Find research datasets worth reusing

opencc-by-4.0Apr 2024View details →

Supplementary data for the paper: "Resilient crystalline admixture in ultra-high performance self-healing concrete under cyclic freeze-thaw with de-icing salts"

Supplementary data for the paper: "Resilient crystalline admixture in ultra-high performance self-healing concrete under cyclic freeze-thaw with de-icing salts" Open data concerning experimental work. This study investigates the influence of a crystalline admixture (CA) in Ultra-high performance (fibre-reinforced) concrete under freeze-thaw (FT) cycles with de-icing salts with focus on single cracks with a width of around 120 µm, specifically focusing on the ability of the healing products of CA to survive and the ability to re-heal after a healing regime following FT exposure.

Single-crystalline Al tips

This repository contains all input data for simulating the field-evaporation process of specifically oriented single-crystalline APT tips with representing f.c.c. aluminium using TAPSim. Having conducted all simulations, it contains all detector hit information for all ions and each tip. The authors have also personally stored the trajectories of all ions for each tip. However, given the size of these particular additional binary data (approx. 100 - 200GB), they have --- as a compromise --- not yet been uploaded.  They are available from the authors for free upon request. The individual components of the dataset are as follows, where the * is a placeholder/wildcard. *SX_DetectorHits.zip The simulation results here specifically all detector hit maps for each tip and the input files for tapsim *.cif The Crystallography Open Database file used to define the unit cell *MeshgenInput.tar.gz The compressed txt-based inputfiles for MESHGEN with which the inputfiles for TAPSim for each tip were generated. *TapsimInput.tar.gz the compressed binary inputfiles for TAPSim which specify at which locations the atoms and mesh supporting points are located *BaseAndTip.zip The compressed PNG-renditions of where the tip ions and the base plate supporting points are located. The potential Moire effects apparent in the images are likely related to us reducing the rendering resolution. *Everything.zip The compressed PNG-renditions of where the tip ions, the base plate supporting points, and mesh points are located. *Meta.mat and *PRNG.txt The MATLAB workspace variables which hold the orientation of each tip as a rotation matrix and the seed of the random number generator. The rotation matrix is of active sense premultiplying point position vectors from the VESTA output into their actual locations from which subsequently the tip was carved. A few tip orientations did not run, this was solved by rotating them by an additional 0.5 deg random rotor in space about the projected orientation, the corresponding individual rotation matrices are stored in individual *.mat files with the corresponding tip ID in the filename each *.vesta The Vesta inputfile with which sufficiently sized lattice point clouds were written out to subsequently rotate them and carve into a tip. *DetectorHits.tar.gz The compressed raw data in the default TAPSim format encoding where each ion hit the detector surplus additional pieces of information. *.pl Perl scripts with which to extract runtime profile information from the TAPSim command line output *.sh Shell scripts used to execute the batch queue. Two scripts were used. The first, "run_queue_tapsim_job.sh" $1 $2 fires off independent sequential TAPSIM simulations on existent data The second "run_single_tapsim_job.sh" is called by the first script to fire this of. tapsim_r3225.zip The source code used to execute the simulations The *.tar.gz files in the *.zip archives can be unpacked with the unpack_dataset.sh shell script specifically unpack_dataset.sh $1 $2 with the first argument $1 the start ID and the second argument $2 being the tip end ID to unpack. Modification of the script might be necessary if file name of *.tar.gz archive has been changed... Single *.tar.gz files in general can be most conveniently extracted with a Unix/Linux operating system via for instance the command: tar -xvf nameofthefile.tar.gz On Windows WinRAR can be used for decompression

