Introduction

The use of wood is limited by its susceptibility to organisms that discolour and degrade wood. If wood is wet for long periods, wood-degrading fungi will attack it. Thus, with the increase in competition between different building materials, it is important that the durability of wood can be guaranteed for longer periods. Due to the length of service life trials, accelerated durability tests of wood are necessary in order to test new paints and products. Two of the demands that are placed on a good accelerated test for wood durability are (1) that the same degradation processes that take place in reality also occur in the test, and (2) that the method is relatively fast and reproducible (ISO 2001).

Several tests for estimating durability have been carried out in ground contact where decay usually is faster due to the higher moisture contents and the direct colonisation of fungi from the soil (CEN 1989). The problem with these tests is that they are very sensitive to where in the soil the test specimens are put. Even in the same test field one can get different results, as it is only the fungi in the vicinity of the stake that are able to degrade the wood (Nilsson and Daniel 1990). Moreover, while these tests may only be of relevance to wood used in ground contact, they can be applied as an indicator of durability of wood in above ground use. In above ground use, wood is subjected to another type of condition than in the soil. One important factor is that wood above ground has a better chance of drying out between wet periods.

Above ground testing is more relevant for testing wood, which is intended to be used in above ground situations. Above ground, field tests take years to carry out. In order to speed up the experiment, different moisture traps are often used (CEN 1993, 1996b; Johansson et al. 1999). Due to the moisture traps, the moisture content rises, and the wood becomes more susceptible to attacking fungi. But, as in all field tests, it must be noted that the present weather conditions are of importance for the test result (Sell and Zimmermann 1995; Creemers et al. 2002)

Accelerated laboratory tests are often performed under sterile conditions with rot fungi in pure culture. Note that only one type of fungus was used at the time. Since only one fungus was used, there was no interaction between fungal species. However, under natural conditions, fungal species interact and their influence can be positive, negative and even symbiotic (Bärlund 1950). Furthermore, in most of these tests, the fungi are allowed to grow on media in a vessel before the wood is added (Rennerfelt 1947; CEN 1994, 1996a). As a consequence of this test design, the fungus has all the benefits of an excess of nutrients and moisture. Thus, this kind of test only measures the toxicity of the wood extractives or preservatives. Because the wood is not allowed to dry out at any time during the test, the influence of the characteristics of its moisture dynamics is eliminated. These tests can be used for testing preservatives, but they must be considered as being too limited for estimating the above-ground durability.

Experiments in the laboratory with non-sterile soil have been used in different variations, for example, in soil-bed and fungal cellar methods (Edlund 1998; Larsson Brelid et al. 2000; Molnar et al. 1996; Stephan et al. 2000). These tests have the benefit from the action of a natural microcosmology as a part of a natural ecosystem. The disadvantages inherent to such texts are that they are not easily reproducible since the same soil cannot be used twice, and the wood is often not allowed to dry out at any time during the test period.

Even fairly small changes in relative humidity or temperature in the surrounding environment change the average moisture content of wood specimens (de Meijer and Militz 2001; Time 2002). The amount of water uptake depends on inherent factors in the wood, for example, its chemical composition and density. Rot fungi only degrade wood if there is enough water available. For this reason, the moisture characteristics of the wood are of importance for the above-ground durability.

Mould fungi growing on wood are not so dependant on the moisture content in the wood, as they are on the relative humidity in the surroundings. The lowest relative humidity allowing mould growth on Scots pine and Norway spruce has been shown to be about 80% (Viitanen and Ritschkoff 1991). The more hygroscopic the material is, the lower RH is needed (Block 1953). Since mould fungi only grow on the surface of the wood, the wood does not loose its strength due to the attack (Dinwoodie 2000). However, an attack of mould fungi indicates that there are nutriments available at the wood surface (Block 1953). Furthermore, mould can increase permeability and retain moisture on the surface of the wood. Thus, a mould attack can indicate, and even accelerate an oncoming attack of rot fungi.

In accelerated short-term exposures, it is important to combine and compare them with other experiments; for example, in-use condition tests (ISO 2001). In this first trial with the new Mycologg method, Scots pine replicas from an above-ground exposure experiment at Växjö University in Sweden were used. The aim of this paper is to describe a new accelerated durability method, and to compare a number of experimental results with results from a field test.

Materials and methods

A pilot study was performed where Scots pine (Pinus sylvestris) heartwood that was exposed in the Mycologg for 10 weeks was compared to samples exposed in an outdoor field test in southern Sweden for 2 years. As references, Scots pine sapwood samples were used. All samples were kiln dried together (maximum drying temp 61°C), and the surfaces were freshly sawn before testing. Biological discolouring was graded according to Table 1 in the trials. In addition, a microscopic analysis was made on selected samples that were exposed in the Mycologg method.

Table 1 Classification of attack by discolouring fungi
Full size table

The above-ground testing was performed according to Johansson (1999) at the Asa Forest Research station, Sweden (Lat 57°). Three pieces, sawn to size 22×95×500 mm, were screwed together with eight wood screws so that two pieces of wood were joined together lengthwise with a third piece of wood which overlaps the other two (Fig. 1). Fifty-four heartwood and 27 sapwood samples were made. All of the pine heartwood samples came from different trees, and each had a corresponding sapwood sample from the same tree when possible (Many trees did not have enough sapwood to make the necessary samples). All of the samples had horizontal annual rings and were screwed together so that the pith side of the wood was facing the back. The trees came from different stands, and they had different diameters so as to increase the spread among the samples. The annual ring width and the presence of knots were measured in all samples. The outdoor exposure was performed on racks with a 60° inclination to the South. The lower part of each sample was 500 mm above the ground surface. The ground surface was covered with coarse gravel and free from vegetation. All of the samples were weighed each month, and the moisture content (MC) was calculated. The samples were checked for biological activity on their rear side on three occasions during the first 2 years. While only examine the rear (north) side of the sample the eventual effect of UV light on the surface was minor. The temperature and precipitation data were provided from the weather station at the Asa Forest Research station, Swedish University of Agricultural Sciences.

Fig. 1
figure 1

A schematic picture of how the three pieces of wood (22×95×500 mm) in the outdoor experiment were joined together and exposed (Johansson et al. 1999)

Full size image

The samples that were exposed in the Mycologg test consists of 55 heartwood and five sapwood samples. The heartwood samples came from 11 different trees, which provided five replicas of each sample. The trees came from three different origins. All of the sapwood samples came from a single tree. The samples were sawn with horizontal annual rings to 150×20×70 mm (±2 mm) and then sealed on their sides and end grain with silicone (Bostic) to prevent moisture from penetrating. The samples were exposed in the Mycologg for 10 weeks. The Mycologg was originally designed by the company Mycoteam A/S in Norway, to test the durability of paints against discolouring fungi. But it can also be used for estimating the durability of wood panels. It tests the paint’s or the wood’s inherent ability to resist moisture and fungi. The main purpose of the test is to monitor both fungal attack and moisture fluctuations in the panels at the same time, and to monitor the moisture content in the panels as they are subjected to different RH (relative humidity). The equipment consists of three modules: the moisturiser that produces the mist; the three chambers for the 60 test panels; and the control unit containing the computer software and hardware.

The moisturiser is connected to a water pipe with a water filter, which cleans any impurities that may be present in the water. There are two tubes leading in the mist from the humidifier in the short side of the first chamber, and two tubes leading out the mist from the opposite short side (Fig. 2). The three chambers are connected in series, so that the outgoing tubes from the first chamber are connected to the second chamber. The outgoing moisture tubes from the last chamber are connected to a drainage system.

Fig. 2
figure 2

A picture of one of the three Mycologg chambers (365×200×470 mm) with the nearest long side removed

Full size image

The three chambers (dimensions: 365×200×470 mm) are constructed in aluminium and Plexiglas. Each chamber holds 20 test panels, ten in each aluminium frame that makes up the long sides. The test panels are 150×20×70±2 mm in size, according to the European standard EN 927-5 (CEN 2000). The front of the panels faces the inside of the chamber (Fig. 2). Each panel has a sensor on the back, which measures the moisture content of the panel by using differences in electrical resistance. In this trial, the sensor was inserted to a depth of 3 mm (from the front) (Fig. 3). In addition, each chamber has an RH/temperature sensor, which is connected to the control unit. The sensors are connected to the control unit where all values are calculated and stored.

Fig. 3
figure 3

The outside of the Mycologg chamber. The moisture sensors on each test panel are connected to the computer

Full size image

The back of the panel faces the outside, and is thus exposed to the prevailing climatic conditions in the room. If the climate in the room fluctuates too much it will affect the panels. It is therefore optimal to keep the chamber in a climate-controlled room with a constant RH. In this trial, the room temperature was about 20°C and RH approximately 50%.

The control unit consists of a computer running Fuktlogg (Version 6.3). This application registers the moisture content in the test panels as well as the RH and temperature in the chambers every 30 min. The unit also controls the activation of the humidifier in time increments that are set by the operator. In this trial, the pine heartwood was subjected to 100% RH for 72 h (the mist was on), followed by a dry period of 72 h at approximately 50% RH (the mist was off). This cycle was repeated for 10 weeks. This provides good conditions for the growth of discolouring fungi, tested in these trials.

Since the test was performed under non-sterile conditions, any fungal spore present would have had the opportunity to grow. As a further add-on in this trial, a spore solution of Aureobasidium pullulans (CBS 24965) was sprayed on the panels at the beginning of test period. The biological activity was visually measured on three occasions.

Results

Fungal discolouration

After 10 weeks in the Mycologg, the average grading of the fungal growth of the heartwood samples was 0.98. Of the heartwood samples, 42 of the 55 had visible fungal growth (32 were graded “1”, eight were graded “2” and two were graded “3”) (Table 2). Of the 13 unaffected heartwood samples, six were duplicate samples from the same tree. After 26 months of outdoor exposure, the 54 samples had an average discolouring grading of 1.43, which is slightly higher than the grading observed after 10 weeks in the Mycologg. The fungal growth on the backside was most prolific around the joint, showing that the moisture trap had an effect.

Table 2 Sample summary with discolouring indexes
Full size table

All of the sapwood reference samples were graded “3” after 8 weeks in the Mycologg. After 26 months of weather exposure, the average sapwood grading was 2.74. On samples run in the Mycologg, three sapwood and three heartwood samples were examined with a microscope. Several species of mould and blue stain fungi were found on all of the samples. Furthermore, on some sapwood samples there were signs of soft rot attacks near the surface, proving that also rot can develop in the Mycologg. Since the Johansson samples are part of the ongoing project and are still at the field, no microscopic inspection on these samples was performed. However, after a thorough visual examination it is obvious that there are several different species of fungi that is responsible for the discolouration on these samples too.

Moisture contents

The MC’s of the heartwood samples are unsurprisingly low and quite stable when compared to the sapwood samples. This observation holds for samples both in the Mycologg and in the outdoor weather exposure. The average MC’s are presented in Table 2. It is clear that the weather-exposed samples have higher average MC, and that there is also a larger difference between the heartwood samples and the sapwood samples. Note, however, that the MC’s were not assessed by the same technique in the two tests. The weather-exposed samples were weighed once a month, which provided an average MC for the whole sample. The MC of the samples in the Mycologg was measured by electrical resistance at a distance of 3 mm from the exposed surface.

The outdoor method, with its moisture traps is very demanding. This is clearly apparent when one inspects the moisture patterns (see Fig. 4). The MC of the sapwood samples is almost always above or close to the fibre saturation point (approximately 30%). During the autumn and winter, the sapwood samples reach very high MC’s. Furthermore, the MC for the heartwood samples was frequently over 30%, but displayed a lower average and a more stable MC. The total precipitation (per month) and average temperature (per day) during the test period is also shown in Fig. 4.

Fig. 4
figure 4

The moisture content of a typical heartwood (X) and a typical sapwood (filled square) sample during 2 years of weather exposure. The temperature (black line) and precipitation (grey line) are displayed as a 7-day moving average

Full size image

In Fig. 5, the MC dynamics of a typical heartwood sample (lower curve) and a sapwood (upper curve) sample are shown during 10 days in the Mycologg. The heartwood sample’s MC increases when the humidifier is turned on, but not to the same degree as the sapwood sample. The sapwood sample reaches over 30% MC, measured 3 mm from the surface.

