Technical University of Denmark
Browse
VIDEO
Domain_movie_throughout_sample.avi (43.56 MB)
TEXT
README.txt (4.7 kB)
ARCHIVE
Segmentations.zip (67.86 MB)
ARCHIVE
Raw_reconstructions.zip (3.8 GB)
.ZIP
Computed_properties.zip (5.24 GB)
1/0
5 files

Data set for "Freeze-casting uniformity and domains"

dataset
posted on 2024-09-19, 08:27 authored by Rasmus BjørkRasmus Bjørk, Peter Stanley JørgensenPeter Stanley Jørgensen, Cathrine D. Christiansen

The is the data set for the paper [Jørgensen, P. S., Christiansen, C. D. & Bjørk, R., Freeze-casting uniformity and domains, Journal of the European Ceramic Society, 45, 116907, 2025]. The DOI for the publication is 10.1016/j.jeurceramsoc.2024.116907

This dataset contains three folders with the data shown in the article. Besides this, there is a movie showing the domain structure throughout the dataset, as mentioned in the article.

The folder "Raw_reconstructions" contains the raw reconstructed TIF images of the high resolution cutout (4X) and the bottom and top scans in the respective subfolders. These are shown in Fig. 2.

The folder "Segmentations" contains the segmented structure, i.e. split into pores (with a value of 0) and solid (with a value of 1) of the high resolution cutout (4X) and the bottom and top scans in the respective subfolders. The structures are stored as a 3D logical array, corresponding to the physical directions x, y and z, in a Matlab file format. These are shown in Fig. 2 in the article.

The folder "Computed_properties" contains the various computed properties illustrated in the article. The files are individually described below:

Porosity_4X.mat: Contains the porosity as function of the vertical (z) distance for the high resolution cutout shown in Fig. 4 in the article. The data is stored as two arrays in a Matlab file.

PSD_4X.mat: Contains two variables, which are the solid particle size distribution and the pore particle size distribution, shown in Fig 3b in the article. These are two dimensional arrays that in the first column contains the particle diameter in um and in the second column contains the volume covered in percent.

Structure_tensor_bottom.mat: Contains three arrays for the structure tensor for the bottom scan. The first array is the voxel_size array, which simply gives the voxel size also stated in the article. The second array is the image_stack array with dimensions x by y by color by z. There are three colors (RGB) for each image. These are the images visualized in Fig. 6 in the article. The third array is the theta_slice which contains the local orientation angle with dimensions x, y and z. The latter array has been upsampled in x and y, as also described in the article, such that it has onethird the dimensions of the image_stack in x and y.

Structure_tensor_top.mat: Contains three arrays for the structure tensor for the top scan. The variables are identical to Structure_tensor_bottom.mat.

Structure_tensor_domains.mat: Contains two arrays for the structure tensor combined for the top and bottom scans. The array domain_sizes_slice contains the size of the domains with the dimensions of the array being the slice number times the number of color bins, which is taken to be five, times the number of domains with a size corresponding to the number of pixels. As an example the entry domain_sizes_slice(100,1,50) contains the number of domains in color bin 1 with a size of 50 pixels in the 100 vertical slice in the data set. Similarly the array centroid_positions_slice has the same structure. Its first dimension is the slice number, its second dimension is the color bin, its third dimension is a running number for each domain present and the final dimension of two contains the x and y coordinates in pixels of the centroid position of the respective domains. To make the structure an array, there can be empty numbers for high running domain number. As an example centroid_positions_slice(100,1,:,:) contains all domains in vertical slice 100 in color bin 1. There are 365 such domains, meaning that centroid_positions_slice(100,1,366,:) and onwards contains zeros.

Tau_por_dir_1_4X.mat: Contains an array with the relative local flux in the pores in the high resolution cutout in the x-direction, as visualized in Fig. 5 in the article.

Tau_por_dir_2_4X.mat: Contains an array with the relative local flux in the pores in the high resolution cutout in the y-direction, as visualized in Fig. 5 in the article.

Tau_por_dir_3_4X.mat: Contains an array with the relative local flux in the pores in the high resolution cutout in the z-direction, as visualized in Fig. 5 in the article.

Tau_sol_dir_1_4X.mat: Contains an array with the relative local flux in the solid in the high resolution cutout in the x-direction, as visualized in Fig. 5 in the article.

Tau_sol_dir_2_4X.mat: Contains an array with the relative local flux in the solid in the high resolution cutout in the y-direction, as visualized in Fig. 5 in the article.

Tau_sol_dir_3_4X.mat: Contains an array with the relative local flux in the solid in the high resolution cutout in the z-direction, as visualized in Fig. 5 in the article.

Funding

Independent Research Fund Denmark – Technologies and Productions Sciences, project no. 6111-00073B

Novo Nordisk Foundation project NNF21OC0072021 entitled “Superior nature-inspired biomaterials from freeze-casted microchannels”

History

ORCID for corresponding depositor

Usage metrics

    DTU Energy

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC