# Evaluating Dimensional Stability in Modified Wood: An Experimental Comparison of Test Methods

## Abstract

**:**

## 1. Introduction

- Repeated water soaking to saturation followed by oven drying
- Short-term water soaking (swellometer test)
- Long-term (equilibrium) humidity cycling
- Short-term humidity cycling (Harris test),

- Cell-wall impregnation modification (furfurylation),
- Thermal modification using superheated steam (dry conditions)
- Cell-wall chemical modification (acetylation).

## 2. Materials and Methods

#### 2.1. Wood Modifications

#### 2.2. Repeated Water-Soak Test

_{UWS}—volume of unmodified reference biscuit after water saturation; [mm

^{3}]

_{UOD}—volume of unmodified reference biscuit after oven drying; [mm

^{3}]

_{TWS}—volume of modified biscuit after water saturation; [mm

^{3}]

_{TOD}—volume of modified biscuit after oven drying; [mm

^{3}]

_{WS}—water saturated tangential dimensions; [mm]

_{OD}—oven dried tangential dimensions; [mm]

_{WS}—water saturated radial dimensions; [mm]

_{OD}—oven dried radial dimensions; [mm]

#### 2.3. Short-Term Water Soak (Swellometer) Test

_{W}) is calculated for each sample as follows:

_{0}—tangential dimension before soaking; (mm)

_{30}—tangential dimension after 30 min soaking; (mm)

_{w}would only be used where a relevant unmodified reference was available (e.g., matched modified and unmodified samples). In this study matched reference samples were not available for all the treatments, so a published value was used for consistency.

_{w}, two additional metrics were chosen: the percentage of maximum swelling after 3 days (%SW

_{max}), and the percentage of swelling that occurs after 30 min (%SW

_{30}). These are intended to quantify the overall levels of swelling of the samples, and the rate at which they swell, respectively.

_{max}is the maximum swelling (% of initial tangential dimension)

_{final}and T

_{0}are the final and initial tangential dimensions (mm), respectively.

_{30}is the proportion of the swelling that occurs after 30 min soaking, and is calculated as follows:

_{30}is the tangential dimension (mm) after 30 min soaking, and the remaining parameters are defined in Equations (3) and (4).

#### 2.4. Equilibrium Humidity Cycling Test

_{MC}—mass at specified humidity; (g)

_{OD}—mass when oven dry; (g)

_{MC}—tangential dimension at the specified humidity; (mm)

_{OD}—tangential dimension when oven dry; (mm)

_{T}—swelling coefficient in the tangential dimension

_{MC1}—relative humidity at first measurement level; (%)

_{MC2}—relative humidity at second measurement level; (%)

_{MC}—equilibrated tangential dimensions; (mm)

_{OD}—oven dried tangential dimensions; (mm)

_{MC}—equilibrated radial dimensions; (mm)

_{OD}—oven dried radial dimensions; (mm)

#### 2.5. Short-Term (Harris) Humidity Cycling Test

_{90}—tangential dimension at high humidity; (mm)

_{65}—tangential dimension at medium humidity; (mm)

#### 2.6. Statistical Analysis

## 3. Results

#### 3.1. ASE from Repeated Water-Soak Test

^{2}= 0.922), so the percentage tangential swelling would be a more suitable metric to use in situations where calculating ASE by comparing all the samples with an unmodified reference does not make sense (e.g., for commercial samples, or comparing wood species). Further details of this regression analysis can be found in the supplementary data file.

#### 3.2. Swellometer Test

_{w}) values for each wood type are shown in Figure 3. As with ASE above, ASE

_{w}measures the percentage reduction in swelling relative to unmodified wood, so higher values indicate greater dimensional stability. Ideally the average swelling of the control samples would be the same as the published swelling figure, and thus the average ASE

_{w}of the control samples would be zero. Here the control samples have swelled more than the published figure, so almost all control samples have a negative ASE

_{w}. This illustrates the importance of having an appropriate unmodified reference when calculating ASE

_{w}, and why it is worth considering an alternative metric when an appropriate reference is not available. All the wood types have significantly higher ASE

_{w}values compared to the control samples. The TH samples have significantly lower ASE

_{w}values compared to the other modifications.

^{2}= 0.850), showing that the swellometer samples are at, or close to, full saturation at the end of the test. As with the water soak/oven dry test, comparing the maximum tangential swelling has the advantage of not requiring unmodified reference measurements.

_{30}) is shown in Figure 4. The unmodified control samples had swelled to over 90% of their final dimensions in the first 30 min, indicating that they swelled very quickly. The FA samples, on average, only swelled to around 40% of their final dimension, showing that they swelled much more slowly. The AC and TH samples swelled to 70–80% of their final dimensions after 30 min of soaking. It is interesting to note the relative performance of the FA and TH modifications between the ASE results (Figure 2) where both wood types have similar results (indicating similar levels of swelling), and results from swelling after 30 min (Figure 4), where the FA samples have swelled much less than the TH. If soaked for long enough, both wood types will swell a similar amount, but they do so at very different rates. Examples of tangential swelling over time for each wood type are given in the supplementary data file.

#### 3.3. Equilibrium Humidity Cycling

#### 3.4. Harris Humidity Cycling

## 4. Discussion

#### 4.1. ASE Results Obtained from Repeated Water-Soak Test

#### 4.2. Swellometer Test

_{w}has similar limitations to the ASE results above, namely the requirement for an unmodified reference sample as well as the results being dimensionless, and consequently difficult to relate to in-service performance. The total swelling, and percentage swelling after 30 min, appear to give more meaningful results. However, it should be noted that swelling rates will vary with specimen geometry, because water is preferentially absorbed through the end grain of the samples [7], so measured swelling rates may not accurately predict the rate of swelling of a particular component in service.

#### 4.3. Equilibrium Humidity Cycling

#### 4.4. Harris Humidity Cycling

## 5. Conclusions

## Supplementary Materials

## Funding

## Institutional Review Board Statement

## Informed Consent Statement

## Data Availability Statement

## Acknowledgments

## Conflicts of Interest

## References

- Hill, C.A.S. Wood Modification. Chemical, Thermal and Other Processes; John Wiley & Sons, Ltd.: Chichester, UK, 2006. [Google Scholar]
- Skaar, C. Hygroexpansion in Wood. In Wood-Water Relations; Springer: Berlin/Heidelberg, Germany, 1988. [Google Scholar]
- Spear, M.; Walker, J.C.F. Dimensional instability in timber. In Primary Wood Processing: Principles and Practice; Springer: Dordrecht, The Netherlands, 2006; Volome 9781402043932, pp. 95–120. [Google Scholar]
- Noack, D.; Schwab, E.; Bartz, A. Characteristics for a judgment of the sorption and swelling behavior of wood. Wood Sci. Technol.
**1973**, 7, 218–236. [Google Scholar] [CrossRef] - Anonymous. Wood Handbook-Wood as and Engineering Material; General Technical Report FPL-GTR-190; U.S. Department of Agriculture, Forest Service, Forest Products Laboratory: Madison, WI, USA, 2010. [Google Scholar]
- Sargent, R. Evaluating dimensional stability in solid wood: A review of current practice. J. Wood Sci.
**2019**, 65, 1–11. [Google Scholar] [CrossRef] - Čermák, P.; Suchomelová, P.; Hess, D. Swelling kinetics of thermally modified wood. Eur. J. Wood Wood Prod.
**2021**, 79, 1337–1340. [Google Scholar] [CrossRef] - Navi, P.; Sandberg, D. Heat Treatment. In The Thermo-hydro-mechanical Processing of Wood; EPFL Press: Lausanne, Switzerland, 2012; pp. 249–286. [Google Scholar]
- Dong, Y.; Qin, Y.; Wang, K.; Yan, Y.; Zhang, S.; Li, J.; Zhang, S. Assessment of the performance of furfurylated wood and acetylated wood: Comparison among four fast-growing wood species. BioResources
**2016**, 11, 3679–3690. [Google Scholar] [CrossRef] [Green Version] - Cermák, P.; Baar, J.; Dömény, J.; Výbohová, E.; Rousek, R.; Paril, P.; Oberle, A.; Abalová, I.G.; Hess, D.; Vodák, M.; et al. Wood-water interactions of thermally modified, acetylated and melamine formaldehyde resin impregnated beech wood. Holzforschung, 2022; in press. Available online: https://www.degruyter.com/document/doi/10.1515/hf-2021-0164/html(accessed on 23 August 2021).
- Finnish ThermoWood Association. ThermoWood Handbook; Finnish ThermoWood Association: Helsinki, Finland, 2003. [Google Scholar]
- Bongers, F.; Alexander, J.; Marcroft, J.; Crawford, D.; Hairstans, R. Structural design with Accoya wood. Int. Wood Prod. J.
**2013**, 4, 172–176. [Google Scholar] [CrossRef] - Hill, C.; Jones, D. The Dimensional Stabilisation of Corsican Pine Sapwood by Reaction with Carboxylic Acid Anhydrides. Holzforschung
**1996**, 50, 457–462. [Google Scholar] [CrossRef] - Kininmonth, J.A.; Whitehouse, L.J. Properties and Uses of New Zealand Radiata Pine; Kininmonth, J.A., Whitehouse, L.J., Eds.; Forest Research Institute: Rotorua, New Zealand, 1991; Volume One-Wood Properties. [Google Scholar]
- Suchsland, O. The Swelling and Shrinking of Wood: A Practical Technology Primer; Forest Products Society: Madison, WI, USA, 2004. [Google Scholar]
- Sargent, R.; Durbin, G.J.; Dunningham, E.A. Measurement of dimensional stability. In Proceedings of the 8th European Conference on Wood Modification, Helsinki, Finland, 26–27 October 2015. [Google Scholar]
- Rowell, R.M.; Wang, R.H.S.; Hyatt, J.A. Flakeboards made from aspen and southern pine wood flakes reacted with gaseous ketene. J. Wood Chem. Technol.
**1986**, 6, 449–471. [Google Scholar] [CrossRef] - Wdma, T.M. 2-1999 Test Method to Determine the Short Term Anti-Swell Effectiveness of Treating Systems; The Window & Door Manufacturers Association: Chicago, IL, USA, 2009. [Google Scholar]
- Rowell, R.M.; Banks, W.B. Water Repellency and Dimensional Stability of Wood. Gen. Tech. Rep. FPL-50; United States Department of Agriculture, Forest Service, Forest Products Laboratory: Madison, WI, USA, 1985; 24p. [Google Scholar]
- DIN 52 184; Testing of Wood; Determination of Swelling and Shrinkage. Deutsches Institut für Normung E.V.: Berlin, Germany, 1979.
- Harris, J.M. The Dimensional Stability Shrinkage Intersection Point and Related Properties of New Zealand Timbers; Forest Research Institute, New Zealand Forest Service: Wellington, New Zealand, 1961. [Google Scholar]
- RStudio Team. RStudio: Integrated Development for R. RStudio; PBC: Boston, MA, USA, 2020; Available online: http://www.rstudio.com/ (accessed on 23 August 2021).
- Dilba, G.; Bretz, F.; Guiard, V. Simultaneous confidence sets and confidence intervals for multiple ratios. J. Stat. Plan. Inference
**2006**, 136, 2640–2658. [Google Scholar] [CrossRef] - Djira, G.; Hasler, M.; Gerhard, D.; Segbehoe, L.; Schaarschmidt, F. Mratios: Ratios of Coefficients in the General Linear Model. Available online: https://CRAN.R-project.org/package=mratios (accessed on 7 September 2021).
- Rowell, R.M.; Ibach, R.E.; James, M.; Thomas, N. Understanding decay resistance, dimensional stability and strength changes in heat-treated and acetylated wood. Wood Mater. Sci. Eng.
**2009**, 4, 14–22. [Google Scholar] [CrossRef] - Chauhan, S.S.; Aggarwal, P. Effect of moisture sorption state on transverse dimensional changes in wood. Holz Als Roh-Und Werkst.
**2004**, 62, 50–55. [Google Scholar] [CrossRef] - Altgen, M.; Willems, W.; Hosseinpourpia, R.; Rautkari, L. Hydroxyl accessibility and dimensional changes of Scots pine sapwood affected by alterations in the cell wall ultrastructure during heat-treatment. Polym. Degrad. Stab.
**2018**, 152, 244–252. [Google Scholar] [CrossRef]

**Figure 2.**Anti-shrink efficiency values for each modification type (excluding the first water-soak oven-dry cycle). Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level).

**Figure 3.**ASE

_{w}values for each wood type (calculated following 30 min soaking). Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level).

**Figure 4.**Percentage of total swelling that has occurred after 30 min (%SW

_{30}). Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level).

**Figure 5.**Equilibrium moisture content (EMC) values measured for each sample, and at different humidity levels (excluding the first humidity cycle). Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level). EMC values for all humidity conditions were pooled when calculating the significance groupings.

**Figure 6.**Radial (

**a**) and tangential (

**b**) dimension change for each 1% change in relative humidity (h

_{R}and h

_{T}, respectively). Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level).

**Figure 7.**Swelling anisotropy ratios (T/R) calculated from the ratio of swelling coefficients. Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test.

**Figure 8.**Percentage change in (

**a**) radial and (

**b**) tangential direction after 24 h at 90% RH for a 1% change in RH. Superscripted letters on the x-axis indicate significant differences according to Tukey’s HSD test (95% confidence level).

Step | Temperature | Relative Humidity |
---|---|---|

1 | 25 °C | 60–70% RH (High) |

2 | 25 °C | 90–95% RH (High) |

3 | 25 °C | 60–70% RH (Low) |

4 | 25 °C | 60–70% RH (Medium) |

**Table 2.**Mean swelling anisotropy ratios (T/R) for each wood type, plus standard deviation values (SD) in parentheses.

Wood Type | T/R Ratio (SD) |
---|---|

Control | 1.91 (0.28) * |

FA | 1.91 (0.46) |

TH | 1.54 (0.26) * |

AC | 3.97 (4.04) |

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |

© 2022 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Sargent, R.
Evaluating Dimensional Stability in Modified Wood: An Experimental Comparison of Test Methods. *Forests* **2022**, *13*, 613.
https://doi.org/10.3390/f13040613

**AMA Style**

Sargent R.
Evaluating Dimensional Stability in Modified Wood: An Experimental Comparison of Test Methods. *Forests*. 2022; 13(4):613.
https://doi.org/10.3390/f13040613

**Chicago/Turabian Style**

Sargent, Rosie.
2022. "Evaluating Dimensional Stability in Modified Wood: An Experimental Comparison of Test Methods" *Forests* 13, no. 4: 613.
https://doi.org/10.3390/f13040613