Dynamic

When setting a tomographic experiment several parameters are adjusted by the instrument operator including X-ray energy, exposure time, frame rate, rotary stage speed, scanning mode (sequential or interlaced angles etc.). Here we include a series of datasets collected at different experimental conditions aiming to capture fast evolving samples in 3D.

Noisy Data

Sometimes data collection speed requirements, e. g. in evolving samples, impose very short exposure times generating noisy data. Here we include a dataset, Dorthe_F_002, in which the exposure time much shorter than the optimal value. The sample description and the experimental conditions are reported in tables below: and accessible for download under tomo_00031.

To load the data sets and perform a basic reconstruction using tomopy

tomopy recon --file-name tomo_00031.h5 --rotation-axis 484.5

tomo_ID	00031
Image preview
Downloads	tomo_00031
Instrument	APS 13-BM-D
Sample name	Dorthe_F_002
Energy	33.269 keV
Monochromator	double crystal Si (1,1,1)
Scan Range	180 degree
Number of Projections	900
White Fields	20 before
Dark Fields	none
Exposure Time	0.006 s
Frame Rate	80 frames/s
Total Collection Time	11.25 s
PixelSize	3.18 µm
Rotation axis location	484.5

Lower Resolution

This study was optimized for temporal resolution and less for spatial resolution. The experiment was originally designed to follow the propagation of a Cs solution with time in a rock. The spatial resolution could be relaxed at the time of the experiment to provide sufficient time resolution.

In the second phase of the project, the focus moved however towards the smaller reactive inclusions and complementary techniques (x-ray microprobe and destructive chemical tomography with LA-ICP-MS) have been used to investigate the distribution of different elements and phases in selected regions, with sometimes higher spatial resolution than in the original tomographic dataset (see [B19], [B3]).

To load the data sets and perform a basic reconstruction using tomopy

tomopy recon --file-name tomo_00069.h5 --rotation-axis 515.50

tomo_ID	00069
Image preview
Downloads	tomo_00069
Instrument	SLS TOMCAT
Sample name	SLS_03/Cskin1_36__B1
Energy	36.085 keV
Sample-to-detector distance	4 mm
Scan Range	180 degree
Number of Projections	1001
White Fields	20 (10 before - 10 after)
Dark Fields	5
PixelSize	3.7 µm
Rotation axis location	515.500232028

Interlaced Scan

A technique adopted to capture fast evolving sample includes interlaced data collection. In this mode multiple series of continuous 0-180 deg datasets are collected with equally spaced sparce angles. After each 0-180 deg rotation the next data set is collected with angular sampling interlaced to the previous scan.

Below we report the sample description and the experimental conditions for an interlace dataset (tomo_00057) [B13] collected at the Elettra Syrmep beamline.

tomo_ID	00057
Image preview
Downloads	tomo_00057
Instrument	Elettra Syrmep
Sample name	Bone MR
Energy	24 keV
Filter	1 mm Al
Sample-to-detector Distance	210 mm
Scan Range	180 degree
Interlaced Data Collection	20 projections x 36 (0-180 deg) iteration
Total Projections	720
White Fields	20
Dark Fields	20
Exposure Time	0.8 s

Foam data

In this study, we investigate the rheology of liquid foams by fast synchrotron X-ray tomographic microscopy [B37]. Foams are complex cellular systems which require artifact free tomographic reconstruction for a reliable quantification of their time-dependent properties such as deformation fields of bubbles. In our example we acquire X-ray projections of the liquid foam flowing through a constriction and being rotated around the tomographic axis. The experiment was performed at the TOMCAT beamline of the Swiss Light Source using the fast acquisition setup [B34].

To load the data sets and perform reconstruction use the tomopy_rectv.py python script.

