Natural, accelerated, and simulated weathering of wood: A Review

Abstract

This review considers three aspects of the weathering of wood – natural weathering, accelerated weathering, and simulated weathering. Natural weathering begins when unprotected wood, such as an unpainted board, is exposed to cycles of solar radiation and rain. Unpainted barns and fenceposts take on a gray coloration and their surfaces may become rough, loosened, or checked with the passage of time. The underlying causes of such changes involve ultraviolet light, the effects of cyclic wetting and drying, and the action of certain fungi. Accelerated weathering tests have been used not only to evaluate the effectiveness of varnishes and paints, but also to aid in the understanding of factors affecting natural weathering. Simulated weathering usually has the goal of quickly and conveniently changing the appearance of fresh wood to give the impression of weathering. This might increase its appeal for various decorative purposes. Information about simulated weathering, though largely absent from the scientific literature, is very much alive in social media. This article considers the science behind all three types of weathering in the light of published accounts.

Full Article

Natural, Accelerated, and Simulated Weathering of Wood: A Review

Marcel Kropat, Martin A. Hubbe,* and Frederik Laleicke

This review considers three aspects of the weathering of wood – natural weathering, accelerated weathering, and simulated weathering. Natural weathering begins when unprotected wood, such as an unpainted board, is exposed to cycles of solar radiation and rain. Unpainted barns and fenceposts take on a gray coloration and their surfaces may become rough, loosened, or checked with the passage of time. The underlying causes of such changes involve ultraviolet light, the effects of cyclic wetting and drying, and the action of certain fungi. Accelerated weathering tests have been used not only to evaluate the effectiveness of varnishes and paints, but also to aid in the understanding of factors affecting natural weathering. Simulated weathering usually has the goal of quickly and conveniently changing the appearance of fresh wood to give the impression of weathering. This might increase its appeal for various decorative purposes. Information about simulated weathering, though largely absent from the scientific literature, is very much alive in social media. This article considers the science behind all three types of weathering in the light of published accounts.

Keywords: Ultraviolet light; Solar radiation; Gray coloration; Roughness; Rainfall; Bluestain fungus; Checking

Contact information: Department of Forest Biomaterials, North Carolina State University, Campus Box 8005, Raleigh, NC 27695-8005; *Corresponding author: hubbe@ncsu.edu

Contents

INTRODUCTION

Weathering of wood, in the context of this article, will be defined as a change in appearance and surface properties of unprotected wood when exposed to outdoor weather conditions. The term also can refer to laboratory conditions intended to predict effects of natural weathering in a shorter time period. It might also refer to treatments intended to impart a weathered appearance to wood. The present review is concerned especially with what happens at the surface of exposed wood, rather than such effects as bulk rotting, fatigue, or any chemical changes that might affect the interior properties of a weathered board. Though one can expect that weathering also will affect wood that has been protected by varnish, paint, or other coatings, such situations are not the focus of this article. Some questions to be considered in this article are as follows:

• What are typical changes that occur when wood is exposed to weather conditions that include sunlight and rain?
• What are the principal causes of changes in such attributes as color, roughness, and the appearance of cracks in wood during natural weathering?
• To what extent can accelerated weathering tests predict the effects of natural weathering?
• In what ways can results of accelerated weathering tests reveal the probable mechanisms responsible for natural weathering?
• What are some cost-effective, generally environmentally benign treatments that might be used to mass-produce wood that meets with the approval of customers who are interested in weathered wood for such purposes as decoration, furniture, or design?

Natural weathering is a process that affects not only cut boards, but also dead standing trees and driftwood (Williams 2005). The appearance of old barns, cedar-shake roofs, and fenceposts can impart a feeling of nostalgia. Such wood, when recovered, often can be sold at a premium price in modern applications on account of its appearance. When one considers the huge numbers of wooden structures around the world that either never were protected by paint, or from which protective layers have peeled away, it is understandable that many researchers would be motivated to find out how and why various changes happen to the surface of the exposed wood. The natural processes are also of interest to companies providing products such as paint and varnishes, the purpose of which is, in part, to protect wood from natural weathering (Panek and Reinprecht 2014, 2016; Turkoglu et al. 2015a,b; Kart et al. 2018). Important earlier reviews of the topic of natural weathering include those of Feist (1982, 1989), Feist and Hon (1984), Kranitz et al. (2016), and Cogulet et al. (2018), among others.

The motivation to employ accelerated weathering protocols is a matter of practicality; one desires information about the resistance of various wood materials to weathering – either in their untreated state or when somehow protected – and in many cases one cannot afford to wait many months or years to find the answers. In addition, accelerated weathering conditions make it possible to specify such conditions as UV radiation attributes, moisture levels, and temperatures. Though accelerated tests can give answers more quickly, the results might not be trusted. Questions may be raised regarding whether the accelerating weathering conditions allow a fair prediction of changes that would have occurred as a result of natural weathering. Fortunately, as will be shown, a great many studies have been carried out in parallel, allowing direct comparisons of matched sets of wood specimens exposed either to outdoor weathering or laboratory conditions. The following are some notable review articles focusing on accelerated weathering methods and their use (Back and Sandström 1982; Feist and Rowell 1982).

Various methods can be used to modify wood in an attempt to give an impression of weathering, often with the goal of increasing its sales price. Genuine weathered wood, say from old barns, is often limited in supply, and a high amount of labor may be required to deconstruct the old structure to recover the wood. Thus, a key goal of those who wish to simulate such effects is to decrease the cost of delivering product to customers who desire a product that has a weathered appearance. The effects of weathering on aesthetic perceptions have been studied by systematic interviews (Zanetti et al. 2003).

Unlike the topics of natural weathering and accelerated aging, the topic of simulating aging, i.e., modifying wood with the intention of making it appear to be weathered, has been addressed sparsely in the scientific literature. Instead, there has been a voluminous output of YouTube videos and social media postings related to the topic. Part of the objective of this review article is to discuss what can be learned from the social media information and the extent to which the observations have a correspondence with the scientific literature.

Various aspects related to wood weathering have been covered in earlier review articles and monographs. The words “wood aging” often have been used in studies focusing on the effects of time, under dry conditions; such effects have been reviewed by Cavalli et al. (2016) and Kranitz et al. (2016). Sell (1975) and Cogulet et al. (2018) discuss wood weathering in the context of requirements for surface treatments, including paints and varnishes, when the goal is to prevent weathering. A book by Zabel and Morrell (1992) covers the broad topic of wood decay, in addition to discussing bluestain issues that occur during natural weathering or wood. Researchers studying the natural weathering of wood have benefited from the excellent monographs of Feist (1982, 1984) and also from a chapter by Feist and Hon (1984). The present review article covers natural, accelerated, and simulated weathering of wood, with attention to governing mechanisms and recent trends.

NATURAL WEATHERING

This section, dealing with natural weathering of wood, will consider three main aspects. First, typical observations will be described, in terms of changes in color, roughness, crack development, and wood properties. Second, the influence of site conditions, such as the direction relative to the sun, will be considered. Third, mechanisms that have potential to explain effects related to natural weathering will be described.

Typical Observations

Color

The most widely noted effect of long-term exposure of unprotected wood to outdoor weather, including periods of sunlight and rain, is a change in wood’s surface color. Commonly reported trends are summarized in Table 1, along with selected references. To summarize, the initial changes in color toward yellower or browner hues have been described as relatively rapid. Such changes are often perceptible relative to unexposed wood within the first days of outdoor exposure. The change in color during this initial period has been described as linear with time (Dunningham et al. 1992). In the longer term, however, the wood becomes less yellow and less red, tending toward a gray color. Whether or not the final stage can be regarded as “darkened” or “bleached” differs from case to case, since wood species have large differences in their initial color before exposure. Wood’s initial color, prior to weathering, also will be influenced by such factors as drying conditions, thermal treatments, and periods of storage.