Fig. 5
figure 5

The moisture content of a typical heartwood sample (lower curve) and a sapwood (upper curve) sample during 10 days in the Mycologg

Full size image

Discussion and conclusion

A prediction of service life can be based on data from in-use conditions, but very often one needs an accelerated test to obtain results more quickly. Accelerated tests are limited since they focus on specific applications only. It is not possible to simultaneously accelerate each and every agent that contributes to ageing. Consequently, one has to carefully decide on the most important process (ISO 2001). The Mycologg method makes possible a specific study focused on the discolouration and degradation of fungi in conjunction with forced water cycles. Compared with other studies, this method has the benefit of monitoring the fungal attack simultaneously with the changing water dynamics in the test panel.

The fungal discolouration after 10 weeks in the Mycologg test corresponds to about 1–2 years of weather exposure. Because an outdoor test is very unpredictable and dependent on the weather, it is not possible to determine an exact corresponding time. Note too that the outdoor field test described in this paper is also somewhat accelerated because of the presence of moisture traps and unprotected end grains in the samples, thereby making the MC higher and the samples’ affinity to fungi even higher. The 10 weeks in the Mycologg with the moisture loads described above should perhaps correspond to a much longer time period than 2 years of normal use.

The electrical resistance in the Mycologg was measured at a position of three mm from the surface of the sample in this trial. This means that the average MC of the whole sample is much lower. It is merely higher close to the surface. The pins that measure the MC are adjustable, so it is possible to measure the MC at a different depth if it is so desired. The moisture patterns found in the Mycologg trial are similar to the patterns found in the weather exposure, with high and fluctuating moisture patterns for the sapwood samples, and low and stable patterns for the heartwood samples.

The moisture load in the Mycologg was enough to achieve substantial fungal discolouration to a level corresponding to about 2  years of weather exposure. However, the moisture load cycles used in the present study may not be optimal. The time intervals of the moisturiser must be carefully designed so as to provide data that is required in accordance with the aim and the range of the study. In this study, soft rot attacks were evident near the surface on some sapwood samples, but in order to test more than discolouring fungi, the exposure time must be prolonged and also the moisture load must be higher. Also, while testing durability of paint, for example, longer moisture cycles are required in order to make the wood sufficiently wet. The frequency of the exposure to moisture can also be altered in many ways.

With respect to evaluations, they should be made at narrow intervals. In addition to the results stored in the computer, several visual inspections must be made throughout the test period. To decide when the test panels have reached the point of pass or fail, it is useful to include one or several reference panels in the test. In this first trial, Scots pine sapwood was used, as it is known to have a lower durability against fungi than heartwood. The reference samples can be compared to the test panels so as to determine the point between failure and non-failure for the samples that are investigated.

The test has its disadvantages since it is restricted to water uptake in no other direction than through the front of the test panels. In addition, since the test is performed during non-sterile conditions, the number of, and the different species of fungal spores is unknown. This can also depend on season of the year when the test is performed. At the same time, it is a more natural condition then using monocultures. One way to reduce this problem is to add a specified spore solution. The spores in the solution are provided an advantage over the naturally existing spores, because they are sprayed on the surface in a (preferably) high concentration. The solution can be chosen according to a number of different preferences, for example selecting the fungi which is know to cause problem on the tested material.

Another disadvantage is the lack of influence from weathering. This can be solved by running the same test samples in a weather-o-Meter (Blümer and Nussbaum 2001) that efficiently subjects samples to cold, warmth, rain and UV light. The controlled water cycles in the Mycologg make the method reproducible. Thus is demonstrates that this method can be a useful tool for estimating the durability of wood. As was shown in this study, 10 weeks in the Mycologg corresponds well to a number of years of outdoor exposure with respect to discolouring fungi. The Mycologg provides for a more natural contamination with a variety of different rots and moulds, when compared with many other methods. It is also a good method for monitoring moisture fluctuations in panels with fungi contamination so as to investigate how the panels respond to different RH cycles.