Next Article in Journal
Quantitative Testing and Analysis of Non-Standard AEB Scenarios Extracted from Corner Cases
Next Article in Special Issue
Testing Protective Clothing with an Emphasis on Subjective Opinions
Previous Article in Journal
Novel Cooling Strategy for a Hybrid Photovoltaic/Parabolic Dish Concentrator
Previous Article in Special Issue
Impact of Filling Stations: Assessing the Risks and Consequences of the Release of Hazardous Substances
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Applied Sciences
Volume 14
Issue 1
10.3390/app14010170
applsci-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Effect of Spruce Wood Density on Selected Fire-Technical Parameters during Thermal Loading

by
Patrik Mitrenga
*,
Linda Makovická Osvaldová
and
Milan Konárik
Department of Fire Engineering, Faculty of Security Engineering, University of Žilina, 010 26 Žilina, Slovakia
*
Author to whom correspondence should be addressed.
Appl. Sci. 2024, 14(1), 170; https://doi.org/10.3390/app14010170
Submission received: 5 December 2023 / Revised: 15 December 2023 / Accepted: 21 December 2023 / Published: 24 December 2023
(This article belongs to the Special Issue Advanced Technologies in Environment Protection and Environmental Risk Assessment)
Browse Figures
Versions Notes

Abstract

:
The paper evaluates the effect of spruce wood density on the parameters of mass loss and mass loss rate during exposure to thermal load. The intention was to determine whether the effect of density is still evident after the application of flame retardants to the test samples. Groups of samples with different densities under the same retardant treatment were compared. The differences in densities of the compared groups of samples were different for each flame retardant. Water-soluble flame retardants based on inorganic salts were used. For testing, a simple test method was used in which the samples were exposed to direct flame from a Bunsen burner. The results of the study are the findings of how wood density affects the burning process of the samples treated with flame retardants. Statistical evaluation of the experimental results shows a significant effect of wood density on the monitored parameters even when flame retardants are used. For a difference in sample densities of 244 kg·m−3, there was a density dependence of the mass loss rate, with the lower density samples having a higher mass loss rate (0.158%·s−1) over the whole experimental period compared to the higher density samples (0.077%·s−1). The ANOVA test also demonstrated the influence of density on the mass loss of the samples at the above density difference. At lower density differences (51 kg·m−3 and below), the effect of sample density on the observed parameters was no longer evident. The fire spread rate parameter was also investigated. Here, a linear correlation between the difference in sample densities and the difference in the values of the above parameter at high and low densities is observed with a reliability coefficient R2 = 0.99.

1. Introduction

Wood is a widely used construction material for both exterior and interior applications [1], and its global production has been steadily increasing, surpassing 800 million tonnes per year in 2015. In Europe, the wood industry comprises approximately 400,000 enterprises, representing approximately one fifth of all manufacturing businesses [2]. One of the main drawbacks of wood and other wood-based materials in all forms, when used as a building material, is its flammability [3,4,5,6,7,8,9,10], limiting its fire resistance [11,12,13].
To address this concern, efforts are being made to develop protective measures and flame retardants [14]. However, accurately testing these flame retardants is crucial to detect even minor differences in their effectiveness. In Europe, several testing methods [15,16,17,18,19] are applied to classify products into reaction-to-fire classes [20]. However, these are insufficient for assessing flame retardants. Therefore, other, much more sensitive methods are also used to compare the effects of flame retardants. Thermal analysis methods are used to study the properties of materials as temperature changes. Of these, differential scanning calorimetry (DSC), which measures the heat released or absorbed during the heating process, is used in particular. Differential thermal analysis (DTA) can be used to determine the oxidation process. Thermo-gravimetric analysis (TGA) can be used to measure the change in mass of the sample during the heating or cooling process [21,22,23]. For the analysis of gases produced under heat load, the determination of heat release and mass loss rate can be performed using the cone calorimeter method [24]. There are other non-normal methods for the evaluation of flame retardants that are sensitive enough to investigate the combustion behaviour of these materials. Several authors [25,26,27,28,29,30,31,32] have already evaluated some flame retardants using different methods [9,33].
Among the suitable test methods, one involves exposing test samples to direct flame [34,35]. In this procedure, actual sample masses are recorded over time, and subsequently, the mass loss rate can be calculated. It is one of the most significant fire engineering parameters used in the past in mathematical models of forest fire spread [36] or in modelling the fire resistance of wooden structures [7,37]. It is also possible to determine the heat release rate HRR from this parameter under certain conditions [38]. The specific testing conditions were already set by the researchers according to the type and kind of material. To compare the effectiveness of different wood retardant treatments with sufficient accuracy, it is necessary to minimise other factors that affect wood combustion. These are mainly the chemical composition, structure, density, material surface, moisture content and geometric shape of the wood [39,40,41,42]. Chemical composition and structure are mainly determined by the type of wood. Experimental tests are always conducted on the same species of wood when the retardation effects are compared. Also, the shape of the samples (dimensions), the moisture content of the wood and the material treatment must be identical. The above factors (species, moisture content, dimensions and treatment of the samples) shall be minimised in the experiments.
However, a significant, frequently neglected, factor is the density of the wood. Wood density indicates the mass per unit volume of wood [43], usually given at a specific moisture content because this directly influences it [44,45]. It is calculated as the ratio of the mass and the volume of wood at the same moisture content. According to several authors, wood density influences the combustion process in addition to physical and mechanical properties [46,47]. Wood with a higher density is more difficult to ignite than wood with a lower density because it has better thermal conductivity and more intense heat dissipation from the material surface to the internal zones. This effect on ignitability is particularly pronounced in large wood-based materials [48,49]. In order to achieve better fire-resistant properties in the wood, wood densification is carried out, which removes the intercellular space. Delignification, i.e., the reduction in lignin and hemicellulose content, is also part of the densification process [50,51]. According to several studies, wood densification significantly increases fire retardation [52,53,54]. The use of flame retardants in the densification process further increases the flame resistance of wood [55,56,57]. However, there are no scientific studies yet that provide relevant evidence on the extent to which wood density affects the combustion process of undensified wood treated with flame retardants.
This paper aims to investigate how the density of spruce wood influences the burning process of wood samples treated with flame retardants. In particular, the study is applicable in the evaluation of the effectiveness of different wood retardant treatments. As already indicated, the experiments are performed by a test method in which the samples are exposed to a direct flame. The criteria evaluated will be mainly the rate of flame retardancy, mass loss and other additional parameters. We also chose this method because of the simplicity of the test equipment, and we designed the test procedure to reflect the conditions of a real fire. The samples for the experiment will be selected from spruce wood, as it is the most widely used wood species in the construction industry [58,59].

2. Materials and Methods

2.1. Materials

The samples of spruce wood (Picea abies) with dimensions 200 × 100 × 10 mm were used for the experiment. They were selected to be free of lumps, resins and other defects that could affect the combustion process. Before the experiment and the application of flame retardants, all samples were placed in the laboratory for two weeks at a temperature of 21 °C and a humidity of approximately 40% to ensure that the samples’ moisture content matched the end-use conditions. The final moisture content of the samples before applying the flame retardants was approximately 12%.
For the first experiment, untreated samples were first used. Due to the high variability in the measured data caused by the samples being burnt through, samples treated with flame retardants were used for further experiments to see if the wood density would affect the experimental results with such samples.
Three different wood flame retardants were applied to three groups of samples. The applied retardants were transparent, water-soluble, colourless coatings based on inorganic salts. Their specific composition is not specified here as the aim of this paper is not to evaluate the retardants themselves. They were applied to all sides of the samples, including their edges, in one coat using a brush. The average amount of coating varied between 60 and 80 g·m−2 depending on the specific retardant (according to the instructions of the manufacturers of the individual flame retardants). The retardants on the samples were dried for two months in a room with a temperature of 21 °C and a humidity of about 40%.
The samples were supplied by an external contractor. When measuring the densities of the individual samples, we found a very large variability. The aim was to take advantage of this variability and create groups of samples with various density differences. To investigate the effect of density on the mass loss rate and mass loss, each studied sample group included ten higher- and ten lower-density samples when one type of flame retardant was applied. The designations of each sample group are specified in Table 1. The average sample densities used in the experiment and the range of variation of the densities of the individual samples within a test set are depicted in Table 2.
As demonstrated in Table 1 and Table 2, the samples were intentionally selected to exhibit varying mean densities within each group. For each retardant coating, the samples comprised both higher (>450 kg∙m−3) and lower (<450 kg∙m−3) densities, with the magnitude of density differences differing for each coating. The boundary between high and low densities was determined approximately as the median of the densities of all samples used for the research. This design enabled us to establish the relationship between the difference in sample densities and the resulting testing parameters.
The densities of each sample were calculated based on the exact masses and volumes of the samples. The volume of the samples was calculated based on the exact dimensions of the samples. The length of the samples was measured in mm at the centre of the sample to the nearest 0.5 mm. Width was measured with a calliper to hundredths of millimetres at both ends and then averaged. Thickness was measured with a calliper at three points of the specimen to hundredths of a millimetre and then averaged.

2.2. Test Equipment and the Procedure

The test apparatus consists of a gas cylinder, a flow valve, a flow regulator, a burner, a burner holder, a sample holder, precision scales weighing in grams to two decimal places and a computer. Figure 1 demonstrates a schematic diagram with the actual elements of the apparatus.
The flow regulator on the cylinder (3) ensures a constant flow of 30 mbar, without interruptions and fluctuations. The burner also has a gas flow regulator (7) to allow precise adjustment of the flame height, which in our case is 10 cm. The samples (10) are fixed in the holder (9) at an angle of 45° to the horizontal plane. With this setup, we also captured the flame propagation. During the experiment, the samples were exposed to a flame heat source applied from the underside, with 1 cm of flame in direct contact with the sample. Propane–butane gas was used as the fuel. The flame temperature of the propane–butane used is 1950 °C, and its calorific value is 44 MJ·kg−1 [60]. Scales (8) recorded the actual sample mass at 10 s intervals. Measurement accuracy was ensured by a Mettler Toledo MS1602S/M01 scale (Greifensee, Switzerland) weighing hundredths of grams. Automatic mass recording at set intervals was provided by the BalanceLink 4.2.0.1 programme (Mettler Toledo, Switzerland). The total testing time per sample (exposure to the flame heat source and mass recording) was 10 min. After this time, the mass recording was stopped. Figure 2 is a real photo of the main part of the test equipment during the experiment. The variability of the average values did not exceed the values that were characteristic of untreated wood.
The study [34] confirms the suitability of this method for testing spruce wood and its treatments. The variability of the values of the parameters tested by this method does not exceed the characteristic variability values for untreated spruce wood.

2.3. Evaluation and Calculation

The main evaluation criterion was the relative mass loss rate, calculated from the measured mass data over time according to the Formula (1).
v r = m ( τ ) − m ( τ + Δ τ ) m ( τ ) · Δ τ · 100 ,
where vr is the relative mass loss rate [%·s−1], m(τ) is the mass of the sample at a time (τ) [g], m(τ + Δτ) is the sample mass at a time (τ + Δτ) [g] and Δτ is the time interval at which the mass is read [s]. The time interval for reading the mass was 10 s in our case.
Another datum significant for the statistical detection of the dependence of different average sample densities on the experimental results is the mass loss calculated by the Formula (2).
δ m p ( τ ) = m − m ( τ ) m · 100 ,
where δmp(τ) is the mass loss at time (τ) [%], m is the original mass of the sample before the experiment [g] and m(τ) is the mass of the sample at time (τ) [g].
Next, we investigated the fire spread rate parameter, which is the ratio of the maximum value of the mass loss rate and the time to reach this value. This value is determined as the average for the first peak of the mass loss rate according to Formula (3) [61].
R f s = v r τ ( v r ) ,
where Rfs is the fire spread rate [%·s−2], vr is the relative mass loss rate [%·s−1] and τ(vr) is the time to reach the first peak of the maximum mass loss rate [s].
As this value increases, the fire spread rate also increases [61].
The results presented below are specified as average values.
The dependence of mass loss on sample density was verified by one-factor analysis of variance (ANOVA) using the statistical software “R” (The R Project for Statistical Computing) version 4.1.2.

3. Results and Discussion

In our case, the fundamental parameters for monitoring the influence of the density of spruce wood samples on the combustion process are the mass loss and the mass loss rate. The tested samples were carefully selected to ensure consistency, free of knots, visible resins and any visible deformations or cracks. They displayed identical wood texture and moisture content before coating and were processed by the same method—smooth planning. However, during the experiment, it was observed that some samples burned through from the sides to the top, resulting in a significant increase in the mass loss and, consequently, the mass loss rate. The flame retardants used dilute the combustible gases released from the wood by heat and thus make it more difficult to ignite. In the above cases, the effect of the retardants was apparently insufficient, which may have caused the spontaneous combustion of the samples. This may be due to the application of less retardant to the samples or an overall lower concentration of active substance in the individual retardants combined with a higher flame intensity on the samples, where after a certain time the retardant failed to produce sufficient non-flammable gases to prevent spontaneous combustion. Why this was the case for only some samples is not entirely clear. It may be due to the different properties of the individual samples (chemical composition, resin, porosity, etc.) or possible variations in the testing.
To prevent these outliers from significantly affecting the average values, they were excluded from the overall evaluation. The identification of these extremes was performed statistically using box plots displaying the values of the total mass loss over the 600 s testing time for each group of samples. These box plots were constructed using the statistical software “R” (The R Project for Statistical Computing) version 4.1.2 and are presented in Figure 3.
From the plot in Figure 3, we can see that there were outliers in the total mass loss for each group of samples. The highest number of outliers (three) occurred in the lower-density samples with the application of retardant 3 (R3-LD). It was mainly due to the low variability in the other measured data from that group. The number of outliers for each group of samples was random and did not depend on the treatment applied or the sample density. We then excluded the extreme values (marked by red dots) from further evaluations to minimise their undesired influence on the overall evaluation of the experiment.
The mean values of the mass loss of the samples after the experiment, including the standard deviation, and the maximum mass loss rate are given in Table 3.
The untreated samples exhibited a variability of up to 18% mass loss for the higher-density samples and approximately 55% mass loss for the lower-density samples. This is because the untreated samples experienced spontaneous combustion, with some cases even burning through the side edges to their top.
From the beginning of the experiment to the testing time of 300 s, we monitored significant differences in the mass loss rate of the samples as a function of their density. This can be seen in Figure 4. The lower-density samples displayed a higher mass loss rate than the higher-density samples from the beginning of testing. The highest mass loss rate of the U-LD samples (up to a time of 300 s) was 0.130%·s−1, reached in 60 s. The U-HD samples had the highest mass loss rate (up to a time of 300 s) of 0.100%·s−1, reached in 70 s. The highest mass loss rate of the U-LD samples during the whole testing period was 0.22%·s−1 at a testing time of 450 s due to spontaneous sample combustion.
The sudden increase in the mass loss rate of untreated samples occurs at a testing time of about 300 s. In this case, it was the second peak of the maximum mass loss rate, which, according to several authors [62,63,64], occurred in untreated wood. It caused spontaneous combustion or ignition of the samples. It had a significant telling value. The later the ignition occurs, the safer the material is from a fire protection point of view.
In contrast, untreated samples with higher density (480 kg·m−3) did not exhibit a second peak in the mass loss rate during the experiment. If the testing time were extended, it would probably appear since, from a testing time of 540 s, mass loss rate started to increase significantly until the end of the experiment. However, the second peak in the maximum mass loss rate would occur much later than for the lower-density samples. Thus, it can be argued that the lower-density samples are easier to ignite, in line with [48,49]. Additionally, throughout the experiment, lower-density samples consistently displayed higher mass loss rates compared to higher-density samples.
It was also because of the significant variability in the measured mass loss due to the combustion of untreated samples that we decided to perform experiments using flame retardants on the samples. The assumption was that the retarding effect would prevent overburning of the samples, which would contribute to an objective evaluation of the influence of sample density on the experimental results.
We can conclude that for all groups of flame retardant-treated samples, the variability of the final mass loss over the 600 s experiment time is relatively low (up to 5% mass loss) (see Figure 3). The retardant treatment had a significant retarding effect on the combustion process and, in most cases, prevented their spontaneous combustion.
A one-factor analysis of variance (ANOVA) was performed to determine the mass loss dependence on the density of the wood samples. The analysis was always performed between two sample groups with the same retardant treatment but different densities. The results are presented in Table 4 for the 300 s test time mass loss and Table 5 for the 600 s test time mass loss.
The results of the ANOVA test at 300 s test time indicated statistically significant differences in mass loss as a function of density for retardant 1-treated samples. Samples with retardant treatment 1 demonstrated a significant dependence with a p-value of p = 1.56 × 10−5. However, for the sample groups treated with retardants 2 and 3, there was no longer any mass loss dependence on sample density. For these sample groups, the average difference in densities was within 50 kg·m−3. Similar results were also found at a testing time of 600 s (Table 5).
For the samples treated with retardant 1, we observed a significant difference in the mass loss rates among the higher- and lower-density samples (Figure 5). The average difference in densities here was approximately 244 kg·m−3. We also saw the highest difference in the mass loss rate in the first few seconds of the experiment (up to a testing time of about 90 s), where the lower density samples (R1-LD) displayed twice the maximum mass loss rate as the higher density samples. Although this phenomenon only occurred in the first 90 s of the experiment, it resulted in a significant difference in the final mass loss, as indicated by the dependence of the mass loss on the density of the samples in Table 4 and Table 5. The maximum mass loss rate of the R1-LD samples was reached at a testing time of 50 s with a value of 0.158%·s−1, while the R1-HD samples had their highest mass loss rate at a testing time of 70 s with a value of 0.077%·s−1. However, differences in the mass loss rates were still observed up to a testing time of about 300 s (R1-LD samples had a noticeably higher one). After this time, the mass loss rates of both the higher- and lower-density samples were virtually identical due to the retardant treatment of the samples, which prevented them from spontaneous combustion and had a significant effect on the mass loss rate.
Samples treated with retardant 2, which had a difference in densities of approximately 50 kg·m−3, and samples treated with retardant 3, with a difference in densities of about 8.5 kg·m−3, showed significantly lower variations in the mass loss rates (Figure 6 and Figure 7), which were negligible. The differences primarily occurred in the maximum mass loss rate within the initial seconds, with a difference of 0.002%·s−1 for samples treated with retardant 2 and 0.010%·s−1 for samples treated with retardant 3. Consequently, the average mass loss rate of the higher- and lower-density samples was virtually identical throughout the experiment. These differences also had no significant effect on the total mass loss, as confirmed by Table 4 and Table 5, where the dependence of the mass loss on the sample density was not evident for samples R2 and R3.
From the point of view of fire spread, not only the maximum value of the mass loss rate was significant, but also the time to reach this value. The ratio of these two parameters (Equation (3)) provided us with the rate of fire spread. The higher the value, the more the material contributed to the fire spread [61]. Table 6 demonstrates the fire spread rate (Rfs) values for each group of samples. These values were averaged for the first peak of the mass loss rate (up to a time of 100 s). Here, we observed that as the difference in densities between the sample groups increased, the difference in Rfs values also increased. The fire spread rate between two sample groups with identical treatment was always higher for the lower-density (LD) samples. For example, the samples with the highest density difference R1 exhibited a fire spread rate Rfs difference of 0.00206, whereas the samples with the lowest density difference (R3) had a difference in Rfs values of 0.00014.
Higher differences in densities are logically associated with the higher differences in fire spread rates, as demonstrated by the graph in Figure 8. The strong linear dependence of the difference in fire spread rate on the difference in sample densities with a reliability coefficient of R2 = 0.99 supports this assumption. Based on the above results in Figure 8 and Table 6, it is evident that spruce wood with lower density exhibits a higher fire spread rate, which has significant implications for testing wood flame retardants.
The above experimental findings demonstrate the significant impact of spruce wood sample density on test results, specifically the evaluation criteria of mass loss rate, mass loss and fire spread rate.
For samples treated with retardants, the experimental outcomes were likewise influenced by sample density, depending on the density difference. Samples with an approximate density difference of 244 kg·m−3 exhibited significantly different maximum mass loss rates, with lower-density samples showing higher rates. However, when flame retardants were applied, there was no second peak in the mass loss rate, confirming the retardants’ effect in prolonging ignition time, as reported by [62], effectively preventing combustion initiation.
The mass loss parameter, which is dependent on sample density, was also significantly affected (confirmed by ANOVA tests in Table 4 and Table 5). As the density difference decreased, the dependence of mass loss rate and mass loss on sample density also decreased. Figure 6 and Figure 7 reveal that with density differences up to approx. 50 kg·m−3, there were no significant differences in the mass loss rate, and the dependence of mass loss on sample density was also not confirmed (Table 4 and Table 5).
These findings underscore the vital role of sample density in wood testing and the effectiveness of wood retardant treatments. However, the results are subject to the variability of sample densities selected for testing. Experiments revealed that lower variability in sample densities minimised the impact on test results. For precise testing, a difference in sample densities of up to 50 kg·m−3 is optimal, as supported by ANOVA tests (Table 4 and Table 5) and fire spread rate versus time plots (Figure 5, Figure 6 and Figure 7). Nevertheless, it is challenging, if not impossible, to select samples with identical density. In addition, other properties (e.g., material processing) and defects or deformations (excessive resin content, bulges, cracks, etc.) that can also affect testing results should be considered when selecting samples [65].