Reconstruction by Gridrec

python tomopy_rectv.py dk_MCFG_1_p_s1_.h5 --type subset --nsino 0.75 --binning 2 --frame 95

Reconstruction by the method with suppressing motion artifacts [B46] requires module rectv that can be installed from https://github.com/math-vrn/rectv_gpu. In this case, the algorithm run with option –tv True

python tomopy_rectv.py dk_MCFG_1_p_s1_.h5 --type subset --nsino 0.75 --binning 2 --tv True --frame 95

tomo_ID	00080
Image preview
Downloads	tomo_00080
Instrument	SLS TOMCAT
Sample name	dk_MCFG_1_p_s1
Energy	16 keV
Sample-to-detector Distance	250 mm
Scan Range	180 degree
Continuous Data Collection	300 projections x 130 (0-180 deg) iteration
Total Projections	39000
White Fields	512
Dark Fields	512
Exposure Time	0.7 ms
Frame rate	840 deg/s
PixelSize	3 µm
Rotation axis location	1008

Fuel cell data

This data was first provided to TomoBank to be used as part of TomoChallenge, please see https://tomochallenge.github.io/ for more details and for information about how you can participate. Sub-second X-ray tomographic microscopy was exploited to investigate liquid water dynamics in a fuel cell during operation. During the experiment, the cell was rotated continuously around the tomographic axis and three tomographic datasets, each consisting of 60 consecutive scans, were acquired. Each of the three datasets (60 scans) was acquired within 6 seconds, the interval period between each dataset was approximately 7 seconds. The outer boundaries of the cell were slightly outside of the field-of-view, leading to interior tomography. The experiment was performed at the TOMCAT beamline of the Swiss Light Source using the fast acquisition setup coupled with high-numerical-aperture macroscope optics [B4].

At the start of the experiment the cell was completely dry. During operation, water generated on the surface of the catalyst layer (Pt based, bright area in image preview) started to propagate through the porous fiber layer (carbon based gas diffusion layer located between the four channels), emerging in the channels. A large water droplet can be clearly visually detected in the reconstructed channel region after the 115th time step (e.g. slice 364).

The image preview is an example of a fuel cell slice reconstructed by Gridrec from phase retrieved [B12] projections and cropped to the region of interest. The flow field (FF) consists of carbon based material, the gas diffusion layer (GDL) of carbon fiber and the catalyst coated membrane (CCM) of a polymer, coated with a Pt based catalyst. Image courtesy of Hong Xu (Paul Scherrer Institut).

To load the dataset and perform basic reconstruction use the the tomopy_rectv_fc.py python script

python tomopy_rectv_fc.py fuelcell_i1.h5

tomo_ID	00081
Image preview
Downloads	tomo_00081
Instrument	SLS TOMCAT
Sample name	fuelcell_i1, fuelcell_i2, fuelcell_i3
Energy	Polychromatic radiation (filtered, mean ~ 30 keV)
Sample-to-detector Distance	30 mm
Scan Range	180 degree
Continuous Data Collection	301 projections x 60 (0-180 deg) iteration
Total Projections	18060
White Fields	100
Dark Fields	10
Exposure Time	0.3 ms
Frame Time	0.33 ms
PixelSize	2.75 µm
Rotation axis location	702

In addition to the dynamic datasets, a high-quality post operando scan of the fuel cell in dry state is provided. The scan parameters are specified in the table below.

To load the high-quality dataset and perform basic reconstruction use the tomopy_rectv_fc.py python script:

python tomopy_rectv_fc.py fuelcell_dryHQ_i1.h5 --rotation-axis 702.00 --nproj 1001 --ntframes 1

tomo_ID	00082
Image preview
Downloads	tomo_00082
Instrument	SLS TOMCAT
Sample name	fuelcell_dryHQ_i1
Energy	Polychromatic radiation (filtered, mean ~ 30 keV)
Sample-to-detector Distance	30 mm
Scan Range	180 degree
Total Projections	1001
White Fields	100
Dark Fields	10
Exposure Time	1 ms
PixelSize	2.75 µm
Rotation axis location	702