Single-crystalline Mg tips

This repository contains all input data for simulating the field-evaporation process of specifically oriented single-crystalline APT tips with representing h.c.p. magnesium using TAPSim. Having conducted all simulations, it contains all detector hit information for all ions and each tip. The authors have also personally stored the trajectories of all ions for each tip. However, given the size of these particular additional binary data (approx. 100 - 200GB), they have --- as a compromise --- not yet been uploaded.  They are available from the authors for free upon request. The individual components of the dataset are as follows, where the * is a placeholder/wildcard. *SX_DetectorHits.zip The simulation results here specifically all detector hit maps for each tip and the input files for tapsim *.cif The Crystallography Open Database file used to define the unit cell *MeshgenInput.tar.gz The compressed txt-based inputfiles for MESHGEN with which the inputfiles for TAPSim for each tip were generated. *TapsimInput.tar.gz the compressed binary inputfiles for TAPSim which specify at which locations the atoms and mesh supporting points are located *BaseAndTip.zip The compressed PNG-renditions of where the tip ions and the base plate supporting points are located. The potential Moire effects apparent in the images are likely related to us reducing the rendering resolution. *Everything.zip The compressed PNG-renditions of where the tip ions, the base plate supporting points, and mesh points are located. *Meta.mat and *PRNG.txt The MATLAB workspace variables which hold the orientation of each tip as a rotation matrix and the seed of the random number generator. The rotation matrix is of active sense premultiplying point position vectors from the VESTA output into their actual locations from which subsequently the tip was carved. A few tip orientations did not run, this was solved by rotating them by an additional 0.5 deg random rotor in space about the projected orientation, the corresponding individual rotation matrices are stored in individual *.mat files with the corresponding tip ID in the filename each *.vesta The Vesta inputfile with which sufficiently sized lattice point clouds were written out to subsequently rotate them and carve into a tip. *DetectorHits.tar.gz The compressed raw data in the default TAPSim format encoding where each ion hit the detector surplus additional pieces of information. *.pl Perl scripts with which to extract runtime profile information from the TAPSim command line output *.sh Shell scripts used to execute the batch queue. Two scripts were used. The first, "run_queue_tapsim_job.sh" $1 $2 fires off independent sequential TAPSIM simulations on existent data The second "run_single_tapsim_job.sh" is called by the first script to fire this of. tapsim_r3225.zip The source code used to execute the simulations The *.tar.gz files in the *.zip archives can be unpacked with the unpack_dataset.sh shell script specifically unpack_dataset.sh $1 $2 with the first argument $1 the start ID and the second argument $2 being the tip end ID to unpack. Modification of the script might be necessary if file name of *.tar.gz archive has been changed... Single *.tar.gz files in general can be most conveniently extracted with a Unix/Linux operating system via for instance the command: tar -xvf nameofthefile.tar.gz On Windows WinRAR can be used for decompression

Single-crystalline W tips

This repository contains all input data for simulating the field-evaporation process of specifically oriented single-crystalline APT tips with representing b.c.c. tungsten using TAPSim. Having conducted all simulations, it contains all detector hit information for all ions and each tip. The authors have also personally stored the trajectories of all ions for each tip. However, given the size of these particular additional binary data (approx. 100 - 200GB), they have --- as a compromise --- not yet been uploaded.  They are available from the authors for free upon request. The individual components of the dataset are as follows, where the * is a placeholder/wildcard. *SX_DetectorHits.zip The simulation results here specifically all detector hit maps for each tip and the input files for tapsim *.cif The Crystallography Open Database file used to define the unit cell *MeshgenInput.tar.gz The compressed txt-based inputfiles for MESHGEN with which the inputfiles for TAPSim for each tip were generated. *TapsimInput.tar.gz the compressed binary inputfiles for TAPSim which specify at which locations the atoms and mesh supporting points are located *BaseAndTip.zip The compressed PNG-renditions of where the tip ions and the base plate supporting points are located. The potential Moire effects apparent in the images are likely related to us reducing the rendering resolution. *Everything.zip The compressed PNG-renditions of where the tip ions, the base plate supporting points, and mesh points are located. *Meta.mat and *PRNG.txt The MATLAB workspace variables which hold the orientation of each tip as a rotation matrix and the seed of the random number generator. The rotation matrix is of active sense premultiplying point position vectors from the VESTA output into their actual locations from which subsequently the tip was carved. A few tip orientations did not run, this was solved by rotating them by an additional 0.5 deg random rotor in space about the projected orientation, the corresponding individual rotation matrices are stored in individual *.mat files with the corresponding tip ID in the filename each *.vesta The Vesta inputfile with which sufficiently sized lattice point clouds were written out to subsequently rotate them and carve into a tip. *DetectorHits.tar.gz The compressed raw data in the default TAPSim format encoding where each ion hit the detector surplus additional pieces of information. *.pl Perl scripts with which to extract runtime profile information from the TAPSim command line output *.sh Shell scripts used to execute the batch queue. Two scripts were used. The first, "run_queue_tapsim_job.sh" $1 $2 fires off independent sequential TAPSIM simulations on existent data The second "run_single_tapsim_job.sh" is called by the first script to fire this of. tapsim_r3225.zip The source code used to execute the simulations The *.tar.gz files in the *.zip archives can be unpacked with the unpack_dataset.sh shell script specifically unpack_dataset.sh $1 $2 with the first argument $1 the start ID and the second argument $2 being the tip end ID to unpack. Modification of the script might be necessary if file name of *.tar.gz archive has been changed... Single *.tar.gz files in general can be most conveniently extracted with a Unix/Linux operating system via for instance the command: tar -xvf nameofthefile.tar.gz On Windows WinRAR can be used for decompression

Single-crystalline Zr tips

This repository contains all input data for simulating the field-evaporation process of specifically oriented single-crystalline APT tips with representing h.c.p. zirconium using TAPSim. Having conducted all simulations, it contains all detector hit information for all ions and each tip. The authors have also personally stored the trajectories of all ions for each tip. However, given the size of these particular additional binary data (approx. 100 - 200GB), they have --- as a compromise --- not yet been uploaded.  They are available from the authors for free upon request. The individual components of the dataset are as follows, where the * is a placeholder/wildcard. *SX_DetectorHits.zip The simulation results here specifically all detector hit maps for each tip and the input files for tapsim *.cif The Crystallography Open Database file used to define the unit cell *MeshgenInput.tar.gz The compressed txt-based inputfiles for MESHGEN with which the inputfiles for TAPSim for each tip were generated. *TapsimInput.tar.gz the compressed binary inputfiles for TAPSim which specify at which locations the atoms and mesh supporting points are located *BaseAndTip.zip The compressed PNG-renditions of where the tip ions and the base plate supporting points are located. The potential Moire effects apparent in the images are likely related to us reducing the rendering resolution. *Everything.zip The compressed PNG-renditions of where the tip ions, the base plate supporting points, and mesh points are located. *Meta.mat and *PRNG.txt The MATLAB workspace variables which hold the orientation of each tip as a rotation matrix and the seed of the random number generator. The rotation matrix is of active sense premultiplying point position vectors from the VESTA output into their actual locations from which subsequently the tip was carved. A few tip orientations did not run, this was solved by rotating them by an additional 0.5 deg random rotor in space about the projected orientation, the corresponding individual rotation matrices are stored in individual *.mat files with the corresponding tip ID in the filename each *.vesta The Vesta inputfile with which sufficiently sized lattice point clouds were written out to subsequently rotate them and carve into a tip. *DetectorHits.tar.gz The compressed raw data in the default TAPSim format encoding where each ion hit the detector surplus additional pieces of information. *.pl Perl scripts with which to extract runtime profile information from the TAPSim command line output *.sh Shell scripts used to execute the batch queue. Two scripts were used. The first, "run_queue_tapsim_job.sh" $1 $2 fires off independent sequential TAPSIM simulations on existent data The second "run_single_tapsim_job.sh" is called by the first script to fire this of. tapsim_r3225.zip The source code used to execute the simulations The *.tar.gz files in the *.zip archives can be unpacked with the unpack_dataset.sh shell script specifically unpack_dataset.sh $1 $2 with the first argument $1 the start ID and the second argument $2 being the tip end ID to unpack. Modification of the script might be necessary if file name of *.tar.gz archive has been changed... Single *.tar.gz files in general can be most conveniently extracted with a Unix/Linux operating system via for instance the command: tar -xvf nameofthefile.tar.gz On Windows WinRAR can be used for decompression

opencc-by-4.0Feb 2023View details →

Unit cells and resummed interactions for the calculations in "Systematic Analysis of Crystalline Phases in Bosonic Lattice Models with Algebraically Decaying Density-Density Interactions"

This directory contains all the unit cells with the respective resummed interactions used for the optimisation procedure to obtain the results in the work "Systematic Analysis of Crystalline Phases in Bosonic Lattice Models with Algebraically Decaying Density-Density Interactions[1]". To get an overview of the organization of the directory and a description of the data we recommend the README file. [1]: J. A. Koziol et al., Systematic Analysis of Crystalline Phases in Bosonic Lattice Models with Algebraically Decaying Density-Density Interactions, <a href="https://10.21468/SciPostPhys.14.5.136">10.21468/SciPostPhys.14.5.136</a>, 2023

opencc-by-4.0Jul 2023View details →

Dataset for "A general purpose potential for glassy and crystalline phases of Cu-Zr alloys based on the ACE formalism"

This dataset was used to fit a general purpose machine learning interatomic potential for the Cu-Zr system. It supports the paper "A general purpose potential for glassy and crystalline phases of Cu-Zr alloys based on the ACE formalism".

opencc-by-4.0Oct 2023View details →

Dataset for "Light-Induced Non-Thermal Phase Transition To The Topological Crystalline Insulator State In SnSe"

Dataset relative to the publication "Light-Induced Non-Thermal Phase Transition To The Topological Crystalline Insulator State In SnSe" Stefano Mocatti, Giovanni Marini, Matteo Calandra J. Phys. Chem. Lett. 2023, 14, XXX, 9329–9334 DOI: <a href="https://doi.org/10.1021/acs.jpclett.3c02450">10.1021/acs.jpclett.3c02450</a> The dataset contains the following folders Conv: Convergence tests for total energy, stress tensor and phonon frequencies Pseudo: All pseudopotentials used  Figures: All the data and script to reproduce the figures reported in the manuscript Fign: Simulation data referred to the nth figure of the manuscript The dataset contains also the preprint version of the manuscript (preprint.pdf) and the supporting information file (SI.pdf)

opencc-by-4.0Nov 2020View details →

The gamaproteobacterium Achromatium forms intracellular amorphous calcium carbonate and not (crystalline) calcite- dataset

This is the dataset accompanying the paper published in geobiology and titled The gamaproteobacterium Achromatium forms intracellular amorphous calcium carbonate and not (crystalline) calcite It comprises SEM and light microscopy images, Raman spectra (txt files) and one excel files containing the results of Raman spectra fits.

opencc-by-4.0Dec 2018View details →

Рис. 6. Поперечные (А–Е, З, И) и продольный (Ж) среЗы череЗ органы пиЩеварительной системы моллюска: А – меШок кристаллического стебелька, Б, В – петли киШки, Г, Д – часть желудка с длинными складками на дорсальной части стенки, Е, З – петли киШки с тифлоЗолем, И – боковаЯ часть тифлоЗолЯ. МасШтабные линейки 300 мкм (Е, З, И), 200 мкм (А, В, Г, Д), 40 мкм (Б). мс – меШок кристаллического стебелька, Кс – кристаллический стебелек, Р – реснички, т – тифлоЗоль, lp – lamina propria, г – гемолимфа. Fig. 6. Transverse (А–Е, З, И) and saggital (Ж) sections of the digestive organs: А – crystalline style sac, Б, В – intestinal limbs, Г, Д – part of the stomach with long plicae on the dorsal wall, Е, З – intestinal limbs with typhlosole, И – lateral part of the typhlosole. Scale bars 300 µm (Е, З, И), 200 µm (А, В, Г, Д), 40 µm (B). мс – style sac, Кс – crystalline style, Р – cilia, т – typhlosole, lp – lamina propria, г – hemolymph. in Nodularia vladivostokensis (Bivalvia: Unionidae) from Razdolnaya River (Primorye, Russia)

Рис. 6. Поперечные (А–Е, З, И) и продольный (Ж) среЗы череЗ органы пиЩеварительной системы моллюска: А – меШок кристаллического стебелька, Б, В – петли киШки, Г, Д – часть желудка с длинными складками на дорсальной части стенки, Е, З – петли киШки с тифлоЗолем, И – боковаЯ часть тифлоЗолЯ. МасШтабные линейки 300 мкм (Е, З, И), 200 мкм (А, В, Г, Д), 40 мкм (Б). мс – меШок кристаллического стебелька, Кс – кристаллический стебелек, Р – реснички, т – тифлоЗоль, lp – lamina propria, г – гемолимфа. Fig. 6. Transverse (А–Е, З, И) and saggital (Ж) sections of the digestive organs: А – crystalline style sac, Б, В – intestinal limbs, Г, Д – part of the stomach with long plicae on the dorsal wall, Е, З – intestinal limbs with typhlosole, И – lateral part of the typhlosole. Scale bars 300 µm (Е, З, И), 200 µm (А, В, Г, Д), 40 µm (B). мс – style sac, Кс – crystalline style, Р – cilia, т – typhlosole, lp – lamina propria, г – hemolymph.

opencc-by-4.0Nov 2024View details →

Finite strain continuum phenomenological model describing the shape-memory effects in multi-phase semi-crystalline networks

This dataset comes from the following paper: Matteo Arricca, Nicoletta Inverardi, Stefano Pandini, Maurizio Toselli, Massimo Messori, Giulia Scalet, Finite strain continuum phenomenological model describing the shape-memory effects in multi-phase semi-crystalline networks, Journal of the Mechanics and Physics of Solids, 105955, 2024. <a title="Persistent link using digital object identifier" href="https://doi.org/10.1016/j.jmps.2024.105955" target="_blank" rel="noopener">https://doi.org/10.1016/j.jmps.2024.105955</a> It contains: <ul> <li>"Notes.pdf" describing all the files uploaded</li> <li>. txt experimental data</li> </ul>

opencc-by-4.0Sep 2021View details →

Data from: Improving the Effectiveness of the Solid-Solution-Strengthening Elements Mo, Re, Ru and W in Single-Crystalline Nickel-Based Superalloys

This dataset is the basis for the journal article "Improving the Effectiveness of the Solid-Solution-Strengthening Elements Mo, Re, Ru and W in Single-Crystalline Nickel-Based Superalloys" (<a href="https://doi.org/10.3390/met11111707">doi.org/10.3390/met11111707</a>).  Differential Scanning Calorimetry (DSC), Electron-Probe Mirco-Analysis (EPMA equipped with WDS detectors), Compression Creep and CalPhaD calculation data are included in this dataset. In the resulting article the partitioning and solid solution strengthening behavior of the elements Mo, Re, Ru, and W are investigated in three different alloy collections: "Reference" alloys:  ERBO/1 (based on the commercial alloy CMSX-4), ERBO/13 (optimized alloy: <a href="http://doi.org/10.1088/0965-0393/23/3/035004">doi.org/10.1088/0965-0393/23/3/035004</a>) and ERBO/15 (optimized alloy: <a href="http://doi.org/10.1002/9781119075646.ch4">doi.org/10.1002/9781119075646.ch4</a>) "Model" alloys: ERBO/17, ERBO/18, ERBO/19 and ERBO/32 - experimental SX Nickel-Based superalloys to investigate the influence of Ti and Ta on the partitioning behavior of W "Experimental" alloys: EXP10, EXP11, EXP12, EXP13, EXP14, EXP15, EXP16, EXP17, EXP18 - general investigation of the behavior of Mo, Re, Ru and W, in terms of partitioning and thermophysical properties. _________   CalPhaD data: short meta data file included in the respective subdirectory DSC data: Meta data as header in each file EPMA data: For each alloy a directory includes a general meta data file "0.cnd", an element specific meta data file "*.cnd" and the actual mapping data of each element "*.txt" (element name is listed in the corresponding specific meta data file). All element compositions are given in wt.-% in the mapping data .txt files. Exceptions: "COMPO" maps represent the total amount of detector counts; The values correspond to counts if the mapping .txt files contain non-float (e.g. integer) values (usually this means that there is very little or none of that element present and should therefore not be quantified to a wt.-% value). Creep: short meta data file included in the respective subdirectory

opencc-by-4.0Mar 2021View details →

Data for the article "Strained crystalline nanomechanical resonators with ultralow dissipation"

GDS fabrication masks, ringdown traces and compilations of fit results, exemplary Raman scans, results of relevant FEM simulations and some of the scripts used for generating the plots in the manuscript.

opencc-by-4.0Jun 2022View details →

Dieletric function crystalline 2H-GaS

Txt files containing the dielectric function of bulk single crystal c-axis oriented 2H-GaS (gallium sulfide) both experimental and DFT calculated values. EXPERIMENTAL GaS_bulk_experimental.txt Experimental value of the dielectric function obtain through spectroscopic ellipsometry under the isotropic assumption. DFT CALCULATED DFT calculations on bulk GaS were performed with SIESTA (https://siesta-project.org/siesta/). The two components of the dielectric tensor of GaS were calculated. <ul> <li>in-plane component (E || c-axis): GaS_inplane_epsilon_img.dat and GaS_inplane_epsilon_real.dat</li> <li>out-of-plane component (E  &perp; c-axis): GaS_outofplane_epsilon_img.dat and GaS_outofplane_epsilon_real.dat</li> </ul>

opencc-by-4.0Jun 2024View details →

Predicting particle quality attributes of organic crystalline materials using Particle Informatics

Dataset related to the publication: "Predicting particle quality attributes of organic crystalline materials using Particle Informatics" published in Powder Technology (<a title="Go to table of contents for this volume/issue" href="https://www.sciencedirect.com/journal/powder-technology/vol/443/suppl/C">Volume 443</a>, 1 July 2024, 119927Volume 443, 1 July 2024, 119927). In this work, a novel quercetin solvate of dimethylformamide (QDMF) was studied. The crystal structure was solved using single crystal X-ray diffraction and analysed using synthon analysis and other particle informatics tools (e.g., solvate analyser). The thermal behaviour and thermodynamic stability of QDMF were studied experimentally using Raman spectroscopy, ATR-FTIR spectroscopy, differential scanning calorimetry, and thermogravimetric analysis. A clear relationship between the two-step desolvation behaviour of QDMF and the type, strength, and directionality of the main bulk synthons characterizing the QDMF structure was observed. Additionally, the attachment energy model was used to predict the QDMF morphology, together with facet-specific topology and chemical nature of each of the dominant {001}, {110}, and {200} facets. The {200} facet was found to be significantly rougher than the other two; whereas, the {110} was characterized by a higher percentage of exposed DMF molecules compared to the other two facets. Specific scanning electron microscopy and contact angle measurements were used to experimentally detect differences among the three facets and validate the modelling results.

opencc-by-4.0Dec 2017View details →

Data of the publication: Recent Advances in Rare Earth Doped Inorganic Crystalline Materials for Quantum Information Processing

Data corresponding to the figures of the publication "Recent Advances in Rare Earth Doped Inorganic Crystalline Materials for Quantum Information Processing" by N. Kunkel and Ph. Goldner (https://doi.org/10.1088/1361-648X/aa529a). A text file describes data in each compressed folder, please refer to the caption in the publication for more details.

opencc-by-4.0Oct 2018View details →

Milling of single-crystalline tips supplementary material

This repository contains supplementary data and simulation results to our paper on "Building a Library of Simulated Atom Probe Data for Different Crystal Structures and Tip Orientations Using TAPSim". The data were generated in the revision stage of the proceedings paper to the APT&M2018 Microscopy and Microanalysis special Issue. The source code for our C/C++ tip synthesis tool is available here on Github: https://github.com/mkuehbach/APTTipCarving https://github.com/mkuehbach/APTTipCarving.git The MATLAB script to visualize detector hit maps are available as: FDM_tapsim.m read_dhits_f.m The supplementary data contains several zip compressed folders, which in detail contain: InputDependencies: TAPSim and Meshgen configuration files, physical specification of the Al tip and evaporation conditions scripts: All shell scripts we used to generate the configurations as much as possible automatized. The scripts should be placed in a folder. In this folder the following should be placed i) meshgen.ini, ii) Al*.cfg, iii) the tapsim, meshgen, and bicarving executables and the BiCarving and MeshGen generated output. Once a simulation queue is initiated, it creates individual folders named BatchMill* one for each tip run specifically. These folder contain the TAPSim results. OrientationsCompareToExpUsingInv through datsets *6604, *6606, *6608, *6610, *6612, and *6614.zip These support Figure 3 and 4 of the manuscript with which the effect of increasing tip radii on detector hit intensity maps was studied. 6604 4nm half-sphere radius, 6606, 6nm, ... OrientationsIdentityXRotated: These simulations probe systematic rotations of an identity oriented tip about the x-axis rotating from +5deg in increments of +5degree to +85degree about the x-axis. We used them to assure the consistency of our orientation indexing scheme in addition to the randomly oriented tips. DetailedVerbosingOfTAPSim: These data detail a simulation run which ran instrumented with the VERBOSE option of TAPSim to identify how much run elapsed time was spent in certain sections of the code. IdentifyWhyTAPSimFailsForSomeOris: This bug reports through instrumented TAPSim source code one of the examples in which TAPSim failed by allocating the entire system memory because of getting stuck in a while loop as is detailed further in the manuscript.