4. Conclusions

The study evaluated various fire-technical parameters of spruce wood with different densities, including some samples treated with wood flame retardants. The spruce wood samples were divided into several groups with identical treatments (or no treatment) but different average densities, gradually reducing the density difference to establish a limit for the dependence of experimental results on the difference in sample densities. The experiment utilized a simple method where the samples were exposed to a direct flame at a 45° angle. The monitored parameters were mass loss and mass loss rate, with an additional parameter being the fire spread rate. The conclusions from the experiment can be summarised as follows:
  • Untreated samples exhibited significant variability in the measured data (mass loss variability of untreated samples with low density was 55%, with higher-density samples at 18%), unlike samples treated with retardants, which showed a lower variability of mass loss, approximately 5%. This significant variability in mass loss for untreated samples is mainly attributed to their ignition after a specific time, resulting in a subsequent rapid increase in the mass loss rate. The results confirm, among other findings, that flame retardants significantly prolong the ignition time.
  • Untreated samples with a density difference of about 100 kg·m−3 displayed a clear density dependence on the mass loss. Samples with a lower density have a significantly higher average mass loss rate over the entire duration of the experiment.
  • For samples treated with retardant 1 with a density difference of approximately 244 kg·m−3, a dependence of the mass loss rate on the sample density was observed. The most significant difference was in achieving the maximum mass loss rate, which was twice as high for the lower-density samples compared to the higher-density samples. Also, the mass loss dependence on the sample density was statistically confirmed by the ANOVA test (p-value = 8.46 × 10−5 at a testing time of 600 s).
  • Lower-density samples exhibited higher mass loss and mass loss rate. The study results further indicated that the sample density also affected the fire spread rate parameter (Rfs). As the sample density decreased, the value of Rfs increased, showing a linear relationship between the difference in Rfs values and sample densities with a reliability coefficient of R2 = 0.99.
  • For samples treated with retardants with a density difference of up to 50 kg·m−3, no dependence of the mass loss rate on sample density was observed. The ANOVA statistical analysis did not confirm the mass loss dependence on the sample density. Therefore, the difference in densities of up to 50 kg·m−3 no longer significantly affected experimental results, particularly parameters such as the mass loss rate and mass loss. Consequently, selecting the test samples with such a variation range of densities can be considered optimal without impacting the test results.
The research was aimed at investigating the influence of spruce wood density on mass loss and mass loss rate when using flame retardants. The research was carried out on groups of samples with different densities using three flame retardants. In order to refine the research, it would be advisable to complement the experiments so that several flame retardants were applied for each density respective of the difference in densities. However, such research would be very challenging, especially due to the selection of samples with different densities. In our case of “simplified research”, we had hundreds of samples available to us, from which we selected only 80 suitable for testing.

Author Contributions

Conceptualization, P.M. and L.M.O.; methodology, P.M. and L.M.O.; software, P.M.; validation, M.K.; formal analysis, M.K.; resources, P.M. and L.M.O.; writing—original draft preparation, P.M.; writing—review and editing, L.M.O. and M.K.; visualization, P.M.; supervision, L.M.O. and M.K.; project administration, P.M.; funding acquisition, P.M. All authors have read and agreed to the published version of the manuscript.

Funding

This article was funded by the Grant System of University of Zilina of the project: Experimental Determination of Fire-technical Parameters of Alternative Building Materials and Evaluation of its Fire Safety. Project No. 16961.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are contained within the article.

Acknowledgments

This work was supported by the Grant System of University of Zilina No. 1/2022 (17324).

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Barbu, M.C.; Tudor, E.M. State of the art of the Chinese forestry, wood industry and its markets. Wood Mater. Sci. Eng. 2021, 16, 1030–1039. [Google Scholar] [CrossRef]
  2. Eurostat. Wood Products—Production and Trade. [Online]. Available online: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Wood_products_-_production_and_trade#Wood-based_industries (accessed on 12 June 2023).
  3. Mitu, M.; Razus, D.; Boldor, D.; Marculescu, C. Flammability Properties of the Pyrolysis Gas Generated from Willow Wood. Processes 2023, 11, 2103. [Google Scholar] [CrossRef]
  4. Tureková, I.; Ivanovičová, M.; Harangózo, J.; Gašpercová, S.; Marková, I. Experimental Study of the Influence of Selected Factors on the Particle Board Ignition by Radiant Heat Flux. Polymers 2022, 14, 1648. [Google Scholar] [CrossRef] [PubMed]
  5. Östman, B. Chapter 3—Flammability of wood products. In Flammability Testing of Materials Used in Construction, Transport and Mining, 2nd ed.; Vivek Apte: New Delhi, India, 2021; pp. 61–87. [Google Scholar]
  6. Vandličkova, M.; Markova, I.; Holla, K.; Gašpercová, S. Evaluation of Marblewood Dust’s (Marmaroxylon racemosum) Effect on Ignition Risk. Appl. Sci. 2021, 11, 6874. [Google Scholar] [CrossRef]
  7. Vandlickova, M.; Markova, I.; Makovicka-Osvaldova, L.; Gaspercova, S.; Svetlik, J. Evaluation of African Padauk (Pterocarpus soyauxii) Explosion Dust. BioResources 2020, 15, 401–414. [Google Scholar] [CrossRef]
  8. Kadlicová, P.; Gašpercová, S.; Makovicka Osvaldova, L. Monitoring of Weight Loss of Fibreboard during Influence of Flame. Procedia Eng. 2017, 192, 393–398. [Google Scholar] [CrossRef]
  9. Gašpercová, S.; Makovická Osvaldová, L. Fire Protection in Various Types of Wooden Structures. Civ. Environ. Eng. 2015, 11, 51–57. [Google Scholar] [CrossRef]
  10. Tureková, I.; Turnova, Z.; Balog, K.; Pastier, M. Study of thermal degradation of polymers. Adv. Mater. Res. 2013, 652–654, 1664–1667. [Google Scholar] [CrossRef]
  11. Gao, B.; Wei, S.; Du, W.; Yang, H.; Chu, Y. Experimental Study on the Fire Resistance Performance of Partition Board under the Condition of Small Fire Source. Processes 2021, 9, 1818. [Google Scholar] [CrossRef]
  12. Rusinová, M.; Kalousek, M.; Šlanhof, J. Assessment of the dynamic temperature profile in fire loaded sandwich structures based on wood in comparison with conventional structural systems. Appl. Mech. Mater. 2019, 887, 98–105. [Google Scholar] [CrossRef]
  13. Kučera, P.; Lokaj, A.; Vlček, V. Behavior of the spruce and birch wood from the fire safety point of view. Adv. Mater. Res. 2013, 842, 725–728. [Google Scholar] [CrossRef]
  14. Troitzsch, J.; Antonatus, E. Chapter 3. Flame retardants and flame-retarded plastics. In Plastics Flammability Handbook, 4th ed.; Carl Handes Verlag: München, Germany, 2021; pp. 53–128. [Google Scholar]
  15. EN 13823; Reaction to Fire Tests for Building Products—Building Products Excluding Floorings Exposed to the Thermal Attack by a Single Burning Item. CEN: Brussels, Belgium, 2002.
  16. ISO 1182:2020; Reaction to Fire Tests for Building Products—Non-Combustibility Test. International Standards Organization: Geneva, Switzerland, 2020.
  17. ISO 1716; Reaction to Fire Tests for Building Products—Determination of the Gross Heat of Combustion (Calorific Value). International Standards Organization: Geneva, Switzerland, 2002.
  18. ISO 9239-1; Reaction to Fire Tests for Floorings—Part 1: Determination of the Burning Behavior Using a Radiant Heat Source. International Standards Organization: Geneva, Switzerland,, 2010.
  19. ISO 11925-2:2020; Reaction to Fire Tests—Ignitability of Building Products Subjected to Direct Impingement of Flame—Part 2: Single-Flame Source Test. International Standards Organization: Geneva, Switzerland, 2020.
  20. EN 13501-1; Fire Classification of Construction Products and Building Elements—Part 1: Classification Using Data from Reaction to Fire Tests. CEN: Brussels, Belgium, 2002.
  21. Akash, M.S.H.; Rehman, K. Introduction to Thermal Analysis. In Essentials of Pharmaceutical Analysis; Springer: Singapore, 2020. [Google Scholar] [CrossRef]
  22. Gabbott, P. The Principles and Applications of Thermal Analysis; Wiley-Blackwell: London, UK, 2007. [Google Scholar]
  23. ASTM E1131-20; Standard Test Method for Compositional Analysis by Thermogravimetry. ASTM International: West Conshohocken, PA, USA, 2020.
  24. ISO 5660; Reaction-to-Fire Tests. Heat Release, Smoke Production and Mass Loss Rate. International Standards Organization: Geneva, Switzerland, 2002.
  25. Gašpercová, S.; Marková, I.; Vandlíčková, M.; Osvaldová, L.M.; Svetlík, J. Effect of Protective Coatings on Wooden Elements Exposed to a Small Ignition Initiator. Appl. Sci. 2023, 13, 3371. [Google Scholar] [CrossRef]
  26. Mao, N.; Jiang, L.; Li, X.; Gao, Y.; Zang, Z.; Peng, S.; Ji, L.; Lv, C.; Guo, J.; Wang, H.; et al. Core-shell ammonium polyphosphate@nanoscopicaluminum hydroxide microcapsules: Preparation, characterization, and its flame retardancy performance on wood pulp paper. Chem. Eng. J. Adv. 2021, 6, 100096. [Google Scholar] [CrossRef]
  27. Zhang, L.; Xu, J.; Shen, H.; Xu, J.; Cao, J. Montmorillonite-catalyzed furfurylated wood for flame retardancy. Fire Saf. J. 2021, 121, 103297. [Google Scholar] [CrossRef]
  28. Lu, J.; Jiang, P.; Chen, Z.; Li, L.; Huang, Y. Flame retardancy, thermal stability, and hygroscopicity of wood materials modified with melamine and amino trimethylene phosphonic acid. Constr. Build. Mater. 2021, 267, 121042. [Google Scholar] [CrossRef]
  29. Ma, T.; Li, L.; Liu, Z.; Zhang, J.; Guo, C.; Wang, Q. A facile strategy to construct vegetable oil-based, fire-retardant, transparent and musel adhesive intumescent coating for wood substrates. Ind. Crops Prod. 2020, 154, 112628. [Google Scholar] [CrossRef]
  30. Brahmia, F.; Zsolt, K.; Horváth, P.; Alpár, T. Comparative study on fire retardancy of various wood species treated with PEG 400, phosphorus, and boron compounds for use in cement-bonded wood-based products. Surf. Interfaces 2020, 21, 100736. [Google Scholar] [CrossRef]
  31. Gaff, M.; Kačík, F.; Gašparík, M.; Todaro, L.; Jones, D.; Corleto, R.; Makovická Osvaldová, L.; Čekovská, H. The effect of synthetic and naturalfire-retardants on burning and chemical characteristics of thermally modified teak (Tectonagrandis L. f.) wood. Constr. Build. Mater. 2019, 200, 551–558. [Google Scholar] [CrossRef]
  32. Lee, S.H.; Ashaari, Z.; Lum, W.C.; Abdul Halip, J.; Ang, A.F.; Tan, P.L.; Ling Chin, K.; MdTahir, P. Thermal treatment of wood using vegetable oils: A review. Constr. Build. Mater. 2018, 181, 408–419. [Google Scholar] [CrossRef]
  33. Makovická Osvaldová, L.; Gašpercová, G. The evaluation of flammability properties regarding testing methods. Civ. Environ. Eng. Sci. Tech. J. 2015, 11, 142–146. (In Slovak) [Google Scholar]
  34. Čekovská, H.; Gaff, M.; Osvald, A.; Kačík, F.; Kubš, J.; Kaplan, L. Fire resistance of thermally modified spruce wood. BioResources 2017, 12, 947–959. [Google Scholar] [CrossRef]
  35. Turekova, I.; Markova, I. Ignition of Deposited Wood Dust Layer by Selected Sources. Appl. Sci. 2020, 10, 5779. [Google Scholar] [CrossRef]
  36. MCKenzie, D.; Hessl, A.E.; Kellog, L.K. Using neutral models to identify constraints on low-severity fire regions. Landsc. Ecol. 1996, 21, 139–152. [Google Scholar] [CrossRef]
  37. Janssens, M.L. Modeling of the thermal degradation of structural wood members exposed to fire. Fire Mater. 1998, 28, 199–207. [Google Scholar] [CrossRef]
  38. Martinka, J.; Rantuch, P.; Martinka, F.; Wachter, I.; Štefko, T. Improvement of Heat Release Rate Measurement from Woods Based on Their Combustion Products Temperature Rise. Processes 2023, 11, 1206. [Google Scholar] [CrossRef]
  39. Blažej, A. Chémia dreva; Alfa: Bratislava, Slovakia, 1975; p. 63. ISBN 80-228-0656-0. [Google Scholar]
  40. Dubovský, J.; Berešová, K. Zmena hustoty smrekového dreva vplyvom niektorých fyzikálnych charakteristík. In Proceedings of the 1. Symposium Interakcia Dreva s rôznymi Formami Energie; TU Zvolen: Zvolen, Slovakia, 1996; pp. 23–28. ISBN 80-228-0549-1. [Google Scholar]
  41. Kačíková, D. Dynamika horenia dreva, materiálov a výrobkov na báze dreva. In Selected Processes in the Wood Processing; Technická Univerzita vo Zvolene: Zvolen, Slovakia, 2005; pp. 252–261. ISBN 80-228-1484-9. [Google Scholar]
  42. Karlsson, B.; Quintiere, J.G. Enclosure Fire Dynamics; CRC Press: London, UK, 2000; ISBN 0-8493-1300-7. [Google Scholar]
  43. Danihelová, A. Physics 2; Nikara: Krupina, Slovakia, 2010; 206p, ISBN 978-80-970363-6-2. (In Slovak) [Google Scholar]
  44. Trebula, P.; Klement, I. Drying and Hydrothermal Treatment of Wood; Technical University in Zvolen: Zvolen, Slovakia, 2005; 449p, ISBN 80-228-1421-0. (In Slovak) [Google Scholar]
  45. Požgaj, A.; Chovanec, D.; Kurjatko, S.; Babiak, M. Wood Structure and Properties; Príroda a.s: Bratislava, Slovakia, 1997; 488p, ISBN 80-0700600-1. (In Slovak) [Google Scholar]
  46. Babrauskas, V. Charring rate of wood as a tool for fire investigations. Fire Saf. J. 2005, 40, 528–554. [Google Scholar] [CrossRef]
  47. Hostikka, S.; McGrattan, K.B. Large Eddy Simmulation of Wood Combustion. In Interflam 2001; Interscience Communications Ltd.: London, UK, 2001; Volume 1, pp. 755–762. [Google Scholar]