Yildiz et al. (2011) reported relatively large effects of natural weathering on the color of alder wood that previously had been heat-treated at either 150, 180, or 200 C. The heat-treatment caused progressive darkening of the wood, while the wood became somewhat less red and less yellow. Subsequent natural weathering remarkably lightened the wood and further decreased the red and yellow coloration. Ayadi et al. (2003), Huang et al. (2012a), and Garcia et al. (2014) reported that the color-stability of several types of heat-treated wood during artificial weathering was greater compared to untreated wood. By contrast, Cui and Matsumura (2019) observed decreasing changes in color during natural weathering of Cunninghamia lanceolata (Lamb.) Hook. in the following order: heat-treated at 220 C > untreated > heat-treated at 190 C.

Table 1. Reported Changes in Wood’s Color Due to Natural Weathering

Some of the reported changes cited in Table 1 have been reported by relatively few investigators. Further research may be needed in those areas to back up the findings and to establish whether they have general validity. For example, it would be interesting to better define the conditions (type of weathering, time of exposure, wood type, etc.) governing the existence and thickness of a brown layer below an outer gray layer at the surface of weathered wood (Browne and Simonson 1957). Also, using modern analysis tools, there is an opportunity to characterize the chromophoric substances, e.g. determining whether they are byproducts of the decomposition of lignin.

To summarize the main effects of natural weathering on color, Table 2 reports the average values for color based on 28 sets of reported data. The color is specified based on the CIELab system (Broadbent 2004). A full listing of the source data is provided in Table A (see Appendix). According to the CIELab system, the quantity L* corresponds to the lightness of the material, with 100 representing an ideal white and 0 representing a pure black. The quantity a* indicates the degree of redness (if positive in value) or greenish appearance (if negative). The quantity b* indicates the degree of yellowness (if positive) or blueness (if negative). One unit of change in any of the parameters can be taken as an approximation of a typical perceptible difference for objects having uniform coloration under ideal lighting.

Table 2. Average Effects of Natural Weathering on Color Attributes of Wood after Brief or Long-term Exposure

Notes: Mean values and (standard deviations) shown; § Short time was selected as either the earliest reported non-zero elapsed time of a given study, or if available, a time corresponding to a maximum increase in a* value. §§ Long time always represented the longest duration of exposure for each study. Data from the following studies were considered: Delucis et al. 2016; Gonzalez-C. et al. 2015; Kržišnik et al. 2018; Liu et al. 2017; Machova et al. 2019; Mohebby and Saei 2015; Oberhofnerová et al. 2017; Ozgenc and Yildiz 2016; Panek and Reinprecht 2016a,b; Reinprecht and Panek 2015; and Turkoglu 2015b. There were 28 sets of values averaged for initial and long-term data. There were 27 sets of values averaged for short-term data.

An example of the appearance of weathered wood, including its color, is provided in Fig. 1. In this case, the image on the right corresponds to a tangential surface of the red oak (Quercus rubra) wood board that had been facing the weather. The left-hand image depicts the backside, not facing the weather, of the same board.

Fig. 1. Example of weathered red oak (Quercus rubra) wood. The right side of the board was exposed to the weather, whereas the left side was covered.

The central image shows the end grain, as well as revealing some aspects of roughness on the other faces, especially the weathered surface (see later discussion). Note the grayer and less red appearance of the weathered (right-hand) image in comparison to the left-hand image. The non-uniformity of coloration is also worth noting in this example.

To summarize the findings, the general trend was that wood specimens eventually became darker (lower L* value), less red (a* value lower and nearer to zero), and less yellow (b* value lower and nearer to zero). “Graying” is a general term to sum up all of these trends, when considering the effects of long-term exposure. However, it is important to note that the short-term effects often did not correspond to the long-term trends. In many specific cases, relatively short-term exposure to natural weathering increased the values of a* especially, and sometimes b* as well. In other cases, due to differences either in species or test conditions, the short-term effects were more in line with the long-term effects.

Roughness

A second very commonly reported observation of the effects of weathering is that the wood surface becomes rougher, often on a fine scale. Such observations are summarized in Table 3. Among the publications cited in this review, the term “rough” generally has been used when describing features in a size range from about 0.5 to several mm. Such features include corrugated annual rings and protruding, loose fibers.

Regarding the reports of “corrugated” morphology, coinciding with annual rings, it has been reported that the latewood, which is denser, tends to be more resistant to weathering than the earlywood, which tends to recede more quickly in the course of weathering (Feist 1982; Feist and Rowell 1982; Williams et al. 2001b,c; Williams 2005; Hazneza and Evans 2016; Petrillo et al. 2019). A logical interpretation might be that the denser parts of the wood are simply stronger and more resistant. As a possible alternative explanation, Miniutti (1964) suggested that at least part of the effect might be attributed to crushing of springwood during planing of the wood; upon rewetting of the wood, the crushed structures would be expected to rebound, giving rise to a corrugated surface. Such a mechanism, however, does not explain why the latewood would mainly comprise the raised ridges, as shown in cross-sectional micrographs (Williams 2005). The development of voids at the microscale has been reported only once (Evans et al. 2008) to the best of the authors’ knowledge. Accordingly, this could be a useful area of focus for future research.

Table 3. Reported Increased Roughness of Wood Due to Natural Weathering

An example of a weathered piece of white oak (Quercus alba) wood that shows substantial roughness is provided in Fig. 2. In this case, the right-hand image corresponds to the weathered surface. Note the profile of the right side of the central section of this composite figure. It is apparent here that material had been removed in an uneven manner from the wood surface. The erosion left an uneven surface with some cracks extending inward. The topic of cracks is considered next.

Fig. 2. Example of weathered white oak (Quercus alba) wood. The left side of the board was planed; the right side shows a deep relief due to erosion by weathering.

Cracking

In reporting cracks and related features in weathered wood, it has been noted that very small cracks may appear initially, and these can be followed by larger cracks at longer times of weathering. Reported observations of this type are summarized in Table 4. Microscopic examination of weathered wood by Miniutti (1964) showed that some cracks were within individual beech wood fibers, where they generally followed the microfibril angles of the predominant S2 sublayers. Other micro-scale cracks coincided with the middle lamella regions between some of the adjacent fibers. Xiao et al. (2012) likewise observed cracks within individual softwood tracheids.

Table 4. Reported Cracks and Checks in Wood Due to Natural Weathering

At the macro-scale, Sandberg (1999) observed that the frequency and depths of cracks tend to be much greater on exposed tangential surfaces of wood than on the corresponding radial surfaces. The ratio between tangential and radial surface crack lengths was observed to be 6 in the case of spruce and 13 in the case of pine.

The word “checking” appears to have a range of meanings in the published literature, including large cracks extending outwards from the heartwood of a tree and apparently brought about by extensive shrinkage of the sapwood, which surrounds the heartwood. Also, it can refer to the development of large cracks during drying of certain lumber types, such as acacia (Tenorio et al. 2012). Checking is also a term used for cracks that develop in the end of logs when they are cut and allowed to dry before being delivered to the sawmill (Linares-Hernandez and Wengert 1997). When the word has been used to describe surface effects, it is often used to describe much smaller features. For instance, Feist (1982) stated that “Checking in the wood may eventually grow into large cracks.” Williams (2005) describes the development of checks starting at bordered pits, which are micrometer-scale features at fiber surfaces. Miniutti (1964), Dunningham et al. (1992) and Evans et al. (1994) referred to “micro-checks”. Jebrane et al. (2017) appear to use the word “check” as a synonym for “crack”, as in the phrase “check propagation”. Osawa et al. (2019) observed that some checks were 20 mm in depth from the wood surface. In the present article the word checking will be used when referring to cracks in weathered wood that extend inward from the exposed surface. They are often arranged in a series of parallel cracks. As discussed in a later section, they appear to be related to cycles of wetting and drying of the wood.

Figure 3 provides an example showing the development of micro-cracks in a specimen of red oak (Quercus rubra) wood. Detail 3.1 shows the crack formation starting on a medullary ray but deviating from it and breaking its way through the vessels into the late wood xylem. Detail 3.2 shows a typical crack formation for oak along the medullary ray in radial direction. The crack formation along the rays can partially explain the higher frequency and depth of cracks on tangential surfaces, compared to radial surfaces, as described by Sandberg (1999).