has unexpected problems. If you extend the waiting time, you are more likely to have issues in your measurements from vibrations (e.g. people bumping into the instrument) or unstable environmental conditions (e.g. temperature in the lab fluctuates more than the instrument can handle).
We have worked with automated sorption balances for more than a decade. We are annoyed that the research community has not resolved a fundamental issue such as how to use these instruments to get good data. In October 2021, we therefore initiated a large study covering 20+ labs across the world with automated sorption balances of different brands. We want to help people improve the way they use these instruments by finding the right balance between instrument measurement time and adequate waiting time to get close to moisture equilibrium. We produced vast amounts of data. Surprisingly, we discovered a range of measurement errors that no one had discovered before. We now have new insights on how to use these instruments to get high quality data.
Keywords: moisture equilibrium, measurement time, uncertainty, DVS, sorption
Authors
Emil Thybring
University of Copenhagen, Denmark
Samuel Zelinka
Forest Products Laboratory, US Forest Service, WI, USA
Samuel Glass
Forest Products Laboratory, US Forest Service, WI, USA
Natalia Farkas
Forest Products Laboratory, US Forest Service, WI, USA
Maria Fredriksson
Lund University, Sweden
Login to download the PDF