Understanding the Effect of Knots on Mechanical Properties of Chinese Fir under Bending Test by Using X-ray Computed Tomography and Digital Image Correlation

Round 1

Reviewer 1 Report

Overall, it's a very interesting piece of work and will be of interest to Forests readers.

Abstract: It seems to be fine.

1. Introduction. It is very well-written, the information presented is clear, the scientific gaps are well-stated, the necessary bibliographic references were cited and the objectives are clear.

2. MEthodology. I consider that the preparation of the samples, procedures and data processing were well explained. They were clear. Figure 1 is a very important one. Then, I kindly recommend improving its quality because, in the way that it is, it is very hard to understand. 
Also, I have one question, why have you decided to work with samples containing only sapwood? Considering that heartwood, in certain wood species, has more biological durability and better mechanical properties?

Another question, which standard did you use in the mechanical testing of small clear wood samples and the small samples with knots?

Another question, you worked with 20 samples free-of-knots and 40 with knots. It is understandable to use more samples with knots due to the variability. However, is there any minimum number for both wood groups according to the used standard?

3. Results and DIscussions: All the Figures presented in this section must have a better quality. They are very important, but they have very low quality. 
In Figure 2, what exactly is the difference between "FK" and "FK-A or FK-I"?

As presented by you, "There is no obvious difference of MOE between specimens without knots, knot specimens failure at knots and knot specimens not failure at knots". To what would the authors attribute this?

The authors stated "The existence of knots, therefore, has an influence on the mechanical properties of the wood. The way of influence is significantly variable according to the failure patterns. Specifically, the impact of knots on MOE and MOR is small when the specimens do not failure at knots or failure around knots. Both MOE and MOR can be significantly lower when the specimens failure in knots". Although the finding is true and supported by the results, it is something that is already widely known and does not necessarily bring anything new. Therefore, I recommend rewriting this part to base it on what is already in the literature or something like that.

The authors stated "Internal structural changes instead of structural failure of the specimens contributes to energy absorption and loading force increase. Large plastic domain means big internal structural changes, resulting in large loading force and high MOR." I kindly recommend using references in these sentences because that is a type of information already explained in the literature.

The quality of Figure 6 I very low, which makes it very hard to understand the results. 

4. Conclusions: The conclusion should be sharpened and the ballast removed as it is too lengthy.

I think that the great advantage of this work is that it uses X-ray and DIC to explain the phenomena involved when the specimens are subjected to static bending tests. Therefore, I think that the authors could discuss a little more about this and use more references to contrast or even substantiate the explanations of the results they present.

Minor editing should be done, mainly regarding the editing and writing.

Author Response

Overall, it's a very interesting piece of work and will be of interest to Forests readers.

Abstract: It seems to be fine.

1. Introduction. It is very well-written, the information presented is clear, the scientific gaps are well-stated, the necessary bibliographic references were cited and the objectives are clear.
   Revision: Thanks!
   2. Methodology. I consider that the preparation of the samples, procedures and data processing were well explained. They were clear. Figure 1 is a very important one. Then, I kindly recommend improving its quality because, in the way that it is, it is very hard to understand. 

Revision: The quality of this image has been improved. If necessary, it will be further changed based on the requirements from editors.
Also, I have one question, why have you decided to work with samples containing only sapwood? Considering that heartwood, in certain wood species, has more biological durability and better mechanical properties?
Revision: There are two reasons of this issue. First, the age of trees used in this study is around 9 years and there is limited heartwood in the trees. Second, knots mainly occur in sapwood instead of heartwood. As such, the specimens were cut from sapwood only. The above explanation has been added in manuscript.
Another question, which standard did you use in the mechanical testing of small clear wood samples and the small samples with knots?
Revision: The standard used in test is GB/T 1936.1-2009. It has been added in manuscript.
Another question, you worked with 20 samples free-of-knots and 40 with knots. It is understandable to use more samples with knots due to the variability. However, is there any minimum number for both wood groups according to the used standard?

Revision: Based on the standard, there is no specific requirement for the amount of replicate. We tested 20 and 40 samples that is much more than commonly used 6 replicates.
3. Results and Discussions: All the Figures presented in this section must have a better quality. They are very important, but they have very low quality. 

Revision: The figures will be further changed based on the requirements of the journal.
In Figure 2, what exactly is the difference between "FK" and "FK-A or FK-I"?
Revision: The difference between "FK" and "FK-A or FK-I" is explained in the caption of Fig.2. Fig.3 also presents the meaning of "FK" and "FK-A or FK-I".
As presented by you, "There is no obvious difference of MOE between specimens without knots, knot specimens failure at knots and knot specimens not failure at knots". To what would the authors attribute this?
Revision: This phenomenon is due to shear strain transferal and internal structural changes. The detailed explanation is given in manuscript.
The authors stated "The existence of knots, therefore, has an influence on the mechanical properties of the wood. The way of influence is significantly variable according to the failure patterns. Specifically, the impact of knots on MOE and MOR is small when the specimens do not failure at knots or failure around knots. Both MOE and MOR can be significantly lower when the specimens failure in knots". Although the finding is true and supported by the results, it is something that is already widely known and does not necessarily bring anything new. Therefore, I recommend rewriting this part to base it on what is already in the literature or something like that.
Revision: Indeed, the findings are somehow straight forward but it is still necessary to understand the reasons behind the facts. This part has been declared based on the related literatures.

The authors stated "Internal structural changes instead of structural failure of the specimens contributes to energy absorption and loading force increase. Large plastic domain means big internal structural changes, resulting in large loading force and high MOR." I kindly recommend using references in these sentences because that is a type of information already explained in the literature.

Revision: Thanks for your suggestion. This statement is according to reference 20, which has been declared in manuscript.

The quality of Figure 6 is very low, which makes it very hard to understand the results. 
Revision: The original figure will be submitted based on the requirements of the journal.
4. Conclusions: The conclusion should be sharpened and the ballast removed as it is too lengthy.

Revision: The Conclusion has been shortened.

I think that the great advantage of this work is that it uses X-ray and DIC to explain the phenomena involved when the specimens are subjected to static bending tests. Therefore, I think that the authors could discuss a little more about this and use more references to contrast or even substantiate the explanations of the results they present.

Revision: More discussion and references have been added in manuscript. The importance of tracking failure process and visualizing internal structure has been highlighted.

Reviewer 2 Report

Thank You Authors for this research!

Please increase the quality of all figures (especially Figure 1), numbers and descriptions are not visible - understandable.

See also some comments in the Manuscript text.

Good luck for the next researches!

Comments for author File: Comments.pdf

Author Response

1.Abstract: MOE and MOR for first time show the full names, also in case they are well known.

Another question, why did You choose three and not four point bending, if we know, that for timber as construction material usually we use four point bending e.g. EN 408 etc.? If the answer is shown in rows 135-142, then it is understandable. Another question, it's better to modify ‘Bending test’ to ‘Three point bending’.

Revision: Full name of MOE and MOR has been added. ‘Bending test’ has been changed to ‘three point bending’. The effect of knots on mechanical strength of wood is mainly caused by shear stress. As such, three point bending test is used.

2. Several places in the text should be modified to the superscript of the specification.

Revision: Thanks for your suggestion, it has been revised.

3. Materials and Methods: Cunninghamia lanceolata should be changed to Latin name in Italic.

Another question, dimensions? 300x20x20 mm and not a 300x20x20 mm³.

Revision: Thanks for your suggestion, it has been revised.

Another question, sapwood of these trees regarding with any kind of the standard?

Revision: The age of trees used in this study is around 9 years and there is limited heartwood in the trees. This has been explained in manuscript.

Another question, what were the environmental parameters?

Revision: The environmental parameters are 65%RH and 20℃, which has been added in manuscript.

Another question, what were the orientation of the wood fibers-annual rings in the cross section of the samples?

Revision: Indeed, the orientation of annual rings has non-trivial effect on mechanical properties of the specimens. The direction of loading force is nearly parallel to tangential direction of the specimens, which has been declared in manuscript.

Another question, its not very clear... 20 specimens or two groups by 10 specimens (Clear-1 and Clear-2), understandable. How about 40 specimens or four groups by ten samples? If results (Table 2) show four groups (FK-1, FK-2, FK-3 and FK-4) and four groups (FNK-1, FNK-2, FNK-3 and FNK-4), or each group by 10 specimens. So totally 80 specimens?

Revision: There are 20 specimens without knots and 40 specimens with knots, as such 60 specimens in total. The specimens with knots were further analyzed according to failure types, such as FK, FNK. This has been explained in manuscript.

Another question, is it regarding any standard, I mean distance between lower supports and test speed.

Most of the standards says that specimen should be broken at special time. And the test speed should be find out after test of couple of specimens.

Revision: Indeed, the loading speed was set based on pre-test. On average, the specimens broken at 60s.

Another question, maybe it is fault of my screen...The quality of the all figures need to be show higher resolution.

Revision: The figures will be revised based on the requirements of the journal.

Another question, evaluate the grades of wood reference?

Revision: This parameter is from Chinese national standard (GB/T 155-2017). It has been added in manuscript.

4. Results and discussion: Please, add full name for first and second Y axis. And also for X axis as well (specimens group).

Revision: The full name is quite long and they are very hard to add in axis. Considering they are well explained in the caption, this comment is not revised accordingly.

Another question, how about mean density each group. Could you show in this table?

Revision: The density of these specimens has been added in Table 2.

Another question, to my mind should be rounded to full number or e.g. 5704 MPa etc. Also in the text. Three number, to my mind, should be given or 51.7 MPa. Also in the text. Also here the same 15.9 %. Here make equalization of the results (tree numbers) e.g. 4.00 and 73.5.

Revision: Thanks for you suggestion. In manuscript, the numbers were set to round to 2 decimal places. It will be further adjusted according to the specific requirements of the journal.

5. Conclusions: If suitable, also in Conclusions, to my mind for first time show full characterization and then abbreviation, because some researcher starts with conclusions of the article or "compress thoughts" of any article.

Another question, this is good suggestion for the other researchers for upcoming researches.

Revision: Full name of abbreviations has been added.

Reviewer 3 Report

Understanding the effect of knots on mechanical properties of Chinese fir under bending test by using X-ray CT and DIC

The influence of knots on the mechanical properties of wooden elements is the subject of many scientific articles. They generally indicate a decrease in strength properties as a result of the occurrence of knots in the tested elements. Generally, the authors of these studies used large samples with a length of over 1000 mm. In this article, the authors used small samples which, according to the standards, should not contain wood defects (including knots) in this type of tests. The question arises to what extent tests carried out on small samples provide reliable results, in particular with regard to determining the influence of knots on bending strength. The authors should indicate and explain the grounds on which they used this type of samples for research. Nevertheless, it should be noted that the authors proposed new methods of sample analysis, such as: X-ray computed tomography (X-ray CT) or digital image correlation (DIC), which may be a source of additional information on the impact of knots on the strength of the tested material. In connection with the above, the authors should clarify the following issues:

·     what were the reasons for using small samples for research?

·     the authors state that the samples were obtained from 9-year-old trees. There is no information on how many trees samples were obtained from. The authors should also better characterize the trees from which they obtained samples, e.g. provide the trunk diameter and length, annual rings width, how many samples were obtained from one trunk and from which part of the trunk the samples came from (butt part, top part);

·     the authors report that they prepared 20 samples without knots and 40 samples with knots. How the samples were obtained, i.e. whether the material was cut randomly and then samples without and with knots were selected. Were there any additional criteria for selecting samples with knots, e.g. location of knots, sample cross-sectional area? For example, Fig. 3 shows 3 samples with knots, one of which (marked FNK) is characterized by a much smaller share of knots than the other 2. Similarly, in Fig. 6 and 7 show samples with knots FNK-4 and FK-4, respectively, but it is difficult to conclude that these samples can be compared in tests due to the completely different share of knots;

·     when calculating HTR, the authors refer only to the height of the knot in relation to the height of the sample. Shouldn't the height and width of the knot be analyzed, as well as the way it penetrates the sample (e.g. whether it passes onto a parallel surface);

·  the methodology lacks information about the statistical analysis methods used;

·     the authors unnecessarily repeat the explanations of abbreviations, e.g.: FK, FNK, FK-A, FK-I, HTR, KCD;

·     the readability of figures, in particular 1, 6, 7, should be improved.

Taking into account the above remarks, the article requires a revision.

Author Response

The influence of knots on the mechanical properties of wooden elements is the subject of many scientific articles. They generally indicate a decrease in strength properties as a result of the occurrence of knots in the tested elements. Generally, the authors of these studies used large samples with a length of over 1000 mm. In this article, the authors used small samples which, according to the standards, should not contain wood defects (including knots) in this type of tests. The question arises to what extent tests carried out on small samples provide reliable results, in particular with regard to determining the influence of knots on bending strength. The authors should indicate and explain the grounds on which they used this type of samples for research. Nevertheless, it should be noted that the authors proposed new methods of sample analysis, such as: X-ray computed tomography (X-ray CT) or digital image correlation (DIC), which may be a source of additional information on the impact of knots on the strength of the tested material. In connection with the above, the authors should clarify the following issues:

what were the reasons for using small samples for research?

Revision: The small samples were used based on the standard, which has been added in manuscript. The aim of this study is to deeply understand the effect of knots on mechanical properties of wood. Hence, DIC and X-ray CT were used to unravel the story based on small samples. This idea has been added in manuscript.

the authors state that the samples were obtained from 9-year-old trees. There is no information on how many trees samples were obtained from. The authors should also better characterize the trees from which they obtained samples, e.g. provide the trunk diameter and length, annual rings width, how many samples were obtained from one trunk and from which part of the trunk the samples came from (butt part, top part);

Revision: The samples were cut from the trunks of 5 trees and the diameter of the trees is approximately 120-150mm. The trunks were obtained from 1m to 8m height of the trees. This information has been added in manuscript. For other information, such as annual ring width and how many samples were obtained from one truck, it has not been added in manuscript due to lack of data.

the authors report that they prepared 20 samples without knots and 40 samples with knots. How the samples were obtained, i.e. whether the material was cut randomly and then samples without and with knots were selected. Were there any additional criteria for selecting samples with knots, e.g. location of knots, sample cross-sectional area? For example, Fig. 3 shows 3 samples with knots, one of which (marked FNK) is characterized by a much smaller share of knots than the other 2. Similarly, in Fig. 6 and 7 show samples with knots FNK-4 and FK-4, respectively, but it is difficult to conclude that these samples can be compared in tests due to the completely different share of knots;

Revision: The samples were cut randomly and then the specimens without and with knots were selected. Indeed, many knot factors could affect the mechanical performances of wood. Hence, we only selected the specimens with one knot located at the bottom part of the specimens. Meanwhile, the position of the knots cannot be very far from the loading position. Based on above reasons, the HTR and KCD were analyzed in detail.

when calculating HTR, the authors refer only to the height of the knot in relation to the height of the sample. Shouldn't the height and width of the knot be analyzed, as well as the way it penetrates the sample (e.g. whether it passes onto a parallel surface);

Revision: Based on the standard, the height of the knot is regarded as a crucial parameter instead of width. The related standard has been added in manuscript. The knots penetration data is hard to obtain and not used widely, therefore, this parameter was not used.

the methodology lacks information about the statistical analysis methods used;

Revision: The statistical analysis was performed based on two tails t test using SPSS. Above information has been added in manuscript.

the authors unnecessarily repeat the explanations of abbreviations, e.g.: FK, FNK, FK-A, FK-I, HTR, KCD;

Revision: Other reviewers suggested that the abbreviations should be explained at more places. Hence, the explanations have not been removed.

the readability of figures, in particular 1, 6, 7, should be improved.

Revision: The figures will be revised based on the requirements of the journal.

Round 2

Reviewer 1 Report

I congratulate the authors for making the necessary modifications. 
I still have some other comments:

1) The authors wrote: "There was limited heartwood and the specimens, measuring 300×20×20mm, were cut from the sapwood of these trees". I would kindly recommend elaborating on this by explaining why you only work with sapwood. In addition, regarding the dimensions, I kindly recommend rewriting them differently, for example, 300mm x 20mm x 20mm, or 300( (longitudinal) x 20 (radial) x 20 (tangential), or however you prefer.

2) Regarding the Standard, it seems that you did not mention it in the References. Another question, is it used for both, wood samples with and without knots? I am sorry for asking but I am not familiar with this one. I am used to working with the American, ISO and EN.

Author Response

1) The authors wrote: "There was limited heartwood and the specimens, measuring 300×20×20mm, were cut from the sapwood of these trees". I would kindly recommend elaborating on this by explaining why you only work with sapwood. In addition, regarding the dimensions, I kindly recommend rewriting them differently, for example, 300mm x 20mm x 20mm, or 300( (longitudinal) x 20 (radial) x 20 (tangential), or however you prefer.

 Revision: The following explanation has been used in manuscript. Chinese fir (Cunninghamia lanceolata) trees, nine years old, were harvested in a planted forest located in Zhuzhou city of Hunan province in China. There was only limited heartwood and the trees mainly consisted with sapwood. Hence, the specimens, measuring 300mm×20mm×20mm, were cut from the sapwood of these trees.

2) Regarding the Standard, it seems that you did not mention it in the References. Another question, is it used for both, wood samples with and without knots? I am sorry for asking but I am not familiar with this one. I am used to working with the American, ISO and EN.

Revision: These standards are used to evaluate the mechanical properties of wood specimens without obvious defects. However, there are no specific standards to measure the mechanical properties of specimens with specific defects. Meanwhile, the aim of this work is to compare the difference between normal specimens and the specimens with defects. Hence, this standard was used in this study. The standards have been added in reference list.

Reviewer 3 Report

Understanding the effect of knots on mechanical properties of Chinese fir under bending test by using X-ray CT and DIC

The influence of knots on the mechanical properties of wooden elements is the subject of many scientific articles. Generally, they indicate deterioration of strength properties due to the presence of knots in the tested elements. Nevertheless, it should be noted that the authors of the reviewed article proposed new methods of sample analysis, such as X-ray computed tomography (X-ray) or digital image correlation (DIC), which may be a source of additional information on the influence of nodes on the strength of the tested material.

Compared to the previous version of the article, the authors made appropriate clarifications and introduced changes suggested by the reviewer. The article has been supplemented and corrected.

Author Response

The influence of knots on the mechanical properties of wooden elements is the subject of many scientific articles. Generally, they indicate deterioration of strength properties due to the presence of knots in the tested elements. Nevertheless, it should be noted that the authors of the reviewed article proposed new methods of sample analysis, such as X-ray computed tomography (X-ray) or digital image correlation (DIC), which may be a source of additional information on the influence of knots on the strength of the tested material.

Compared to the previous version of the article, the authors made appropriate clarifications and introduced changes suggested by the reviewer. The article has been supplemented and corrected.

Revision: Thanks!