  48. Kačíková, D.; Netopilová, M.; Osvald, A. Drevo a jeho Termická Degradácia. SPBI Spektrum 45; VŠB—TU Ostrava: Ostrava, Slovakia, 2006; p. 79. ISBN 80-86634-78-7. [Google Scholar]
  49. Osvald, A. Požiarnotechnické vlastnosti dreva a materiálov na báze dreva. In Vedecké štúdie 8/1997/A; ES TU: Zvolen, Slovakia, 1997; p. 51. ISBN 80-228-0656-0. [Google Scholar]
  50. Song, J.; Chen, C.; Zhu, S.; Zhu, M.; Dai, J.; Ray, U.; Li, Y.; Kuang, Y.; Li, Y.; Quispe, N.; et al. Processing bulk natural wood into a high-performance structural material. Nature 2018, 554, 224–228. [Google Scholar] [CrossRef]
  51. Chen, C.; Kuang, Y.; Zhu, S.; Burgert, I.; Keplinger, T.; Gong, A.; Li, T.; Berglund, L.; Eichhorn, S.J.; Hu, L. Structure–property–function relationships of natural and engineered wood. Nat. Rev. Mater. 2020, 5, 642–666. [Google Scholar] [CrossRef]
  52. Xu, Z.; Zhao, W.; Feng, Y.; Tang, X.; Yan, L. Flame-retardant, heat-insulating and char formation properties of densified Pinus sylvestris treated with different compression pressures. Eur. J. Wood Prod. 2022, 80, 1321–1331. [Google Scholar] [CrossRef]
  53. Gan, W.; Chen, C.; Wang, Z.; Pei, Y.; Ping, W.; Xiao, S.; Dai, J.; Yao, Y.; He, S.; Zhao, B.; et al. Fire-Resistant structural material enabled by an anisotropic thermally conductive hexagonal boron nitride coating. Adv. Funct. Mater. 2020, 30, 1909196. [Google Scholar] [CrossRef]
  54. Chen, G.; Chen, C.; Pei, Y.; He, S.; Liu, Y.; Jiang, B.; Jiao, M.; Gan, W.; Liu, D.; Yang, B.; et al. A strong, flame-retardant, and thermally insulating wood laminate. Chem. Eng. J. 2020, 383, 9. [Google Scholar] [CrossRef]
  55. Fan, C.; Gao, Y.; Li, Y.; Yan, L.; Zhu, D.; Guo, S.; Ou, C.; Wang, Z. A flame-retardant densified wood as robust and fire-safe structural material. Wood Sci. Technol. 2023, 57, 111–134. [Google Scholar] [CrossRef]
  56. Xu, Z.; Zhao, W.; Yan, L.; Tang, X.; Feng, Y.; Wang, Z. Processing of Pinus sylvestris L. into a heat-insulating, thermally stable, and flame-retarded material by combining the flame-retardant impregnation and densification treatment. Holzforschung 2023, 77, 762–775. [Google Scholar] [CrossRef]
  57. Chu, D.; Mu, J.; Avramidis, S.; Rahimi, S.; Liu, S.; Lai, Z. Functionalized Surface Layer on Poplar Wood Fabricated by Fire Retardant and Thermal Densification. Part 1: Compression Recovery and Flammability. Forests 2019, 10, 955. [Google Scholar] [CrossRef]
  58. Spruce. Betterwood. Available online: https://betterwood.co/lexikon/fichte/ (accessed on 13 July 2023).
  59. Spruce Wood. lignoma.com Magazine. Available online: https://www.lignoma.com/en/magazine/spruce-wood/ (accessed on 13 July 2023).
  60. Marková, I.; Tureková, I. Teória Horenia a Analýza Spalín; University of Žilina: Žilina, Slovakia, 2023; 219p, ISBN 978-80-554-1930-5. [Google Scholar]
  61. Kačíková, D.; Makovická-Osvaldová, L. Wood burning rate of various tree parts from selected softwoods. Acta Fac. Xylologiae 2009, 51, 27–32. [Google Scholar]
  62. Draxlerová, M.; Harangózo, J.; Balog, K. Effects of retardants on the ignition of Woodmaterials. Environmental and Safety Aspects of Renewable Materials and Energy Sources. Adv. Mater. Res. 2014, 1001, 288–291. [Google Scholar] [CrossRef]
  63. Martinka, J.; Chrebet, T.; Hrušovský, I.; Balog, K.; Hirle, S. Fire risk assessment of spruce pellets. In Applied Mechanics and Materials; Trans Tech Publications: Baech, Switzerland, 2014; Volume 501–504, pp. 2451–2454. [Google Scholar] [CrossRef]
  64. Xu, Q.; Que, X.; Cao, L.; Jiang, Y.; Jin, C.A. Total heat flux on the wall: Bench scale wood crib fires tests. Therm. Sci. 2010, 14, 283–290. [Google Scholar] [CrossRef]
  65. Kačíková, D.; Balog, K.; Tureková, I.; Mitterová, I. Materials in Fire Protection—University Textbook; Technical University in Zvolen: Zvolen, Slovakia, 2011; 367p, ISBN 978-80-228-2317-3. (In Slovak) [Google Scholar]
Applsci 14 00170 g001
Figure 1. Schematic diagram of the test equipment (description: 1—gas cylinder, 2—shut-off valve, 3—flow regulator, 4—gas intake tube, 5—burner holder, 6—burner, 7—gas flow regulator on the burner, 8—scales, 9—sample holder, 10—sample, 11—connection between scales and computer, 12—computer).
Figure 1. Schematic diagram of the test equipment (description: 1—gas cylinder, 2—shut-off valve, 3—flow regulator, 4—gas intake tube, 5—burner holder, 6—burner, 7—gas flow regulator on the burner, 8—scales, 9—sample holder, 10—sample, 11—connection between scales and computer, 12—computer).
Applsci 14 00170 g001
Applsci 14 00170 g002
Figure 2. Main part of the test equipment during the experiment.
Figure 2. Main part of the test equipment during the experiment.
Applsci 14 00170 g002
Applsci 14 00170 g003
Figure 3. Box plots of the mass loss of each group of samples at 600 s testing time with outliers indicated (red dots).
Figure 3. Box plots of the mass loss of each group of samples at 600 s testing time with outliers indicated (red dots).
Applsci 14 00170 g003
Applsci 14 00170 g004
Figure 4. Average mass loss rate of untreated samples over time.
