NanoPaleoMagnetism

We can tick off another major milestone today, as the complete NanoPaleoMagnetism ERC team came together for our first group meeting. Our newly arrived PhD students Claire Nichols, Joy Muraszko and Robers Blukis joined the rest of the gang (Nono, Josh and James) to sample the delights of James’s chocolate density bricks (chocolate, mocha, salted caramel and Nutella in one brownie? Oh my, I need to lie down!), get an overview of the project and an update on the P.I.’s plans for world domination. A lot has been achieved, and we are on schedule so far. With the lab completed, the next few years should be very busy indeed.

ERC Selfie – Roberts, Joy, Claire, Nono, James, Josh and me

I couldn’t resist posting this tantalising first image obtained by new group member Josh Einsle – a 3D reconstruction at the nanometre scale of metallic iron particles inside a grain of dusty olivine from the Semarkona meteorite. The image represents a 9.6 µm by 8.0 µm by 5.24 µm volume (red numbers are in nm), that has been sliced through and imaged every 20 nm using a dual-beam FIB. From this reconstruction we aim to extract particle sizes and shapes, and use these as input into micromagnetic simulations. This will allow us to fully characterise the remanence carrying potential of these grains and ultimately lead to a quantitative model of their paleomagnetic properties.

Fe nanoparticles in dusty olivine: carriers of magnetic signals from the early solar system?

Well the usual rush to submit abstracts for the AGU meeting is finally over. The NPM project will be very well represented this year, with four abstracts submitted directly from the group, and another two with an NPM component submitted by our collaborators in MIT and the IRM. Among our contributions will be post doc Nono Lascu, presenting the first results to come out of our new MTJ scanning magnetic microscope, newly arrived post doc Josh Einsle, presenting the first FIB slice-and-view tomography results of magnetic carriers in natural dusty olivine, PhD student James Bryson, presenting the culmination of his thesis work on nanopaleomagnetism of the pallasite and IVA parent bodies, and myself describing a major milestone for the computational side of the ERC project: FORCulator – a tool for simulating the FORC diagram of strongly interacting particles. It promises to be a busy meeting. See you in SF!

Rich

One of the things highlighted at the recent FORC workshop at Santa Fe was a simple method devised by Ramon Egli to reduce the appearance of artefacts in FORC diagrams along the “remanence diagonal” – the line with slope -1 through the origin of the FORC diagram that represents the region where the applied field to the sample is zero (i.e. remanence measurements). In this region it is often the case that FORCs have a sigmoidal appearance that is not well described by the 2nd order polynomial fit used to smooth the FORC surface (e.g. due to the presence of SP particles). This can result in artefacts along the remanence diagonal and large anomalies where the remanence diagonal intersects the lower edge of the FORC diagram, especially when the smoothing boxes are not square and when large smoothing factors are used (as is the case for VARIFORC smoothing). The solution is simple but very effective: subtract the lower branch of the hysteresis curve from each of the FORC curves to obtain the ‘difference FORCs’. This reduces the sigmoidal component of the signal. The resulting FORC distribution is not affected, but the appearance of artefacts is significantly reduced (see below).

On the left we see the standard bush.txt FORC diagram processed using VARIFORC. The intensity has been enhanced to highlight the artefacts along the lower edge and the remanence diagonal. On the right we see the ‘difference FORCs’ obtained by subtracting the lower branch of the hysteresis curve. The processed FORC diagram now has the artefacts completely eliminated!

Go to our download site to get the latest version of FORCinel 2.03 that includes this new correction feature. See the ReadMe file in the FORCinel package for instructions on how to perform the correction. Highly recommended!

Nanopaleomagnetism had a high profile at the IRM’s 10th Santa Fe meeting last week, with P.I. Richard Harrison presenting two invited talks including work from group members Ioan Lascu and James Bryson. Both talks highlighted how the research we are performing within the group is being enhanced by the software tools we are developing as part of the ERC project (FORCinel, ATHLETICS and the upcoming FORCulator). Copies of the two talks are available from the links below, as well as from our publications page.

Magnetic Microscopy: seeing is believing

The philosophy of FORCs: what to do, how to do it and why

The second talk formed part of a dedicated FORC workshop. Links to the other presentations by Ramon Egli, Dave Heslop and Michael Winklhofer will be uploaded to the FORC workshop website.

It’s the end of the academic year in Cambridge, and we would like to congratulate two of our future PhD students, Claire Nichols and Roberts Blukis, on their outstanding exam performances. Both obtained 1st class marks in their finals and will be starting their PhD projects in October 2014. Pats on the back all round.

A quick note to welcome post doc Josh Einsle to the NPM group. Josh will be acting as our microscopy specialist. Josh’s project is jointly supervised by Prof. Paul Midgley, head of the electron microscopy group in the Department of Materials Science and Metallurgy and expert in tomography. Josh will be working extensively with both our groups to pioneer methods of 3D characterisation of natural magnetic materials, as well as continuing our long-term interest in magnetic imaging using electron holography.

Welcome to the team.

Preparations are well underway for our upcoming trip to the biennial Institute for Rock Magnetism meeting in Santa Fe. This meeting is one of the most productive in the rock magnetism calendar – a focussed gathering of academics, post docs and students all working on cutting edge aspects of rock magnetism. This year the meeting has been extended for one day to accommodate a workshop on FORC diagrams, aimed at bringing new users up to speed with the latest developments in FORC processing and analysis. Nanopaleomagnetism will be there giving invited talks on two central aspects of the NPM project:

“Magnetic microscopy: seeing is believing.”

and

“The philosophy of FORCs: what to do, how to do it and why.”

The first talk will highlight the importance of linking magnetic microscopy observations to the underlying physical state of a sample via micromagnetic simulations, and will include a practical demonstration of our image simulation code ATHLETICS that enables a range of magnetic microscopy images to be simulated (including electron holography, Lorentz microscopy, magnetic force microscopy, X-ray photo emission electron microscopy and scanning magnetic microscopy).

The second talk provides an introduction and overview of FORC diagrams, and will include a practical demonstration of some of the latest features of FORCinel including a new method for isolating the central ridge signal associated with non-interacting single domain particles. The talk will also provide a sneak peak of FORCulator – a flexible tool for simulating FORC diagrams.

The talks and latest software versions of FORCinel and ATHLETICS will be placed online prior to the meeting, so watch this space!

Rich

I am pleased to announce a new significant update to FORCinel (version 2.02). This version includes significant improvements to the user interface of FORCinel and a much more streamlined workflow for loading and processing FORC diagrams, as well introducing some major new features such as the new “Extract Central Ridge” feature (see screenshot below). Go to the download page to download the latest version.

Changes in version 2.02:

1. Made a single Load FORCs command that can now automatically recognise whether file is old format or new format. User no longer has to decide which file type is which.

2. Simplified and streamlined the loading process – no user input is required when loading the data. Operations that needed user input before are now only performed when the user chooses the “Extract Day Plot Option”.

3. Created control panel for loading, processing and smoothing FORCs, together with simplified options for changing the colour scale. Full control of the colour scale is now contained in a separate control panel called “Fine Tune Colourscale”. Buttons have been added to switch between the default and rainbow colour scales, and behaviour is now consistent when using the autoscale and reset colour scales. Added “reset colour scale” button to quickly reset colour scale to its default range.

4. Rearranged menu items into more natural groupings. Items relating to displaying FORCs, smoothing FORCs, changing the colour scale and changing the contours are grouped together. Related to this is a change in the keyboard shortcuts for autoscaling and resetting intensity scales and contour levels.

5. New display option to limit the colour resolution of the default colour scale via a pop-up menu on the main control panel has been added. This enhances the usefulness of the default colour scale by creating a subtle contour effect. Previous version used 500 colours to define the colour scale. Now you can restrict this to a lower value which also reduces the amount of unnecessary noise in the diagram.

6. New major feature: Extract Central Ridge. Uses a new background removal algorithm to allow users to extract and isolate the central ridge signal from samples containing obvious single domain signals. The method works by the user first defining a range of Hu values that define the vertical extent of the ridge. The ridge region is blanked out, leaving only background signal. The Loess smoothing algorithm is then used to fit a smooth function to the background signal and automatically extrapolate across the gap. Removing the smoothed and extrapolated background function leaves the central ridge. The control panel allows you to plot a profile of the extracted ridge and background functions, change the size of the ridge window, and choose the smoothing level for the background fit.

7. Added “Remove lower edge artefact” option. When using VARIFORC, there is typically an artefact on the lower edge of the FORC diagram, where the smoothing box intersects the edge of the measured data range. Removing this region not only looks nicer, but prevents the colour scale from being skewed by large signal values in this region. Signal is removed below a line drawn at 45° to the Hc axis, which intersects the Hu axis at a user-defined value. After removal it is recommended to press the “reset intensity scale” button in order to autoscale the intensity range. Simply change the smoothing factor again to restore the original FORC diagram without the edge removed.

8. Added “Attempt Rescue of Bad Data” option. In rare cases when the fields applied during measurements are not monotonic (due to some unforeseen experimental problem), this can cause the FORC diagram to fail to display (just a blank window appears). I get a couple of requests from users a year with this problem. Running this command will manage to correct for the error in the raw data, allowing the FORC diagram to be plotted as normal.

9. Fixed display bug that caused contours to change style when adding the signal-to-noise contour.

10. Enabled ‘Remove First Point Artefact’ for VARIFORC processed data. Note, however, that this option may require you to increase the smoothing factor along the vertical ridge to a value of AT LEAST 4 in order to have enough points to do the fit.

11. Fixed bug that would show an error when choosing autoscale contour range on a graph that had no contours.

12. Fixed bug in FORCintense that caused incorrect calculation of Preisach distribution.

The NanoPaleoMag project will be one year old tomorrow. On the eve of our first birthday, I thought this would be a good time to look back over all that has happened during the past year, and look forward to what is to come next…

The major milestone defined for the first year was the preparation and installation of key laboratory equipment. This task began in earnest last summer with the arrival of Gary Scott, Bruce, Bruce, Bruce and Bruce and the installation of our magnetically shielded room. Later in the year saw the arrival of our Lakeshore PMC vibrating sample magnetometer/alternating gradient magnetometer. Just a few weeks ago we installed our custom built Micro Magnetics MTJ scanning microscope into the shielded room. Finally (although it is still in its box) we saw the arrival of our last major piece of lab infrastructure: an ASC scientific AF demagnetiser/ARM magnetiser. A big thanks to all the technical staff for helping to get things set up smoothly, and to Julie in the office for handling all the invoices!

Another major task was to find people to make use of all the shiny new kit. First to arrive was post doc Nono Lascu, who brought a wealth of experience in rock and environmental magnetism with him from his time at the Institute for Rock Magnetism. Nono’s main task from now on will be to explore the new opportunities for highly targeted paleomagnetic measurements using the MTJ microscope. June will see the arrival of our second post doc Josh Einsle. Josh is a dual beam FIB applications specialist with many years experience working on both sides of the microscopy fence (both as a researcher and as an employee of FEI). Josh’s main task will be to apply advanced electron microscopy techniques, such as slice-and-view tomography, to our natural samples – allowing us to feed the positions, sizes, shapes and orientations of remanence carriers into computer models of how the sample should behave. Next to arrive in October will be our three new PhD students (Claire, Joy and Roberts), who will each be working on complementary aspects of the project.

In addition to all the work preparing the labs, we’ve also had time to fit in some fundamental research, publish several key papers, make many conference contributions, and release a brand new version of FORCinel. The most exciting and significant development has been the success of our first experiments using synchrotron X-rays to image natural remanent magnetisation at the nanoscale in Fe-Ni meteorites. Together with PhD student James Bryson, we have already achieved one of the long-term goals of the project by successfully demonstrating that paleomagnetic information can be extracted from the Fe-Ni meteorites at sub-micron length scales using XPEEM (the very definition of nanopaleomagnetism).

So one year down, four more to go. Exciting times ahead!

Richard Harrison