Figure 4. Average mass loss rate of untreated samples over time.
Applsci 14 00170 g004
Applsci 14 00170 g005
Figure 5. Average mass loss rate of samples treated with retardant 1 over time.
Figure 5. Average mass loss rate of samples treated with retardant 1 over time.
Applsci 14 00170 g005
Applsci 14 00170 g006
Figure 6. Average mass loss rate of samples treated with retardant 2 over time.
Figure 6. Average mass loss rate of samples treated with retardant 2 over time.
Applsci 14 00170 g006
Applsci 14 00170 g007
Figure 7. Average mass loss rate of samples treated with retardant 3 over time.
Figure 7. Average mass loss rate of samples treated with retardant 3 over time.
Applsci 14 00170 g007
Applsci 14 00170 g008
Figure 8. Dependence of the difference of fire spread rate on the difference of sample densities.
Figure 8. Dependence of the difference of fire spread rate on the difference of sample densities.
Applsci 14 00170 g008
Table 1. Designation of individual sample groups.
Table 1. Designation of individual sample groups.
SamplesHigh-Density Wood Samples
(ρ > 450 kg∙m−3)		Low-Density Wood Samples
(ρ < 450 kg∙m−3)
Samples treated with Retardant 1R1-HDR1-LD
Samples treated with Retardant 2R2-HDR2-LD
Samples treated with Retardant 3R3-HDR3-LD
Untreated samplesU-HDU-LD
Table 2. Average sample densities and their variation range before flame retardant application.
Table 2. Average sample densities and their variation range before flame retardant application.
SamplesAverage Density (ρ)
[kg·m−3]		Variation Range
[kg·m−3]		Difference in Density
(HD-LD) (Δρ)
[kg·m−3]
R1-HD560.6947.86560.69 − 317.00 = 243.69
* (218.92–281.51)
R1-LD317.0014.72
R2-HD462.125.79462.12 − 411.47 = 50.65
* (44.87–57.32)
R2-LD411.476.66
R3-HD455.584.62455.58 − 447.14 = 8.44
* (4.62–12.48)
R3-LD447.1412.48
U-HD480.208.75480.20 − 378.54 = 101.66
* (84.79–115.02)
U-LD378.5421.47
* Variation range of density difference.
Table 3. Average values of mass loss of samples after the experiment and the maximum mass loss rate achieved (indicating the standard deviation).
Table 3. Average values of mass loss of samples after the experiment and the maximum mass loss rate achieved (indicating the standard deviation).
SpecimensTotal Mass Loss (δ)
[%]		Standard Deviation of Total Mass Loss (σδ)
[%]		Max Mass Loss Rate (vr max)
[%·s−1]		Standard Deviation of Max Mass Loss Rate (σvr max)
[%·s−1]
R1-HD19.47±1.160.0816±0.0072
R1-LD23.88±2.060.1650±0.0196
R2-HD19.05±1.500.0953±0.0081
R2-LD19.09±1.100.1016±0.0148
R3-HD20.62±1.630.1173±0.0182
R3-LD20.10±0.530.1262±0.0088
U-HD28.32±6.400.1699±0.3986
U-LD44.12±24.800.4036±0.2933
Table 4. ANOVA of sample mass loss at 300 s testing time.
Table 4. ANOVA of sample mass loss at 300 s testing time.
Samples Treated with Retardant 1, Difference in Density 243.69 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)1107.64107.6439.051.56 × 10−5Strong influence
Residuals1541.342.76
Samples Treated with Retardant 2, Difference in Density 50.65 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)10.4240.42450.4530.51No influence
Residuals1614.9820.9364
Samples Treated with Retardant 3, Difference in Density 8.44 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)10.0120.01240.0110.919No influence
Residuals1315.1751.1673
Table 5. ANOVA of sample mass loss at 600 s testing time.
Table 5. ANOVA of sample mass loss at 600 s testing time.
Samples Treated with Retardant 1, Difference in Density 243.69 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)182.2382.2328.388.46 × 10−5Strong influence
Residuals1543.462.90
Samples Treated with Retardant 2, Difference in Density 50.65 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)10.0110.01050.0060.939No influence
Residuals1627.7431.7340
Samples Treated with Retardant 3, Difference in Density 8.44 kg·m−3
DfSum SqMean SqF valuePr (>F)Influence
Density (ρ)11.0111.0110.6450.436No influence
Residuals1320.3721.567
Table 6. Fire spread rate (proportion of the maximum value of the first peak mass loss rate and the time to reach this).
Table 6. Fire spread rate (proportion of the maximum value of the first peak mass loss rate and the time to reach this).
SamplesR1-HDR1-LDR2-HDR2-LDR3-HDR3-LD
Rfs [%·s−2]0.001090.003160.001530.001890.001600.00174
Difference in Rfs [%·s−2]0.002060.000360.00014
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Mitrenga, P.; Osvaldová, L.M.; Konárik, M. Effect of Spruce Wood Density on Selected Fire-Technical Parameters during Thermal Loading. Appl. Sci. 2024, 14, 170. https://doi.org/10.3390/app14010170

AMA Style

Mitrenga P, Osvaldová LM, Konárik M. Effect of Spruce Wood Density on Selected Fire-Technical Parameters during Thermal Loading. Applied Sciences. 2024; 14(1):170. https://doi.org/10.3390/app14010170

Chicago/Turabian Style

Mitrenga, Patrik, Linda Makovická Osvaldová, and Milan Konárik. 2024. "Effect of Spruce Wood Density on Selected Fire-Technical Parameters during Thermal Loading" Applied Sciences 14, no. 1: 170. https://doi.org/10.3390/app14010170

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop