AtomicJ Users Manual
AtomicJ Users Manual
USER’S MANUAL
April 2021
CONTENT
2
4. ROI modes ………………………………………………….............................................. 51
3
9. ROI Shape Factors dialog …………………………................................................ 91
10. Editing style of ROIs ………………………….…………….......................................... 92
11. Image processing ………………………………......……………................................... 93
4
Written by Paweł Hermanowicz
Malopolska Centre of Biotechnology, Jagiellonian University, Poland
Faculty of Biochemistry, Biophysics and Biotechnology, Jagiellonian University, Poland
[email protected]
AtomicJ is a non-commercial, open source application, licensed under the terms of the GNU
General Public License (GPL). The whole text of the license in distributed with the software.
Please note that AtomicJ is distributed WITHOUT ANY WARRANTY; without even the implied
warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Please refer to the
license for details.
1. Download AtomicJ on your computer. AtomicJ can be downloaded free of charge from
SourceForge (http://sourceforge.net/projects/jrobust/). There are two different
distributions available, for different operation systems:
5
Click on the AtomicJ.exe file (right in the main application directory, icon with yellow
triangle on black background. You may want to create a shortcut on your desktop).
AtomicJ will be opened with the default run parameters – it will use up to 3 GB of the
RAM. AtomicJ.exe may not open if there is not enough memory. In such case follow
steps in B.
If Java 8 or later is not installed on your computer, AtomicJ will be run with the Java
Runtime Environment included in the distribution. Java can be downloaded free of
charge from Oracle website:
http://www.oracle.com/technetwork/java/javase/downloads.
Click on the AtomicJ.jar file (icon with a steaming cup of coffee). AtomicJ will open with
the default run parameters, which depend on the version of Java installed on your
computer. You may want to override those default settings, especially the maximum
memory available for application. To do so, run AtomicJ.jar from the command shell (for
Windows, see appendix How to Open Windows Command Prompt).
On the shell, set the current working directory to the AtomicJ directory (in Windows,
just go to the directory using the cd command) and run the command:
The value memoryMaximum is the maximal amount of memory (in megabytes) that
AtomicJ can use. For example, if you work on a computer with 8GB RAM and you want
to process at once several large force maps, you may use 7GB of memory, using the
following command:
The recordings used in this manual are included in the test files, available at
https://sourceforge.net/projects/jrobust/files/TestFiles/AFMRecordings.zip. You will find there
force curves (collected on prostate cancer cells), as well as a force map and a corresponding
AFM image of fibroblasts. Simulated curves for different tip shapes can be downloaded from
https://sourceforge.net/projects/jrobust/files/TestFiles/SimulatedCurves.rar. The curves are
accompanied by .properties files, which contain information about the recording conditions and
processing settings. Those files can be read-in to AtomicJ before processing (see section E.1.
Working with the processing assistant).
6
C. Structure of the Main window
The graphical Interface of AtomicJ consists of the main window, a number of dialogs for
displaying results, and assistants which help users to accomplish complex tasks.
The main window (Fig. 1) appears immediately after the program is started and remains
accessible throughout the session. It is equipped with a menu bar and a toolbar. All options can
be accessed through the menus from the menu bar, while the most commonly used are also
included in the toolbar.
Toolbar options :
Open (Ctrl + O) (Fig. 1 b) lets you open force curves (without calculating
mechanical properties) or images.
Force curves (Fig. 1 c) shows the Force curves dialog, where the charts with
force curves opened during the session are stored. This action is initially
disabled, as there are no charts to show.
Graphical results (Fig. 1 d) shows the Charts dialog, which stores charts
generated during processing of force curves and force maps. Initially
disabled.
Images (Fig. 1 f) shows the Images dialog. Click to view or process the AFM
images that were opened during the current session. Initially disabled.
Maps (Fig. 1 g) shows the Maps dialog. Click to view and analyze the force
maps generated during the current session. Initially disabled.
7
Fig. 1. Main window. The toolbar icons (a – h) correspond to the commonly used options: a –
process force curves and maps, b – open curves or AFM images,
Menu options
The main menu bar contains three menus and can be navigated with either the mouse, or the
keyboard, using the mnemonics in the brackets:
8
Fig. 2. Preferences dialog lets you specify the maximum
number of processors or cores used by AtomicJ for parallel
processing. AtomicJ is a concurrent application, it can break
larger task into multiple subtasks that are run concurrently
on separate processors. It uses concurrency to process force
curves, read in and save files, generate stack frames,
calculate stack cross-sections and to create charts. All these
tasks are faster if more processors are used.
AtomicJ can open and process different types of AFM recordings, in particular:
• Single force curves - the recordings of force acting between the sample and the
cantilever tip when it is pushed into the sample and then withdrawn.
• Force volume recordings (or force maps)- multiple force curves recorded in a grid. The
grid is specified over the sample and force curves are recorded in each node of the grid.
• Flexible force curves recordings - multiple force curves recorded on the sample at
arbitrary points.
• Images obtained by scanning the sample. The images usually consist of different
channels, for example sample topography, friction or error signal.
All these types of AFM recordings can be viewed with the Open files assistant (Fig. 3). It can be
accessed from the Main Window with the shortcut Ctrl + 0, or through the toolbar icon
9
or the main menu (go to File (Alt + F) → Open). It lets you read in and open multiple AFM
recordings, possibly of different types. The files that you read in to the Open files assistant are
displayed in the list in its center (Fig. 3).
You can read in or delete files to open with the buttons (Fig. 3 a) and the context menu (b):
• to read in a new file, click the Add button (Alt + A). A standard file dialog will be
displayed. It is important to select the right type of AFM files (Fig. 4 a). You do not need
to add all files one by one. If you select a directory, all files of the right type in this
directory and its subdirectories will be automatically read in and added to the list.
You can open all read-in files or just the ones that you selected in the list.
• to open all AFM recordings in the list, click the Open button (Alt + P).
• to open some of the AFM recordings, select the corresponding file names in the list then
move the mouse over the selected files and click the right mouse button. The context
menu (Fig. 3 b) will appear, click Open.
10
• to add new files to the selection, hold down the Ctrl key and click the new files with the
left mouse button.
• to select all files, use the shortcut Ctrl + A.
The force curves are plotted in the Force curve dialog (subsection D.3. Force curve dialog). All
individual curves from force – volume recordings or flexible force curve recordings are plotted in
the Force curve dialog, too. If you wish to generate maps of mechanical properties from these
recordings, you need to process them – see the section E Force curve processing. AFM images
are plotted in the Image dialog – for details see the section J. Working with images.
The files can be also opened in the default application, associated with their extension by your
Operating System. This may be useful when you want to view the files in your AFM application
before processing in AtomicJ. To open files in the default application, select them in the list,
then click the selected files with the right mouse button. The context menu (Fig. 3 b) will
appear, click Open in default application.
The Open files assistant allows you to display the raw data that make up the AFM recordings.
This may be useful, for example, when the original recordings are saved in the binary form. To
display the raw data, select the files of interest in the list, then move the mouse over the
selected files and click the right mouse button. The context menu (Fig. 3 b) will appear. Click
Raw data. The raw data will be displayed in a separate dialog (Fig. 4).
The Raw data dialogs (Fig. 5.) displays numerical data, for example, the coordinates of points of
the read-in force curves, a matrix of image values or a sample displayed in a histogram. The
dialogs contain a central tabbed pane with tables. Each distinct collection of numerical data (ex.
a force curve in Fig. 5) is displayed in a separate table, under its own tab.
11
The Raw data dialogs are equipped with a main menu bar (Fig. 5 a) and a set of buttons (b) that
allow you to print, save to a file or customize the data in the tables:
• to save the currently displayed table to a file, press Ctrl + S or go to File (Alt + F) → Save
(Alt + S) or click the Save button (Fig. 5 b). A file dialog will be displayed, which allows you
to choose where the table should be saved and in which format (TSV or CSV).
• to print the currently displayed table, press Ctrl + P, or go to File (Alt + F) → Print (Alt + P),
or click the Print button (Fig. 5 b).
• to change the number format, go to Customize (Alt + U) → Format data (Alt + F). The
Customize number format dialog will be displayed (Fig. 22). It lets you specify the number
of decimal fraction digits and the decimal separator used in the statistics tables.
Anytime you open a force curve file, a chart is generated in the Force curves dialog. This is a
prototypical dialog for displaying charts, as more complicated dialogs in AtomicJ inherit its basic
structure. We will discuss it in greater detail and skip the redundant information in discussions
of more complex dialogs.
Force curves dialog consists of a main menu bar (Fig. 6 a), a set of buttons (b) for the most
commonly used actions, a central split pane, with the list of force curves on the left (c) and a
panel displaying the corresponding chart on the right (d).
12
o Open all in dialog (Ctrl + Shift + O) opens all the charts for the current sample in new
dialogs
o Clear clears all the charts
o Close closes the dialog
Clear, Save all and Close options can be also accessed via buttons in panel b.
• Preprocess menu provides many preprocessing procedures, that can be applied to all
force curves stored in the Force curve dialog, or just to the selected ones. The menu also
includes options for undoing and redoing the previous preprocessing operations.
13
o Translate lets you shift the curves in the x- or y-
direction.
o Weighted median lets your smooth the curves using the median filter with different
weights assigned to different positions in the filter window. If a weight equals n, then
the corresponding point is repeated n times in the set used for median calculation.
The regular median filter is a special case of weighted median filter, with all weights
set to 1.
o Gaussian lets you smooth the force curves using Gaussian filter, with user-specified
standard deviation.
o Savitzky – Golay lets you smooth or differentiate the force curves, using Savitzky -
Golay filter. The filter is usually applied to regularly spaced points. If the spaces
varies, AtomicJ will use linear interpolation to produce regularly spaced points before
applying Savitzky-Golay filter.
o Local regression lets you smooth or differentiate the force curves, using robust local
regression procedure with tricube weights. More details can be found in the original
paper: Cleveland, W. S. (1979). Robust locally weighted regression and smoothing
scatterplots. Journal of the American statistical association, 74(368), 829-836.
o Convolve lets you convolve the force curves with a kernel of arbitrary size and
elements. Convolution can be used for example to smooth or to differentiate the
curves. If the curve points are not regularly spaced, AtomicJ will use linear
interpolation to produce regularly spaced points before applying convolution.
14
The list of force curves is equipped with a context menu (Fig. 6 e), triggered by right mouse click.
The context menu contains eight actions:
• Process allows you to calculate mechanical quantities from the currently selected force
curves.
• Raw data (the same as in the main menu)
• Save selected opens the Batch save dialog, which will help you to save the charts for the
currently selected force curves.
• Short names toggles between short names (usually filenames) and long names (usually
full paths) of force curves
• Change font opens a font selection dialog. Click to change the font style or color of the
items in the list
• Open all in dialogs (Ctrl + Shift + O) (the same as in the main menu)
• Open in dialogs (Ctrl + O) (the same as in the main menu)
• Delete deletes the selected force curves. Del key acts the same.
15
E. Force curve processing
Processing assistant is the most important assistant (or wizard) in the application, as it is the
entry point for processing of all types of indentation recordings. This is where the force curves
or maps are chosen, the parameters of analysis are specified and where the actual computation
is triggered. The assistant ensures that the necessary parameters are specified in the right order
and within the correct range. It consists of two pages, the first one is for selection of the data
for analysis, the second one is for specification of how the analysis should be performed. The
assistant allows you to create an arbitrary number of batches, each with different processing
settings.
To display the Processing assistant, click on the cog-wheel icon in the main window
(Fig. 1 a). The assistant will pop up, displaying its first page, which is initially empty (Fig. 7).
16
quality. The Patchwork button lets you choose curves from a force map that should
be included in the current batch based on an AFM image. It also lets you partition a
force map into multiple batches. For more details on partitioning curves from a single
force map into multiple batches, see section E.2. Partitioning force maps into multiple
batches.
d) The navigation buttons (<<Back, Next>> and Finish) (Fig. 8 d) for navigating between
different pages of the assistant.
e) The list of curves already added to the batch, equipped with a context menu,
triggered by the right mouse button (Fig. 8 e). The context menu includes options for
displaying the raw data for selected curves (see the section D.2. Raw data dialog),
previewing, opening and deleting selected curves from the batch.
The parameters of the analysis are specified in the second page of the processing assistant. The
settings specified in this page can be saved to a file, with the Export button. The saved settings
can be then read in with Import button, for example to speed up input processing settings for
subsequent batches.
The second page of the Processing assistant consists of three tabs – General, Advanced
and Output. The General tab (Fig. 9) allows for input of the mandatory settings:
• Batch name (subpanel a in Fig 9). The default name is the current batch index.
• The algorithm used for contact point estimation – whether the search for contact point
should be manual or automatic, and if automatic, which estimator should be used
(Contact estimator, panel b).
• How the selected algorithm should work:
17
a) should it take into account the contact model, working on the force –
indentation curve? (option Based on the contact model in the combo-box
Estimation method, panel b)
b) or should it search for the contact point on the force-displacement curve,
ignoring the contact model (option Model independent)?. In this case, you need
to specify the degree of the polynomial fitted to the in-contact part of force-
displacement curve (field In-contact degree, panel e). This option is mainly for
curves collected on a rigid substrate.
• The regression method used for fitting the contact mechanics model to force-
indentation data. Classical L2 is the least squares regression, classical L1 is the least
absolute deviation regression (panel b)
• The shape and size of the AFM tip used to collect the force curves. Contact models for
different shapes are implemented :
a) Cone
b) Sphere (Hertz's approximation)
c) Sphere (Sneddon's solution)
d) Fung’s (hyperelastic material, sphere)
e) Ogden’s (hyperelastic material, sphere)
f) DMT (model of adhesive contact, sphere)
g) JKR (model of adhesive contact, sphere)
h) Maugis (model of adhesive contact, sphere)
i) Pyramid
j) Power-shaped (shape obtained by revolving a power function along the y-axis)
k) Hyperboloid
l) Hyperboloid (Sun’s model of adhesive contact)
m) Blunt cone
n) Blunt pyramid
o) Truncated (i.e. flat - ended) cone
p) Truncated pyramid (Fig. 9, panel c)
Depending on the shape, you will have to specify tip radius (radius of curvature at the
apex), half angle and transition radius (see tip sections in the Appendix M.2.). For blunt
tips, the value of transition radius that ensures smooth merge between spherical cap
and the tip body can be calculated automatically – click the button.
• Poisson’s ratio of the sample (panel d).
• The method of calculating energy of adhesion (panel d). Energy of adhesion can be
calculated from the adhesion force directly measured on the chart (option Calculate
18
from fit) or treated as a fitting parameter (option From fit) This action is enabled only if
the selected model of indentation takes into account adhesive interactions.
• The degree of the polynomial fitted to the off-contact region of a force curve (panel e)
• The spring constant of the cantilever and the optical lever sensitivity (panel f).
Sensitivity has to be specified only if the deflection values are provided as raw
photodiode output. Otherwise this field will be greyed out. Sensitivity is used to
convert voltage into deflection expressed in microns. It can be calculated based on a
force curve recorded on a clean, rigid substrate (i.e. mica), with the help of the
Calibration dialog (Fig. 14).
The Advanced tab (Fig. 10) allows for input of the optional settings:
Fig. 10. Second page of the
Processing assistant. The Advanced
tab is visible. It allows you to specify
preprocessing - how the curves
should trimmed (a) and smoothed
(b), and to specify the maximum load
or depth of indentation (c) that can
be used to calculate mechanical
properties. For thin samples, you can
choose to correct the bottom effect
(d), i.e. effect of a rigid substrate on
the apparent stiffness of a thin
sample. Bottom effect corrections
are available for spherical and
conical tips.
19
• The extent of cropping – how much of the extreme parts of the raw force curves should
be cropped (panel a in Fig. 10). You can either:
a) fill in the fields in the panel a;
b) or use the Cropping dialog (Fig. 11), which lets you specify cropping on the chart,
by dragging markers with the mouse. The Cropping dialog is displayed with the
Select button.
• The maximal indentation or maximal load (panel c in Fig. 10). If these parameters are
set, parts of the force - indentation curve above the specified maximal values are
rejected.
• Smoothing filter (panel b in Fig. 10). Two low-pass filters are available:
a) Local regression.
20
Half width determines the number of points to the left and to the right of the
current point that are included in the moving filter window. Larger half width
ensures better removal of noise, but smaller features may be lost or distorted.
Degree determines the degree of the polynomial fitted to the points within the
moving window. Higher degrees ensure that higher moments of data are preserved.
• Bottom effect correction (panel d in Fig. 10). Bottom effect is the effect of the rigid
substrate on the apparent stiffness of the sample. Thin samples on rigid substrates
appear stiffer than they really are. You can specify the thickness of the sample either as
an absolute value, or choose a topography image (discussed below). The correction
depends on whether the sample is bonded to the substrate or not (checkbox
Adherent). The corrections are available for two tip shapes: cone and sphere.
Using topography for bottom effect correction: If you recorded the topography image
of your sample, AtomicJ can use it to find the thickness of your sample in the positions
where the curves where recorded. You should:
If a force curve file does not contain information about the recording position, the
application will assume that the sample was of infinite thickness, unless you have
specified explicitly the thickness in the Thickness field (Fig. 10 panel d).
By default, three charts are generated for each curve. For larger sets of curves, this may result in
large consumption of memory. 4 GB of RAM is enough to draw around 30 000 - 60 000 charts,
depending on the size of curves.
Proper specification of the processing settings is crucial for obtaining high quality results.
Processing settings influence the estimates of material properties both directly and indirectly,
through the contact point estimates. Automatic contact point estimators implemented in
AtomicJ can handle large sets of force curves and are robust against various artifacts in the force
curves. Manual estimation of contact point can be used for small sets of curves. In case of large
sets of poor force curves, it may be advisable to compare the results obtained with the
automatic and manual estimation, for a small subset of curves.
22
The quality of the automatic contact point estimates depends on how well the assumed force –
indentation model describes the true relationship. There are several sources of discrepancies
between the model and the reality in AFM indentation experiments. Analyzing force curves, we
usually assume that the sample is linear elastic, i.e. that the relationship between stress and
strain is linear. This assumption is only satisfied if the indentations are small relative to the
sample thickness and the probe is not sharp. If you suspect that at least some curves may
exceed the limit of linear elasticity, you may use robust estimators to fit model to data (LTS, LTA,
HLTS or HLTA). Robust estimators will reject the parts of force – indentation data that grossly
deviate from the model. Apart from the material properties of the sample, force – indentation
relation depends on the tip shape and the thickness of the sample (for equations, see Appendix
M.2. Force - indentation relations.) If the sample is thin, the cantilever feels underlying, rigid
substrate, which leads to overestimation of Young’s modulus. The corrections for this effect
have been proposed for indentation with spherical (Dimitriadis et. al 2002) and conical tips
(Gavara and Chadwick 2012). These corrections are implemented in AtomicJ. Corrections based
on the Lebedev-Ufliand (1958) exact solution to the problem of a thin sample indented with a
spherical or conical tip have been also implemented. The exact, numerical solution was
approximated using piecewise polynomials. The error due to approximation is negligible. They
were derived specifically for use in AtomicJ and are available as the Lebedev-Chebyshev
correction for sample thickness in the Advanced tab of the Processing assistant..
The degree of a polynomial fitted to the off-contact may influence estimation of the contact
point. A straight line (i.e. degree 1) can be fitted if the off-contact parts of the force curves in
the batch are linear. However, the off-contact parts of force curves often exhibit some
curvature, so fitting a polynomial of degree higher degree usually works better. For single curves
polynomial of degree 2 is often optimal. Curves in force maps may exhibit more artifacts in the
off-contact part. In such cases you may consider fitting of a polynomial of degree 3.
The Poisson’s ratio is a property of the sample – it tells how big lateral contraction (or
expansion) will accompany axial stretching (or compression). In principle, in order to obtain
accurate results of analysis, the correct value of Poison’s ratio for our sample has to be
measured independently. However, biological samples are often assumed to be incompressible,
i.e. that their Poisson’s ratio is 0.5 (for example Radmacher 1996, Mathur 2001). Although
overestimation of Poisson’s ratio leads to underestimation of Young’s modulus. This error is
relatively small if the sample is thick. In the extreme case, when you mistake a thick
compressible sample (Poisson’s ratio 0) for an incompressible (0.5) one, Young's modulus is
reduced by 25 %. However, the effect of Poisson’s ratio on the apparent stiffness of the sample
becomes large when the sample is thin and bonded to a rigid substrate, as is the case with
adherent cells.
23
The spring constant of the cantilever has to be measured in order to convert cantilever
deflections into forces. Various methods of measuring cantilever’s spring constant have been
described (Burnham et al 2003). Automatic measurement of spring constant usually can be
performed with programs operating AFMs. Overestimation of cantilever spring constant leads
to overestimation of Young’s modulus.
Burnham NA, Chen X, Hodges CS, Matei GA, Thoreson EJ, Roberts CJ, Davies MC, Tendler SJB (2003) Comparison of
calibration methods for atomic- force microscopy cantilevers. Nanotechnology 14: 1 - 6.
Dimitriadis EK, Horkay F, Maresca J, Kachar B, Chadwick RS (2002) Determination of elastic moduli of thin layers of
soft material using the atomic force microscope. Biophys J 82:2798–2810.
Gavara N and Chadwick RS (2012) Determination of the elastic moduli of thin samples and adherent cells using
conical AFM tips. Nat Nanotechnol. 7: 733–736.
Radmacher M., Fritz M., Kacher CM., Cleveland JP, Hansma PK (1996) Measuring the Viscoelastic Properties of
Human Platelets with Atomic Force Microscope. Biophys J 70 : 556 – 567.
Mathur AB, Collinsworth AM, Reichert WM, Kraus WE, Truskey GA (2001) Endothelial, cardiac muscle and skeletal
muscle exhibit different viscous and elastic properties as determined by atomic force microscopy. J Biomech 34
(12): 1545 – 1553.
Force maps usually contain regions with very different mechanical properties. Often they
contain both curves collected on the sample and on the rigid substrate to which the sample is
attached. Also the sample itself may be heterogeneous. Animal cells often exhibit huge
heterogeneity, for example the properties of cell body are different from the properties of thin
lamellipodia. There are many reasons for this heterogeneity, among them differences in
cytoskeleton, in the strength of adherence to substrate and differences in thickness.
AtomicJ offers multiple ways to deal with this variation. You may try several approaches:
1. setting the upper limit on the depth of indentation (Fig. 10 c). Any larger indentations
will be discarded from the force – indentation data.
2. setting the upper limit of the applied force (Fig. 10 c). Any larger forces will be discarded
form force - indentation data.
3. using robust estimators (HLTS, HLTA, LTA or LTS) (Fig. 9 b) for fitting the model to force –
indentation data. Robust estimators will automatically discard the parts of force-
indentation data that significantly deviate from the assumed model, usually the large
indentation parts.
4. using the modified models for spherical or conical tips which take into account finite
sample thickness, especially if you have a topography image which can be used by
AtomicJ to find sample’s thickness in different locations (Fig. 10 d, Fig. 12).
24
You may use all this approaches together or any combination of them. You can also use
different approaches for different parts of a force curve. To do so, you need to partition a force
map into several batches of curves, using the Partition force map assistant (Fig. 14, Fig. 15, Fig.
16). This assistant can be opened from the Processing assistant (for details on using Processing
assistant, see section E.1. Working with the Processing assistant):
• to open Partition force map assistant, add a force map to the Processing assistant, then
click the Patchwork button (Fig. 8 c)
25
Using the buttons in Fig. 16 a you can
switch between different ROI shapes:
1. polygonal
2. rectangular
3. elliptical
4. arbitrary shape
Once you selected all patches, click the Finish button (At + F). If you have not covered the image
completely with patches, AtomicJ will ask you whether the curves not included in any patch
should be discarded, or whether a new batch should be created for them (Fig. 17). This means
that when you want to partition a force map just into two patches, you really need to draw just
one of them, and the curves outside it can be automatically added to another batch. You may
also discard these curves, for example when you want to remove from the map the curves
collected on a rigid substrate.
26
Fig. 17. When some curves are not included in any of the selected patches, AtomicJ will ask you whether a new
batch should be created for these curves.
The third page of the Partition force map assistant also allows you change the style of ROIs for
the current patch, as well as to calculate the statistics, shape factors (i.e. area) and to draw
histograms for the selected ROIs. These actions can be accessed from the context menu. To
display the context menu, move the mouse over the image chart and click the right mouse
button.
E.4. Sensitivity calibration
Usually, cantilever deflection is measured with the optical lever technique. Force curves are
recorded as a relationship between voltage generated in the photodiode and the position of the
cantilever. Before processing, voltage should be multiplied by the optical lever sensitivity in
order to convert it into cantilever deflection. Optical lever sensitivity tells how much voltage
corresponds to a unit of deflection. It can be determined based on calibration curves, i.e. curves
recorded on a hard substrate (ex. mica) with the help of the Calibration dialog (Fig. 18).
There is no upper limit on the number of calibration curves you can use. If multiple curves are
used, the Calibration dialog will calculate average sensitivity. The Add button allows you to
choose new calibration curves from your computer file system. You have to choose at least one.
The newly chosen curves will appear in the list (Fig. 18 a). For each calibration curve, select the
range for calculation of the optical lever sensitivity (b). The range should encompass the linear,
in-contact part of the curve. The limits of this range can be dragged with mouse. The sensitivity
is the inverse slope of approach part of the curve.
Fig. 18. Dialog for calculation of the
optical lever sensitivity. Multiple
force curves can be used for
calibration (a), average sensitivity is
then calculated (c). The calibration
is done based on the part of the
approach curve within the blue-
shaded area (b), whose bounds can
be modified with the mouse. The
current sensitivity, the average
sensitivity and the number of
curves used for calibration so far
are displayed in panel c. The
average sensitivity will be used
for force curves processing.
27
The navigation buttons (Fig. 18 d) allow you to commit you selections or to move to other
curves. Click Apply to commit sensitivity calculated from the current curve and range, Skip to
move on to the next curve, Finish to finish calibration (current average sensitivity will be used
as the final result) and Cancel to cancel calibration – no result is returned. The Clear button (Fig.
18 a) discards all calculated values (you may choose new set of calibration curves and repeat
calibration).
TIPS!
• You can choose all calibration curves from a single directory at once – just select the
directory and AtomicJ will read in all force curves in this directory.
• You can choose a force map and use the curves recorded on hard substrate as
calibration curves – click Substrate button to select the rigid substrate on an image.
• You can skip a low quality calibration curve – click Skip without clicking Apply.
• You can zoom-in the in-contact part of the calibration curve.
28
E.5. Manual selection of the contact point
Contact point can be selected manually or semi-manually with the Contact point selection
dialog (Fig. 19).
Fig. 19. Dialog for manual determination of the contact point position.
• select the contact point with a left mouse click on the chart a and adjust its position
dragging it with the mouse. The current coordinates of the contact point are displayed
in the fields d.
• enter the coordinates in the fields d. The contact point appears on the chart, where it
can be repositioned.
You can choose whether to use both or just a single coordinate of the selected contact point,
using the radio buttons e. In the latter case, the missing coordinate is found automatically. For
example, if you specify the Y coordinate to be 100 nN, the application will find the point on the
curve closest to 100 nN. Please remember that the force values are calculated from raw
photodiode readings, so they are shifted by a constant corresponding to the voltage signal
generated by the laser bounced off the unengaged cantilever.
You can preview what will be the results of processing if you apply the current contact selection.
The Live preview dialog (Fig. 20, you can display it with the button c) shows Young’s modulus
and transition indentation corresponding to the current contact point position, together with
the force-indentation and pointwise – modulus curves. It gets instantly updated when you move
29
the contact point (ex. dragging it with mouse). This dialog can be used to demonstrate the
importance of proper selection of the contact point.
Once you have chosen the right point, you can apply it with the Apply button (Fig. 19 panel f).
The Apply to all button applies the current coordinates to all remaining curves. This is useful for
fast, cursory analysis, for example, of curves collected in a single location. The Skip button
allows you to skip the low quality curves that should not be included in the analysis.
Fig. 20.Live preview of the force – indentation and pointwise modulus – indentation curves, generated
based on the contact point selected manually.
30
Working with the results of curve processing
Numerical results of processing of force curves are displayed in the Results dialog. It is equipped
with a menu bar (Fig. 21 a), a toolbar (b) and a set of buttons (c). All options can be accessed
through the menus from the menu bar, while the most often used ones are also accessible from
the toolbar or via buttons.
Toolbar options:
Show histograms (Alt + W) opens the Histograms dialog, which stores all
histograms generated from the processing results in the current session. It is
initially disabled, as there are no histograms to show.
Pooled histogram (Alt + O) lets you generate a histogram for all the batches
combined.
31
Batch histograms (Alt + B) lets you generate a separate histogram for each
batch.
Descriptive statistics (Alt + D) opens the Live results statistics dialog. Click to
view mean, median, variance and other statistics for all batches and pooled
results.
Show box plots opens the Box plots dialog, which stores all box plots generated
from processing results in the current session. It is initially disabled, as there are
no box plots to show.
Add box plots (Alt + X) opens the Box plots assistant. Click to generate box plots
for selected subsamples of processing results.
Mark on the maps marks the position corresponding to the selected entries on
maps of the mechanical quantities
Menu actions:
The menu bar consists of four menus and can be navigated with the mouse or using keyboard
mnemonics in brackets:
32
o Show histograms (Alt + W) (also in the toolbar)
o Pooled histogram (Alt + O) (also in the toolbar)
o Batch histograms (Alt + B) (also in the toolbar)
Clear, Save, Print and Close actions can be also accessed through the respective buttons in the
button panel (Fig. 21 c).
The results are displayed in a table (Fig. 21 d). Each row corresponds to one force curve. The
table contains the calculated values of Young’ modulus, coordinates of the contact point,
transition indentation, transition force, adhesion force, the most important processing settings
– spring constant, optical lever sensitivity, tip shape – and the batch name. The format of the
data in the table can be customized via the customize menu (see above).
Individual or all rows of the result table can be selected. You can select multiple rows:
• to select non-contiguous rows, right-click with the mouse while holding down the Ctrl
key
• to select a range of contiguous rows, click the first row and then the last row, while
holding down the Shift key
33
• to select all rows, press Ctrl + A.
The results dialog is equipped with a context menu (Fig. 21 e). It contains options which are
performed on the currently selected rows:
• Copy (Ctrl + C) copied the rows to the clipboard. The rows can be pasted to other
applications, including MS Excel.
• Select all (Ctrl + A) selects all rows in the table
• Jump to graphs shows the corresponding charts in the Graphical results dialog, if they
are available
• Recalculate opens the Processing assistant. Click to recalculate the curves corresponding
to the selected rows. New results are added to the table, the old ones stays.
• Recalculate and replace acts as Recalculate, but deletes the old results
• Selection histogram opens the Histogram assistant. Click to generate a histogram for the
selected rows.
• Selection statistics opens a dialog displaying statistics for the selected rows.
• Delete deletes the selected rows.
Format of the results can be modified with the Customize results table dialog (Fig. 22). It allows
you to change the number format, in particular:
34
Each change to the format of results is saved as the application’s default settings (in the registry
in the case of Windows), so you do not have to customize it again the next time you run
AtomicJ. The format is also preserved when results are saved to a text file.
You can produce histograms of the results with the help of the Histogram assistant, which
gathers all the necessary settings. It consists of a variable number of pages that depends on the
number of samples to be plotted. If multiple samples are available, the first page (Fig. 23 left)
allows you to select which of them should be plotted. The next pages are the binning pages (Fig.
23 right), one page for each sample. If there is just one sample (ex. when you plot pooled results
for all batches), the assistant consist of a single binning page (Fig. 23 right).
In the binning page each histogram should be named (panel a). You can choose the binning
manually (filling either the bin width or bin count field), or use an automatic binning method
(panel b), one of the following:
• Sturges k = log 2 n + 1
3.5
• Scott’s h = 3
n
• Square root k = n
2 IQR
• the Freedman-Diaconis h = 3
n
where n – the sample size, k – bin Count, h – bin width, σ – sample standard deviation, IQR –
interquartile range, x is the smallest integer not less than x.
If the extreme values in your sample are dubious (ex. negative values of Young’s modulus), you
may want to discard a fixed percentage of largest/smallest values with the help of panel c.
Alternatively, you can change the range of data that are displayed in the histogram (panel e).
AtomicJ supports both ordinary and cumulative histograms. Bin height can be specified as the
count (number of values within the bin), its decimal logarithm, probability (i.e. the frequency of
values within the bin) or probability density (frequency divided by the width of the bin). A
theoretical distribution can be fitted to the sample and plotted together with the histogram.
AtomicJ supports the following distributions:
• Normal
• Log-normal
• Laplace
• Lorentz.
35
The log-normal distribution is not available when the sample contains negative values
Fig. 23.The first (left) and the second (right) page of the histogram assistant. The first page allows you to
specify the quantities for which you want to draw histograms, while the second page allows you to
specify the binning parameters
Histograms are stored in the Histogram dialog (Fig. 24). Its structure is similar to the Force curve
dialog. It contains a main menu bar (Fig. 24 a), a set of buttons (b) for the most commonly used
actions, a central split pane, with a list of samples on the left (c) and a tabbed panel displaying
the corresponding charts on the right (d). Charts for different quantities are located under their
own tabs. Basic information about fit, sample and binning are displayed in the panel e. This
information gets updated when you navigate between different histograms (clicking on
different tabs or selecting different the sample).
The main menu bar contains four menus. The File, Chart and Measure menus contain the same
options that we discussed in the subsection D.3. Force curve dialog. The fourth menu, Data (Alt
+ D) is specific for the Histogram dialog. It contains six actions:
36
• Statistics (Ctrl + T) opens a statistics dialog. Click this to view mean, median, variance
and other statistics for all batches and pooled results.
• Modify binning (Ctrl + B) opens the Binning dialog, which allows you to modify the
currently displayed histogram. It contains the same fields as the second page of the
Histogram assistant (Fig. 23 right). Histogram is instantly updated when you change the
settings in the Binning dialog. Thanks to instant updates, you can easily determine what
is the effect of the changes. For example, you can see how changes in the number of
discarded results on the quality of fit. The changes can be reverted with the Undo
button.
• Fix zero (Ctrl + F) subtracts the smallest element of the sample from all the items in the
sample. After this operation the smallest element equals 0.
• Add constant (Ctrl + D) allows you to add a constant to all items in the sample. You may
use it when you want to fit lognormal distribution to the sample that contains negative
values.
• Transformations allows you to transform the sample, by square root, logarithmic (with
base 10) or reciprocal transformation. Note that square root and logarithmic
transformations require that all the items in the sample are positive. If necessary, get rid
of non-positive items with the Add constant action.
Fig. 24. Result histogram dialog. It stores all histograms that you have generated from the results of force
curve processing. The list of the samples (here named group of curves) (c) allows you to navigate
between histograms (d). For each sample several histograms can be generated for different quantities –
for example for Young’s modulus and for transition indentation.
37
F.5. Creating box plots
Box plots (also known as box-and whisker plots) are useful for comparing distributions of
multiple samples. They carry less information then histograms, but are much more compact , so
they can be recommended when the number of samples is considerable. Box plots depict
several statistical properties of the sample (Fig. 25):
• median
• upper quartile
• lower quartile
• arithmetic mean
• upper fence, i.e. the greatest value less than median + 1.5 times interquartile range
• lower fence, i.e. the smallest value greater than median – 1.5 times interquartile range
• outliers, i.e. data greater than upper fence or less than lower fence
To create box plots from results of force curve processing, you click the Add box plot
toolbar item in the Results dialog (see section F.1). The Box plot assistant will popup. Its first
page contains checkbox with names of all quantities that were calculated from the force curves.
Select the quantities that you want to plot and click the Next button to move to the next page of
the assistant.
38
The Box plot assistant now
displays its second page, which
allows you to select the force curves
that will comprise the first sample.
Click the Add button to choose the
force curves from the processing
results. The number of samples
visualized in a box plot is not limited
– to add new samples, click the Next
button. When you are done, click
the Finish button in the Box plot
assistant. Box plots will be then
created.
You can test hypotheses about means of samples with the inference assistant. It consists of two
pages. The first page (Fig. 26 left) is for selection of the statistical test. Currently, t-tests for one
and two samples are available. The second page (Fig. 26 right) is for specifying the samples and
test parameters (for two-sample t-test, you have to specify the significance level, choose
whether the test should be one – or two tailed and whether sample variances are to be
regarded as equal. Samples are independent of batches – single sample may contain results
from multiple batches.
Click the Edit button to add new results to the
sample. The Sample editor will appear (Fig. 25). It
allows you to add new results to the sample and
choose which quantity you are interested in.
Once you have chosen the samples, click the Finish
button to run the test. A new dialog with test results
(Fig. 27) will pop up.
Fig. 25. Sample editor, allowing you to choose which
39
curves should be included in the sample and which
quantity should be analyzed statistically.
The Test results dialog is equipped with the main menu and a set of buttons. The following
actions can be performed:
• Save (Ctrl + S) saves the test results to a text file (TSV or CSV)
• Print (Ctrl + P) prints the test results
• Close dismisses the dialog.
Fig.26. First (on the left) and second (right) page of the inference assistant.
40
F.7. Descriptive statistics
Descriptive statistics for processing results are displayed in the Live statistics dialog (Fig. 28). It
contains a separate table for each quantity under its own tab (Fig. 28 c). Each table contains
statistics for all batches taken together and each batch separately, in the successive rows. The
content of the tables cannot be directly modified by the user, however, it gets instantly updated
when results are deleted from the results dialog or new results are added. The Results dialog
also lets you calculate descriptive statistics for a selected subset of results, which are displayed
in a similar dialog.
41
Fig. 28. Live statistics dialog. The statistics are grouped in several tables (c), each for a different quantity,
under its own tab. Each table contains statistics for all batches taken together (i.e. all results in the
Results dialog) and for each batch separately.
Statistics dialogs can be displayed from the Results dialog (Fig. 21):
• to display descriptive statistics for all results, click the toolbar icon (Fig. 21 b) or go
to Statistics (Alt + T) → Descriptive statistics (Alt + D).
• to display descriptive statistics for part of results, select the corresponding rows in the
results table, then move the mouse over the selected rows and click the right mouse
button. Click Selection statistics in the context menu (Fig. 21 e).
The statistics dialogs are equipped with the main menu (Fig. 28 a) and a set of buttons (b), that
allow you to print, save to a file or customize the calculated statistics:
• to save the currently displayed table to a file, use the shortcut Ctrl + S, or click the Save
button (Fig. 28 b), or go to File (Alt + F) → Save (Alt + S). In each case, a file dialog will be
displayed, which lets you choose where the table should be saved and in which format
(TSV or CSV).
• to print the currently displayed table, use the shortcut Ctrl + P, or click the Print button
(Fig. 28 b), or go to File (Alt + F) → Print (Alt + P)
42
F. 8. Recalculation
You can recalculate all curves, the selected curves or only the curves from a particular batch. If
you want to delete old values, select the checkbox Delete old calculations, otherwise both old
and new results will be stored. Independently, you may specify whether the old curve charts
should be removed (select Delete old curve charts ) and whether the maps of mechanical
quantities (if any) should be modified (select Modify maps in place), or whether new maps
should be created and old ones kept intact.
The Graphical results dialog stores charts generated during processing of force curves. Three
different types of charts can be generated for each curve (depending on the settings in the
Processing assistant):
a) estimated contact point. It can be dragged with the mouse if the icon is
pressed in. The curve is then recalculated instantly with the new contact point.
b) transition point. It can dragged it with the mouse if the icon is pressed in.
The curve is then recalculated instantly, with the new transition point.
c) adhesion force marker, a vertical line segment, whose x coordinate is the position
of largest adhesion, and its length is equal to the magnitude of adhesion force. It
is visible only if the adhesion force has been calculated for the plotted force
curve. It can be dragged and its ends can be moved with the mouse if the icon
is pressed. To draw or modify additional adhesion markers, press
d) fit of the indentation model, determined by the selected tip shape and sample
thickness. The fit is drawn for the branch used in calculations (either withdraw or
approach)
43
• the force – indentation curve, annotated with:
a) transition point
b) fit of the indentation model
a) transition point
b) fit of the indentation model, which is always a straight, horizontal line. The
position of fit corresponds to overall Young’s modulus.
The Graphical results dialog contains a main menu bar (Fig. 30 a), a toolbar (e), a set of buttons
(b) for the most commonly used actions, a central split pane, with the list of processed force
curves on the left (Fig. 30 c) and the tabbed panel displaying the corresponding charts on the
right, with each type of chart under its own tab (d). The menu bar contains four menus. The File
menu and Measure menu contain the options we discussed in the section D.3. Force curve
dialog.
Toolbar options:
Correct contact point - if pressed , you can drag the contact point with the
mouse. The force curve will be instantly recalculated, also the corresponding
maps and result dialog entries will be updates.
Correct transition point - if pressed , you can drag the transition point with the
mouse. The force curve will be instantly recalculated, also the corresponding
maps and result dialog entries will be updated.
Add adhesion marker – if pressed, you can draw and drag new adhesion
markers with the mouse. New columns are added to the result dialog, storing
the corresponding adhesion force values. Useful when there are multiple
adhesion events in the curve.
Correct adhesion – if pressed, you can drag the adhesion marker (a vertical bar
in the plot) and move its ends. The adhesion force will be instantly recalculated
and the corresponding maps and result dialog entries will be updated.
Mark on the maps marks the position corresponding to the selected curves on
maps of the mechanical quantities, if they are available.
44
Measure lets you measure distances on the chart.
Crop lets you crop the end of all the curves or the selected ones.
Gaussian smoothing lets you smooth the curves (all or the selected ones) with
Gaussian filter.
Median smoothing lets you smooth the curves with the median filter.
Convolve
• Live chart style (Ctrl + L) opens a dialog that lets you modify
chart style, named the Live chart style dialog (see section
H.1. Working with Live chart dialog). Click this to modify the
way the force curves are plotted.
• Raw data opens a raw data dialog (see the section D.2. Raw
data dialog), containing the coordinates of points for the
selected force curves.
• Overlay (Ctrl + Y) lets you plot the results for two force
curves in a single chart. It opens the Overlay assistant,
where you can select the curves to overlay. This feature is
useful when you want to compare two different force
curves or to compare the results of processing obtained for
the same curve, but with different settings, as in Fig. 31.
• Fix contact fixes contact point in the force curve plot that is
currently displayed.
• Fix contact for all fixes contact point in all force curves plots stored in the dialog.
45
The list of the processed curves has a context menu containing one additional action – Jump to
results, which shows the numerical results for the selected force curves in the Results dialog.
46
G. Mouse actions on charts
The effects of mouse clicks or movements over charts depend on which mouse mode is active.
Only one mode is active at a time. You can change the mode through the main menu or a
toolbar, if available. There are seven main mouse modes:
• Normal mode
• Measure distance mode
• Extract profile mode (or Extract slice for stacks)
ROI modes:
For line charts and histograms, only the normal and the measure distance mode can be used. All
mouse modes are available for maps, images and stacks. For slices you can use all the modes
except the extract profile mode.
The following actions work in the normal mode. They also work for all other modes:
a) left click in the left upper corner of the zoom-in area, then drag the mouse
downwards and leftwards. The area highlighted in semi-transparent bluish color
will be zoomed-in after you release the mouse.
b) roll up the mouse-wheel. The zoomed-in area will be centered around the mouse
cursor.
c) click Zoom-in action, accessible from the context menu. You can zoom in both or
just a single axis. The context menu is triggered by a right mouse click.
47
a) left click anywhere in the chart, then drag the mouse in any direction except in
the left down direction.
b) roll down the mouse wheel.
c) click Zoom-out action, accessible from the context menu. You can zoom out both
or just a single axis. The context menu is triggered by a right mouse click.
• Panning – to pan a chart (also when zoomed), hold down the Ctrl key, then left click
anywhere in the chart and drag the mouse
• Adjusting title, legend and scale bars positions – to move a title, legend or a scale bar
drawn inside the plot (i.e. within the data area), move the mouse over it, press the left
button and drag. The element you caught will follow the mouse movements.
In this mode, you can measure distance between two selected points on the chart or the length
of a polyline drawn on the chart. If the measurement is done on an image or map chart, it will
be copied to all other channels from the same image. Thanks to this feature you can easily
compare the size and position of a given structure in all channels.
• to switch on/off this mode, click Measurements in the menu bar, then select the type of
the measurement in the drop-down menu. Three options are available: Line
measurement, Polyline measurement and Free hand measurement. In the Image dialog
and the Map dialog, you can click one of the toolbar icons with a pair of compasses
instead: (line), (polyline) or (free hand measurement).
• to add a new measurement, click the left mouse button on the start point, then move the
mouse towards the end point. To finish drawing a linear measurement, just click the left
48
button at the endpoint. For polyline measurements, you can add nodes with left mouse
click, while to finish drawing a polyline or free hand measurement, click the right mouse
button. When drawing is finished, a distance measurement dialog will pop up, containing
detailed information about the length and position of each measurement line.
• to translate a measurement, move the mouse over the measurement line, press the left
mouse button and drag the mouse. The measurement line will follow the mouse.
• to modify a measurement, move the mouse over the measurement and single click the
left mouse button. Moveable circles will appear at the ends of the line measurement, or in
the nodes of the polyline/free hand measurement. To reposition one of the endpoints or
nodes, move the mouse over the circle, press the left button and drag the mouse. To hide
the circles, single – click the left button on the measurement line.
• To remove or add a vertex to a polyline or free hand profile, move the mouse over its
edge and single click the left mouse button. Circles will appear at the vertices. To remove
a vertex, single click the left button on the circle while holding the Ctrl key down. To add
a new vertex, single click the left button on a vertex circle while holding the Shift key
down. The clicked vertex will be split into two adjacent vertices.
• to remove a measurement, move the mouse over the measurement line and double click
the left mouse button.
The style of the measurement lines can be modified with the Measurement style dialog (go to
Measure → Measurement style). For details, see the section H.2. Edit measurement style.
In this mode, you can extract height profiles from maps, images or stacks along an arbitrary
section line, polyline or free-hand path. For multi-channel images and maps, profiles are
automatically extracted from all the channels along the profile. This makes it easier to find
correlations between the imaged properties.
1. in the Map or Image dialog, go to Profiles (Alt + P), then select the type of the profile
in the drop-down menu. Three options are available: Line profile, Polyline profile
and Free hand profile. Alternatively, you can click one of the toolbar
49
icons with crossed knives: (line), (polyline) or (free hand
profile)
2. in a Stack dialog, go to Sections (Alt + T), then select either Line section, or Polyline
section, or Free hand section in the drop-down menu.
• to add a new profile, click the left mouse button on the start point, then move the
mouse towards the end point. To finish drawing a linear profile, just click the left button
at the endpoint. For polyline profiles, you can add nodes with left mouse click, while to
finish drawing a polyline or free hand profile, click the right mouse button. The section
line will be drawn on all channels of the current image and the cross-section dialog will
pop up, containing chart with height profile extracted from all channels.
• to translate a profile, move the mouse over it, press the left mouse button and drag the
mouse. The line will follow the mouse.
• to modify a profile, move the mouse over it and single click the left mouse button.
Moveable circles will appear at the ends of the line profiles or the nodes of the
polyline/free hand profiles. To reposition one of the endpoints or nodes, move the
mouse over the circle, press the left button and drag the mouse. To hide the circles,
single – click the left button on the line.
• to remove or add a vertex to a polyline or free hand profile, move the mouse over its
edge and single click the left mouse button. Circles will appear at the vertices. To remove
a vertex, single click the left button on the circle while holding the Ctrl key down. To add
a new vertex, single click the left button on a vertex circle while holding the Shift key
down. The clicked vertex will be split into two adjacent vertices.
• to remove a profile, move the mouse over the cross-section line and double click the left
mouse button.
Profiles are equipped with triangular, moveable markers. For each marker, another marker
is drawn in the cross-section chart in the same position. Both in the original image and the
cross-section chart markers can be moved, added or removed with the mouse.
• to add a new marker, move the mouse over the section line and click the right mouse
button or double click the plotted curve in the cross-section chart.
50
• to move a marker, move the mouse over, press the left mouse button and drag the
mouse. The marker will follow the mouse.
• to remove a marker, move the mouse over and double click the left mouse button.
The style of section lines can be modified with the Profile style dialog (click Profiles in the main
menu bar, then Profile style). For details, see the section J.4. Editing style of profile lines.
G.4.ROI modes
ROI modes allow you to select regions of interest of different shapes. Various actions can be
then applied to these areas. For multi-channel images and maps, ROIs are selected for all the
channels at once.
• to switch on/off this mode, go to ROI (Alt + R), then select from the drop-down menu
the ROI mode of your choice, one of the following: Polygon ROI, Rectangular ROI,
Elliptic ROI or Free hand ROI. In the Image dialog and the Map dialog, you can select
the right shape from the toolbar.
• to create a rectangular or elliptic ROI, click the left mouse button at the desired
position. A small ROI will be drawn, which can be then reshaped.
• to create a polygon ROI, click the left mouse button, move the mouse to the next
vertex and click the left mouse button there. Repeat this until all vertices are added.
To finish, click the right mouse button.
• to create a free-hand ROI, click the left mouse button and move the mouse. ROI
boundary will follow mouse movement. To finish, click the right mouse button.
• to resize a rectangular or elliptic ROI, move the mouse over its edge or a corner,
press the left mouse button and drag the mouse. To keep the aspect ratio fixed, hold
the Alt key down. To resize the ROI around its center, hold the Ctrl key down while
dragging the mouse. To both keep the ROI aspect ratio fixed and to resize it around
the center, hold the Alt gr key down or alternatively both the Ctrl and Alt keys.
• to reshape a polygon or free hand ROI, move the mouse over its edge and single click
the left mouse button. Moveable circles will appear at the vertices. To reposition one
51
of them, move the mouse over the circle, press the left button and drag the mouse.
To hide the circles, single – click the left button on the line.
• To remove or add a vertex to a polygon or free hand ROI, move the mouse over its
edge and single click the left mouse button. Circles will appear at the vertices. To
remove a vertex, single click the left button on the circle while holding the Ctrl key
down. To add a new vertex, single click the left button on a vertex circle while holding
the Shift key down. The clicked vertex will be split into two adjacent vertices.
• to translate a ROI, move the mouse over it, press the left mouse button and drag the
mouse. The ROI will follow the mouse. If a ROI is composed of multiple disjoined
regions (ex. after merging of distinct ROIs), each ROI can be moved separately. To
move all regions at once, hold the Ctrl key down while dragging the mouse.
• to rotate a ROI, go to ROI (Alt + R) → Rotate ROIs or press in the icon in the
toolbar. Move the mouse over a ROI, press the left mouse button and drag the
mouse. The ROI will get rotated around the center of its bounding box. If a ROI is
composed of multiple disjoined regions (ex. after merging of distinct ROIs), each ROI
can be rotated separately. To rotate all regions of a composite ROI at once around the
center of its bound box, hold the Ctrl key down while dragging the mouse. To rotate
all regions of a composite ROI at once around the center of one of its components,
hold the Shift key down while dragging the mouse.
• to remove a ROI, move the mouse over its edge and double click the left mouse
button.
You can also draw ROIs which includes whole image, except for a hole (Fig. 33 B). The shape of
the hole depends on the current ROI mode. For example, when drawing holes is switched on
and the Rectangular ROI mode is on, you can select a ROI which includes the whole image
except for a rectangle. To draw ROIs with a hole, click ROI in the menu bar, then select in the
drop-down menu the Draw ROI hole item
A B
52
a b
The style of ROIs can be modified with the ROI style dialog (go to ROI → ROI style). For details,
see the section J.9. Editing style of ROIs.
AtomicJ gives you a full control over the style of charts. This flexibility in customizing chart style
combined with the wide range of supported files allows you to create high quality figures.
There are a few ways to customize the style of a chart, depending on the type of data it shows.
Most chart features can be modified with the Live chart style dialog (Fig. 34), which is available
for every type of charts and stacks. In addition, color gradients for stacks, image charts and map
charts can be modified with the Color gradient dialog and the Gradient range dialog. The style
of regions interest can be modified with the ROI style dialog (section J.9.), the style of
measurement lines with the Measurement style dialog (section H.2.), while the style of profile
lines with the Profile style dialog (section J.4.).
The Live style dialog will help you to modify the appearance of almost every detail of every type
of chart. To display the Live chart style dialog you can either:
1. move the mouse over the chart and click the right mouse button. A context menu will
appear. Click the Live chart style action, the first item in the context menu.
2. or double-click an axis, scale bar or the frame of a legend. The live chart style dialog will
be displayed. The tab corresponding to the clicked element will be selected.
3. or go to Chart → Live chart (Ctrl + L).
53
Fig. 34. Live style dialog gives you
almost full control over the styles
used to render different elements
of charts, such as data series, axes,
title, and legends.
The dialog consists of the main menu bar (Fig. 34 a), a set of buttons (b) for the most commonly
used actions, and the central tabbed pane. Each tab contains an editor with controls for
modifying style of different chart elements (d).
The main menu bar contains the Style menu (Alt + T), with six options:
• Reset to defaults (Ctrl + R) resets the style for the current chart element to the default
style (see above).
• Undo (Ctrl + U) reverts the changes introduced to the current chart element.
• Batch apply all (Ctrl +Shift + B) applies the style for all chart elements to all charts of the
same type in the same dialog. Charts of the same type are usually grouped under the
same tab in a chart dialog.
• Save all as defaults (Ctrl + Shift + S) saves the style for all chart elements to an OS-
dependent backing store. From now on, this style will be used to as the default style to
draw the charts of the same type as the currently modified chart.
54
• Reset all as defaults (Ctrl + Shift + R) resets the style for all chart elements to the default
style
• Undo all (Ctrl + Shift + U) reverts the changes introduced to any of the chart elements
The chart is instantly updates when its style is edited. This feature helps you to find the best
style for the chart, as you can immediately see what the effects of the changes you are
introducing are.
The Live chart dialog consists of separate editors, each for different chart element. The number
and type of editors depend on the type of chart. Those editors are available for every chart:
a. Series editor
Series editor lets you modify the way the data are renderer. Different forms of this editor are
available for different types of charts. A series editor for a line chart is displayed in Fig. 34.
A more detailed description of series editors specific for histograms, images and maps can be
found in later sections.
b. Plot editor
Plot editor (Fig. 35) lets you modify the style of the plot, understood here as the inner area of
the chart, where data are actually draw. The editor allows you to set the plot background, to
specify the stroke of plot’s outline, hide or show range and domain gridlines, as well as to
change their stroke and color. The interval between gridlines is always equal to the interval
between the ticks on the corresponding axis. In addition, you can select the plot orientation.
There are two options:
1. Vertical orientation - the domain axis is horizontal, the range axis is vertical. This is the
usual orientation.
2. Horizontal orientation - the domain axis is vertical, the range axis is horizontal.
55
c. Axes editor
Axes editor (Fig. 36) contains separate subeditors for each axis. Most line charts contain two
axes – domain and range axis, while stacks, as well as image and map channels contain also a
third axis – depth axis, which is drawn next to the gradient bar. Those subeditors give you
control over every detail of the axis. Here, you can hide or show the axis, invert it or change its
range. You can also modify its main label, tick marks, ticks labels, axis line, position in the chart
or the number format (ex. the decimal separator).
You will usually want to keep the style of all axes the same. Axes editor allow you to choose
whether the axes are allowed to differ in their style:
• to ensure that changes in style of one axis are also applied to other axes, select (press
in) the chain link button. When selected, its icon changes from to . .
The Axes editor can be opened as a tab in the Live chart dialog, but it can be accessed even
faster:
• to open the Axes editor, move the mouse cursor over one of the axes and double click
the left mouse button.
d. Title editor
The Title editor (Fig. 37) lets you add or modify the title of the chart. It consists of three tabs:
1. Text tab allows you to set text of the title, its font and color. Instead of simple color, you
can fill the letters with a gradient (check in the Use gradient box and then click the
Select button to choose the gradient).
56
2. Frame tab allows you to choose whether frame should be drawn around the title. You
can select the color and stroke of the frame, and the color of the title background.
3. Position tab allows you to choose whether the title should be drawn inside or outside
the area of the plot occupied by plotted data and in which position. Here you can also
set the title padding and margins. Note that the title can be simple dragged with the
mouse on the chart – for details see the section G. Mouse actions on chart.
The Title editor can be opened as a tab in the Live chart dialog, but it can be accessed even
faster:
• to open the Title editor, move the mouse cursor over the title and double click the left
mouse button.
Scale bar (Fig. 38) editor contains two subeditors – one for the domain (usually horizontal) scale
bar, the other for the range scale bar (usually vertical). Here you can hide or show the scale
bars, change their position and length. The default length of a scale bar is equal to the interval
between axis ticks. The editor allows you to select the stroke and color of the scale bars, as well
as the size and font of its labels. In addition, the scale bar position can be set. Note that the
scale bars can be simple dragged with the mouse on the chart – for details see the section G.
Mouse actions on charts.
The Scale bar editor can be opened as a tab in the Live chart dialog, but it can be accessed even
faster:
• to open the Scale bar editor, move the mouse cursor over the scale bar and double click
the left mouse button.
57
Fig. 38. The editor for scale bars
f. Legend editor
Legend editor (Fig. 39) lets you change the style and position of a legend. If the chart has
multiple legends, for each of them a separate editor is included in the Live chart style dialog.
The Legend editor contains two or three (depending on the type of the legend) tabs:
1. Position tab lets you choose whether the legend should be drawn inside or outside the
area of the plot occupied by plotted data and in which position. Here you can also set
the legend padding and margins. Note that the legend can be simple dragged with the
mouse on the chart – for details see the section G. Mouse actions on charts.
2. Style tab lets you set the font and color used for text in the legend. You can also choose
whether frame should be drawn around the title, select the color and stroke of the
frame, and the color of the title background. Note that legends for image and map
charts contain depth axes that can be edited from the Axes editor, but not from the
Legend editor.
3. Strip tab lets you set the outline and width of the color gradient strip that is the main
part of legends for image and map charts. This tab is included in the Legend editor only if
the edited legend contains a strip.
58
g. Chart editor
Chart editor (Fig. 40) lets you change the chart padding, background color and whether
antialiasing should be used. In addition, the lock aspect ratio action allows you to specify
whether natural aspect ratio of the chart (i.e. determined by unit lengths on the axes) should be
preserved when the dialog is resized.
All types of charts support distance measurements. The style of measurement lines can be
edited through the Measurement style dialog (Fig. 41). To display Measurement style dialog,
got to Measurement →Measurement style.
The dialog (Fig. 41) lets you set different style for finished measurements and the measurement
lines that are being drawn. The measurement lines get instantly updated when you introduce
changes to their style, which makes it easier to find the optimal style parameters.
59
You can edit the stroke of the measurement lines,
as well as the color and font of the labels. The X
distance and Y distance checkboxes controls
whether measurements of distance in the x – and
y – direction (Fig. 42 B) should be also drawn, or
skipped (A). If the quantities on the x – and y –
axis are different, the Euclidean distance cannot
be calculated, so only the distances in the
Fig. 42. The simple measurements (A) and
directions of axes are reported.
measurements with distances in x – and y –
directions (B)
Actions accessible through buttons (Fig. 41 b):
• Batch apply (Alt + B) applies this measurement style to all charts of the same type in the
same dialog. Charts of the same type are usually grouped under the same tab in a chart
dialog.
• Save as defaults (Alt + S) saves this measurement style to an OS-dependent backing
store. In case of MS Windows, the backing store is simply the Windows Registry. From
now on, this style will be used to as the default style to draw measurement lines on the
charts of the same type as the currently modified chart.
• Reset to defaults (Alt + R) resets the style of measurement lines to the default style (see
above).
• Undo (Alt + U) reverts the introduced changes.
H.3.Gradient dialog
The gradient dialog (Fig. 43) allows you to modify the color gradient used to find the color for
each z-axis value (ex. height in topographical images). The range and the colors that form the
gradient are important when you want to bring out interesting features of your image, map or
stack.
The Range panel is where you specify the range of the gradient. The minimal value will
be colored with the first color in the gradient, the maximal value with the last one. You can
specify the range in three ways:
60
3)
The Full range panel reports what is the smallest and the
largest value in the data.
The ROI masks panel allows you to specify a different color, independent of the gradient, which
will be used for the parts of the chart outside or inside regions of interest.
Color gradient editor (Fig. 45) allows you to modify the current color gradient or to choose a
new one. A gradient is determined by its color nodes, their position and the method of
interpolation between them. The control over the color gradient is provided by interactive
gradient bar (Fig. 45 a), a set of buttons (b) and widgets that control how colors are interpolated
(c).
61
• to reverse the order of colors, click the Invert button
• to choose a new gradient among the built-in or user-
defined gradients, click the Built-in button. The Select
gradient dialog (Fig. 47) will be shown (see section H.6.).
• to specify the number of colors in the palette, use the
Palette size field. This is the total number of different
colors in the gradient, obtained through interpolation of
the color nodes.
• to change the method of interpolation between color
nodes (Fig. 44), use the Mixing combo box.
• to read in a gradient saved in a file, click the Import
button
• to save the current gradient to a file, click the Export
Fig. 44. Colors making cup gradient can
button be linearly mixed (a) or form bands (b).
62
H.5. Gradient range dialog
The Gradient range dialog contains the data histogram plotted against the color gradient used
for their visualization (Fig. 46). The area painted with the color gradient is flanked by markers.
The gradient range can be specified in several ways:
• to select the range manually, move the mouse over one of the markers flanking
colored area, press the left mouse button and drag the marker. Release the left
button when the marker is in the right position.
• to choose the automatic range, which contains all the data except the distribution
tails, go to Range (Alt + G) → Restore automatic (Ctrl + R) .
• to choose the full range, which fits all the data, go to Range (Alt + G) → Restore full
(Ctrl + Shift + R)
You can modify histogram binning to reveal all its features. Go to Data (Alt + D) → Modify
binning (Ctrl + B)
Fig. 46. Gradient range dialog. To modify the range of values that the gradient is
applied to, drag with the mouse the lines on the edges of colored area.
63
H.6. Select gradient dialog
64
I. Saving charts to files
AtomicJ gives you much flexibility in saving charts. It supports multiple image formats, both
raster (TIFF, JPEG, JPEG2000, Windows Enhanced Metafile, GIF, PPM, Bitmap and AVI for
movies) and vector formats (EPS, PS, PDF, SVG), as well as text formats (TSV and CSV). You can
save a single chart or multiple charts at once. The actions described in this section are applicable
to any type of charts, except the part on saving as a movie, which is applicable only to stacks.
You can specify different option for each of the file formats, which influence the quality and size
of the saved images. The dialogs for saving charts are equipped with panels that collect those
settings. Here, they are shown after short description of options available for different formats.
i. Size of the output page. If Custom is selected, you can specify the size in the Width and
Height fields, in printer’s point (pts., equal to 1/72 inch). These fields are greyed out if a
predefined size is selected. The following sizes are available: A0 – A6 , B0 – B6, Executive
(522 x 756 pts.), International (595 x 791 pts.), Legal (612 x 1008 pts.), Letter (612 x 792
pts.), Ledge (792 x 1224 pts.)
ii. Margins. The following options are available: None (no margins are left), Small (20 pts.),
Medium (30 pts.), Large (40 pts.).
Please note that if the Keep aspect ratio checkbox is selected, the margins may be made
larger to keep the original aspect ratio.
iv. Fit to page. If this checkbox is selected, the chart is drawn as large as possible, with the
only constraints being the size of page, margins, as possibly the aspect ratio, if the Keep
the aspect ratio checkbox is selected.
v. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the chart is preserved –
when you change the width, the height is updated accordingly, and the other way round.
65
vi. Save only data area. If this checkbox is selected, only the inner part of the chart, where
the image is actually drawn, is saved. Axes are left out, just are those legends or titles
that are draw on the margins of the chart.
Fig. 48. Dialog for setting options for the EPS file format. Dialogs for the PDF and PS formats are
identical.
i. Size of the image file. Enter the Width and Height of the image in printer’s points.
ii. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the chart is preserved –
when you change the width, the height is updated accordingly, and the other way round.
iii. Save only data area. If this checkbox is selected, only the inner part of the chart, where
the image is actually drawn, is saved. Axes are left out, just are those legends or titles
that are draw on the margins of the chart.
Fig. 49. Dialog for setting options for the PNG file format. Dialogs for the SVG, GIF, PPM and
bitmap formats are identical.
66
TIFF. You can specify:
i. Compression. You can save the file Uncompressed, or use the LZW (lossless Lempel-Ziv
and Welch algorithm) or PackBits (lossless) compression.
ii. Size of the image file. Enter the Width and Height of the image in printer’s points.
iii. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the chart is preserved –
when you change the width, the height is updated accordingly, and the other way round.
iv. Save only data area. If this checkbox is selected, only the inner part of the chart, where
the image is actually drawn, is saved. Axes are left out, just are those legends or titles
that are draw on the margins of the chart.
Fig. 50. Dialog for setting options for the TIFF file format.
i. Quality. The image quality can be specified as a number between 1 (highest quality) and
0 (lowest).
ii. Size of the image file. Enter the Width and Height of the image in printer’s points.
iii. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the chart is preserved –
when you change the width, the height is updated accordingly, and the other way round.
67
iv. Save only data area. If this checkbox is selected, only the inner part of the chart, where
the image is actually drawn, is saved. Axes are left out, just are those legends or titles
that are draw on the margins of the chart.
Fig. 51. Dialog for setting options for the JPEG file format.
JPEG 2000.
ii. Quality. The image quality can be specified as a number between 1 (highest quality) and
0 (lowest). Quality cannot be set when lossless compression is used.
iii. Size of the image file. Enter the Width and Height of the image in printer’s points.
iv. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the chart is preserved –
when you change the width, the height is updated accordingly, and the other way round.
v. Save only data area. If this checkbox is selected, only the inner part of the chart, where
the image is actually drawn, is saved. Axes are left out, just are those legends or titles
that are draw on the margins of the chart.
68
CSV, TSV. These are text formats. You can specify:
ii. Trailing zeroes. The checkbox controls whether the trailing zeroes are shown.
iv. Thousand separator (space, comma, dot or apostrophe. The Use separator checkbox
control whether the thousand separator is shown.
Fig. 52. Dialog for setting options for the CSV file format. The dialog for TSV format is identical.
AVI. Stack can be saved as a movie in this format. You can specify:
i. Compression. You can save the movie Uncompressed, or use the PNG
compression (lossless, please note that you may need a codec to open the file!)
or JPEG compression (lossy).
ii. Backwards. If this checkbox is selected, the movie is saved in reverse (as if played
backwards)
iii. First frame. The first frame of the stack that will be included in the movie. You
can set this to a higher value to skip the initial frames of the stack, which often
appear noisy due to low signal-to noise ratio.
iv. Last frame. The last frame of the stack that will be included in the movie. You can
set this to a smaller value to skip the last frames of the stack.
69
v. Rate. The rate of frames per second. Changing the frame rate results in the
changes of the movie length, because the number of frames is constant.
vi. Length. The length of the movie. Changing the movie length results in the
changes of the frame rate, because the number of frames is constant.
vii. Size of the frames. Enter the Width and Height of the frames in printer’s points.
viii. Keep aspect ratio. If this checkbox is selected, the aspect ratio of the stack
frames is preserved – when you change the width, the height is updated
accordingly, and the other way round.
ix. Save only data area. If this checkbox is selected, only the inner part of the chart,
where the image is actually drawn, is saved. Axes are left out, just are those
legends or titles that are draw on the margins of the chart.
Fig. 53. Dialog for setting options for the AVI file format. The dialog allows you specify the
compression method, the range of the movie (i.e. which frames should be included), its speed
and the size of an individual frame.
70
I.2. Saving single charts
1. click File (Alt + F) in the menu bar in the dialog where the chart is displayed, then click
Save (Alt + S) in the drop-down menu.
2. move the mouse over the chart, then click the right mouse button. Click the Save as
item in the context menu.
3. use the shortcut Ctrl + S.
In each case, the Save dialog will be
displayed (Fig. 54). It lets you specify
where the chart is to be saved (the
Browse button, Alt + B), select the
format and specify the saving options
(the Options button, Alt + P). The
saving options are format – specific
and are described above.
71
I.3. Saving multiple charts
AtomicJ allows you to save at once multiple charts displayed in the same dialog. Saving charts to
files is parallelized (i.e. uses multiple processor threads), except when they are compressed in
ZIP file. Parallelization gives sizeable performance gain on multi-core processors. Multiple charts
can be saved through the Batch save dialog, To display this dialog, you can either:
1. click File (Alt + F) in the menu bar in the dialog where the chart is displayed, then click
Save all in the drop-down menu.
2. click the Save all button at the bottom of the chart dialog
3. use the shortcut Ctrl + Shift + S.
The Batch save dialog will be then displayed. The dialog is where you specify which charts are to
be saved and where, in which format, and how to name the files. It consists of three tabs:
Fig. 56. Dialog for saving multiple charts at once. The tab with format settings is displayed.
72
specific parameters that are described in
the subsection Options for file formats.
1. Prefix – by default the name of the type of chart plus underscore (ex. “Young’s
modulus_”).
2. Root – the central part of the name, by default the name of the file with the AFM
recording corresponding to the chart.
3. Suffix - by default a number incremented for each chart of a given type.
1. Arbitrary string. The string can be typed into the appropriate field in the File
naming tab.
2. Name of the file with the AFM recording corresponding to the chart
3. Serial number, i.e. the file with the AFM recording corresponding to the chart.
The initial serial number, used for the first chart, can be set in the Initial serial #
field.
4. Type of the chart plus underscore (ex. “Young’s modulus_” ).
73
I.4. Saving multiple stack frames
You can save at once all or a subset of frames from a stack. Saving charts to files is parallelized
(i.e. uses multiple processor threads), except when they are compressed in ZIP file.
Parallelization gives significant performance gain on multi-core processors. Multiple frames can
be saved through the Save all frames dialog. This dialog can be display from the Stack dialog in
the following ways:
1. click File (Alt + F) in the menu bar in the Stack dialog, then click Save all frames in the
drop-down menu.
2. use the shortcut Ctrl + Shift + S.
The Save all frames will be then displayed. The dialog is where you specify which frames are to
be saved and where, in which format, and how to name the files. This dialog is similar to the
Batch save dialog, described in the previous section. It consists of three tabs:
Format tab. Here you can specify the options for the AVI format (currently the only movie
format available) that are described in the subsection Options for file formats.
File naming tab. This tab is similar to the File naming tab in the Batch save dialog, described in
the previous section.
74
J. Working with images
Atomic Force Microscope records different type of sample images, depending on the mode of
operation. Usually the AFM image consists of several channels, each for different quantity,
because multiple quantities may be simultaneously recorded during scanning (topography, error
signal and friction, for example). We will refer to a set of image channels simultaneously
recorded in the same scan area as an Image recording. All types of image recordings, whatever
the physical meaning of their channels, are read - in in the same way, described in the section D.
Viewing AFM recordings, and are stored in the same Image dialog (Fig. 59).
Fig. 59 The Image dialog stores the AFM images opened in the current session. It contains a list of image
files (d) and a central panel (c), displaying each image channel under its own tab. Images can be
manipulated and analyzed through action accessible from the main menu bar (a), toolbar (b) and a set
of buttons (c).
The Image dialog is equipped with a menu bar (Fig. 59 a), a toolbar (b) and a set of buttons (c)
for the most commonly used actions and a list for navigating between the read-in image
recordings (d). The list is equipped with a context menu (d). The central panel (e) displays the
channels for the currently selected image recordings, each under its own tab.
While working with images, you will use mainly the actions included in the menus from the
menu bar and the toolbar, which provide easy access to the most frequently used actions.
75
Toolbar options :
Draw histogram (Ctrl + D) lets you generate histograms for whole image
channels.
Draw ROI histogram (Ctrl + Shift + D) lets you generate histograms of values
inside or outside selected regions of interest (ROIs).
Show histograms (Ctrl + H) opens a histogram dialog with all the histograms
generated for images in the current session. It is initially disabled, as there are
no box plots to show.
Show box plots opens the Box plots dialog, which stores all box plots generated
for image ROIs in the current session. It is initially disabled, as there are no box
plots to show.
Draw ROI box plots. opens the Box plots assistant, which lets you generate box
plots for selected ROIs. See also section F.5. Creating box plots.
Statistics (Ctrl + T) shows statistics (including mean, median, variance) for each
image channel.
Statistics ROIs (Ctrl + Shift + T) shows statistics for the selected ROIs
Polygon ROI toggles polygonal Regions of Interest (ROIs), changing the way
mouse clicks are interpreted by the chart. If this action is switched on you can
add a new polygonal ROI with the mouse. See the section J.5. Creating and
modifying Regions of Interest.
Wand ROI tool lets you build a ROI which includes a part of image with values
close to the value in an image position selected with the mouse.
76
ROI rotation lets you rotate a ROI around the center of its bounding box.
Individual components of a composite ROIs (i.e. multiple ROIs merged) can be
either rotated separately, or all rotated at once (around the center of a
particular component if the Shift key down, around the center of the whole
composite ROI if the Ctrl key is down).
Line measurement lets you draw a new straight line measurement on the image
with the mouse.
Free hand measurement lets you draw a polyline measurement with the
mouse. The measurement line will follow the movement of the mouse.
Line profile lets you draw profile lines on the image with the mouse. The cross-
section dialog appears, displaying height profiles for all profile lines.
Polyline profile lets you draw a polyline profile (i.e. composed of several
connected line segments) with the mouse. Click left mouse button to add new
profile node (segment of the polyline), click right mouse button to finish
drawing of the profile.
Free hand profile lets you draw a polyline profile with the mouse. The profile
will follow the movement of the mouse. Click right mouse button to finish
drawing of the profile.
Convolve opens the Convolution dialog, which lets you convolve the image with
a kernel of arbitrary size.
Overlay (Ctrl + Y) lets you add channels from another image as new layers to the
charts generated for the currently selected image. It brings up the Overlay
assistant. Charts for the currently selected image are copied and new image is
created.
Gridding lets you increase or decrease the number of pixels in image through
bilinear, bicubic, nearest neighbor and Sibson interpolation. The SIbson’s
interpolation is suitable for scattered images (not on a regular grid. )See:
Smoothing dialog.
77
Lock aspect - when switched on (the button is pressed in), the natural aspect
ratio of the image (i.e. determined by its width and height in microns) is
preserved when the dialog is resized.
Menu options:
The menu bar contains four menus and can be navigated with the mouse or using keyboard
mnemonics and accelerators in brackets:
• File menu (Alt + F) (the same as in the Force curve dialog, see section D.3.)
• Measure menu (the same as in the Force curve dialog, see section D.3.)
• Chart (Alt + C) menu lets you modify the chart style:
o Live chart style (Ctrl + L) opens the Live chart style dialog.
Click to modify the way image is plotted.
o Edit color gradient (Ctrl + G) opens the Color gradient
dialog. See Color gradient edition for details.
o Gradient histogram (Ctrl + Shift + G) opens the Gradient
range dialog, where you can easily set the range of color
gradient against the image histogram. See Color gradient
edition for details.
o Resize (Ctrl + M) (also in the toolbar)
o Overlay (Ctrl + Y) (also in the toolbar
o Lock aspect (also in the toolbar)
• Data menu:
o Draw histogram (Ctrl + D) (also in the toolbar) opens the
Histogram assistant, which lets you generate histograms
for each of the image channel.
o Draw ROI histograms (Ctrl + Shift + D) (also in the toolbar)
opens the Histogram assistant, which lets you generate
histograms for the area inside ROIs. You can create at once
a histogram for each channel of the selected image.
o Show histograms (Ctrl + H) (also in the toolbar) opens the Live histogram dialog,
displaying histograms for the whole image channels.
o Live ROI histogram (Ctrl + Shift + H) opens the Live histogram dialog, displaying
histograms of the data inside regions of interest for each channel of the currently
78
selected image. These histograms are instantly updated when you move, add or
delete a ROI.
o Show box plots (also in the toolbar) displays the box plots generated for image ROIs in
the current session.
o Live ROI box plots opens the Box plot dialog, displaying box plots of the data inside
regions of interest for each channel of the currently selected image. These histograms
are instantly updated when you move, add or delete a ROI.
o Statistics (Ctrl + T) (also in the toolbar) opens a statistics dialog, containing mean,
median, variance and other statistics for each image channel (see the section J.6.
Descriptive statistics for ROIs).
o Statistics for ROIs (Ctrl + Shift + T) (also in the toolbar) opens a statistics dialog, with
statistics calculate for each image channel, only for the parts of images inside the
regions of interest.
o Raw data (Ctrl + W) opens the Raw data assistant, where you can select the image
channels for which you want to view raw data. The raw numerical values
corresponding to pixels in images are then displayed in a raw data dialog (see section
D.2 Raw data dialog).
o Raw data for ROIs (Ctrl + Shift + W) opens the Raw data assistant, where you can
select the image channels for which you want to view raw data – only for those pixels
that are inside regions of interest are then displayed in a raw data dialog.
• Process menu:
o Fix zero (Ctrl + F) subtracts the lowest value from the current image channel. For
example, if the lowest value on a topography image is 0.2 µm, this value is subtracted
from each pixel in the image.
o Remove background (Ctrl + I) allows you to subtract a polynomial fitted to the whole
image, or the area that is inside or outside regions of interest.
o Match lines
o Line fit correction
o Add function allows you to add a linear function (a plane) to the whole image channel,
or to the area inside or outside the selected ROIs.
o Threshold data allows you to replace with a constant value the parts of the image that
are either greater or lesser than a threshold. This can applied to the whole image
channel, or to the area that is inside or outside the ROIs.
o Replace allows you to replace the data inside or outside the ROIs with a linear
function. In particular, you can replace them with a constant.
79
o Transform is a submenu for basic image operations: flipping (i.e. reflection through a
vertical or horizontal line that goes through the center of the image), transposition
(reflection through diagonal ) and rotation.
o Filter is a submenu with several image processing actions for edge detection (Sobel,
Prewitt, Laplacian of Gaussian), enhancing (sharpening, unsharp mask), noise removal
(weighted median filter, Gaussian blur), derivative calculation and general convolution.
o ROI style opens a dialog for editing the style with which the
ROIs are drawn.
o Apply mask (Ctrl + K) opens the ROI mask dialog, which lets
you apply a distinct color to the area inside or outside
regions of interest.
o Shape factors for ROIs (Ctrl + Shift + U) opens the ROI
shape factors dialog with detailed information about the
geometry of each region of interest selected on the current
image. See the section J.9. ROI Shape factors dialog.
o Correct ROI plane (Ctrl + Shift + I) fits a plane the data
inside regions of interest and subtract it from the whole
image. This action is useful for background removal.
The next four actions allows you to draw new regions of interest:
o Polygon ROI switches on drawing Polygon ROIs (the same as in the toolbar)
o Rectangular ROI switches on drawing Rectangular ROI (the same as in the toolbar)
o Elliptic ROI switches on drawing Elliptic ROIs (the same as in the toolbar)
o Free hand ROI switches on drawing Free hand ROIs (the same as in the toolbar)]
The next actions allows you to perform set-theoretic operations on regions of interest:
80
The next action controls the behavior of all ROI selection modes:
o Draw ROI hole controls whether in ROIs modes the shape that you draw with the
mouse is the inside or outside the ROI. If this action is selected, the shape is treated as
a hole in a larger ROI that encompasses all image
The last two actions allow you to import or export ROI to a file.
o Export ROIs lets you export any number of ROIs to a file. Created files have the .roi
extension.
o Import ROIs lets you read in a file containing single or multiple ROIs.
The last two actions allow you to import or export profiles to a file.
o Export profiles lets you export any number of profiles to a file. Created files have the
.profile extension.
o Import profiles lets you read in single or multiple profiles from a file with the .profile
extension.
81
J.2. Creating profiles
Profiles show how the quantity visualized in an image channel changes along an straight line,
polyline or an arbitrary path, drawn with the mouse. The Image dialog lets you draw and modify
profiles of arbitrary length and position. When you draw a profile on one channel, it is
automatically drawn on the others, and the profiles are extracted from all channels at once. This
feature makes it easier to compare and find correlations between profiles from different
channels.
Profiles in the Image dialog are equipped with small, moveable triangular markers. Their
position corresponds to the position of another set of triangular markers, drawn on the profiles
itself. Markers help you to find out how exactly the features visible in the images correspond to
the features noticeable in profiles.
Height profiles are drawn in a separate Profile dialog, but they are modified in response to
actions in the Image dialog (ex. image processing), to ensure the consistency between images
and profiles. In the Image dialog, profile lines can be translated or modified, and the
corresponding height profiles are immediately updated. They are also updated when you
process the image itself, for example by removing the background.
• to extract a new profile, click Profiles (Alt + P) in the menu, then select the type of the
profile: Line profile, Polyline profile or Free hand profile. This will turn on the Extract
profile mode, in which you can draw the profile line with the mouse. Please refer to the
section G. Mouse actions on charts for more details. The Profile dialog will be displayed
(Fig. 60).
• to translate a profile, move the mouse over it, press the left mouse button and drag the
mouse. The line will follow the mouse and the cross-section will be instantly updated.
• to move a profile vertex, move the mouse over it and single click the left mouse button.
Moveable circles will appear at its ends of line profiles and nodes of the polyline/free
hand profiles . To reposition one of the endpoints or nodes, move the mouse over the
circle, press the left button and drag the mouse. To hide the circles, single – click the left
button on the line. The cross-section will be instantly updated.
• to remove or add a vertex to a polyline or free hand profile, move the mouse over its
edge and single click the left mouse button. Circles will appear at the vertices. To remove
82
a vertex, single click the left button on the circle while holding the Ctrl key down. To add
a new vertex, single click the left button on a vertex circle while holding the Shift key
down. The clicked vertex will be split into two adjacent vertices.
• to remove or add a vertex to a polyline or free hand profile, move the mouse over its
edge and single click the left mouse button. Circles will appear at the vertices. To remove
a vertex, single click the left button on the circle while holding the Ctrl key down. To add
a new vertex, single click the left button on a vertex circle while holding the Shift key
down. The clicked vertex will be split into two adjacent vertices.
• to remove a profile, move the mouse over the cross-section line and double click the left
mouse button.
• to add a new marker, move the mouse over the section line and then click the right
mouse button. In the Image dialog, a new triangular marker will be drawn on the profile
line. In the Profile dialog, a new triangular marker will be drawn in the corresponding
position on the height profile.
• to move a marker, move the mouse over the triangular marker, press the left mouse
button and drag the mouse. The marker will move along the section line.
• to remove a marker¸ move the mouse over the triangular marker, then double click the
left mouse button.
• to add a new marker, move the mouse over the plotted profile. When the mouse cursor
changes to a cross, you can add markers by clicking the left mouse button. In the Image
dialog, a new marker will be drawn in the corresponding position on the profile line.
• to move a marker, move the move over the triangular marker, press the left mouse
button and drag the mouse.
• to remove a marker¸ move the mouse over the triangular marker, then double click the
left mouse button.
83
Other image dialog action for profiles:
• to view the length, angle and position of the profile lines, go to Profiles (Alt + T) →
Profile geometry in the Image dialog. A dialog with geometrical parameters of profile
lines will be displayed.
• to edit the style of profile lines, click Profiles (Alt + T) → Profile style in the Image
dialog. The Edit profile style dialog will be displayed. For more details on how to use this
dialog, see section H. Customizing style of charts.
The Profile dialog (Fig. 60) stores charts of height profiles. The center of the dialog is occupied
by a list of image names and a tabbed panel, displaying profiles for different channels. Multiple
height profiles can be plotted on a single chart, if multiple profile lines were drawn on the
image. The Profile dialog is equipped with the main menu bar (Fig. 60 a) and a set of buttons (b)
for saving and clearing profiles. To show profile dialog from the Image dialog, use the shortcut
Ctrl + Shift + X, or go to Profiles (Alt + P) → Show profiles.
Fig. 60 Profile dialog, containing height profiles for two different section lines for each channel
of an image. Topographical profiles are visible.
84
J.3. Selecting interpolation method for height profiles
Creating a height profile requires interpolation to find the values corresponding to the positions
between the centers of pixels in the image. AtomicJ supports three interpolation methods:
You can also specify the number of points calculated for the cross-section, i.e. its resolution. For
image profiles, the default value is always 200. The choice of interpolation method influence the
visual appearance of the height profile, and the influence is greatest for low resolution images
(Fig. 62). To change the interpolation method, use the Profile interpolation dialog (Fig. 61). To
display it, use the shortcut Ctrl + I or click Interpolation (Alt + I) in the main menu bar of the
cross-section dialog, then click Change section interpolation (Alt + S).
Fig. 61. Dialog for specifying the method of interpolation and the number of points making up a height
profile. Nearest neighbor, bilinear and bicubic spline interpolation methods are supported.
The section interpolation dialog allows you to select the method of interpolation and the
number of points that make up the profile. The profiles are immediately updated to reflect the
introduced changes. The dialog is equipped with three buttons.
1. The Apply to all button (Alt + A) applies the interpolation to profiles. This button is
disabled for horizontal and vertical cross-sections.
2. The Reset button (Alt + R) discards all the changes you have made.
3. The Close button (Alt + C)
85
J.4. Editing style of profile lines
The style of profile lines can be edited with the Profile style dialog (Fig. 63). To display Profile
dialog, go to Profile (Alt + P ) → Profile style.
Fig. 62. Influence of interpolation method on the appearance of a height profile. The height profile was
cut along a diagonal of a 128 x 128 AFM topography image. A – Nearest neighbor interpolation, B –
Bilinear interpolation, C – Bicubic spline interpolation.
The Profile style dialog allows you to set different style for finished profile lines and the profile
lines that are still being drawn. Profile lines get instantly updated when you introduce changes
to their style, which makes it easier to find the optimal style parameters. You can edit the stroke
86
of the profile lines, the size of the arrowhead that shows the direction of the profile, and the
color and font of the labels.
• Batch apply (Alt + B) applies this profile line style to all charts of the same type in the
same dialog. Charts of the same type are usually grouped under the same tab in a chart
dialog.
• Save as defaults (Alt + S) saves this profile line style to a OS-dependent backing store. In
case of MS Windows, the backing store is simply the Windows Registry. From now on,
this style will be used to as the default style to draw profile lines on the charts of the
same type as the currently modified chart.
• Reset to defaults (Alt + R) resets the style of profile lines to the default style (see above).
• Undo (Alt + U) reverts the introduced changes.
Often it is necessary to restrict analysis of an AFM image to a selected region. For example, you
may need to compare distributions of some quantity in different regions of a heterogeneous
sample, or to exclude substrate from the analysis. AtomicJ allows you to select Regions of
Interest (ROIs) for this purpose. ROIs are supported by images, maps, stacks and stack cross-
sections.
You can draw any number of ROIs on a single image with the mouse. ROIs are automatically
drawn on all image channels. Four types of ROIs are available – rectangular, elliptic (including
circular), polygonal and free hand (of arbitrary shape). Before you add one, you have to choose
the type of ROI you need.- The current type of ROI determines how mouse actions on the chart
– clicks and movements – are interpreted to create a ROI. In other words, each type of ROI is
associated with its own mouse input mode – for details, see section G. Mouse actions on charts.
Different types of ROI can be easily drawn on the same image - you can always add a ROI of
other shape, just switching to different ROI type.
• to select the polygonal ROI mode, press in the toolbar icon or go to ROI (Alt + R)
→ Polygon ROI
• to select the rectangular ROI mode, press in the toolbar icon or go to ROI (Alt +
R) → Rectangular ROI
87
• to select the elliptic ROI mode, press in the toolbar icon or go to ROI (Alt + R) →
Elliptic ROI
• to select the free hand ROI mode, press in the toolbar icon or go to ROI
(Alt + R) → Free hand ROI
Once you have selected the type of ROI you need, you can draw and modify it with the mouse.
• to create a rectangular or elliptic ROI, click the left mouse button at the desired
position. A small ROI will be drawn, which can be then reshaped.
• to create a polygon ROI, click the left mouse button, move the mouse to the next vertex
and click the left mouse button there. Repeat this until all vertices are added. To finish,
click the right mouse button.
• to create a free-hand ROI, click the left mouse button and move the mouse. ROI
boundary will follow mouse movement. To finish, click the right mouse button.
• to resize a rectangular or elliptic ROI, move the mouse over its edge, press the left
mouse button and drag the mouse. Hold down the Alt key to keep the aspect ratio
constant. Hold down the Ctrl to resize the ROI from its center. Hold down the Alt gr key
(or the Alt and Ctrl at the same time) to resize the ROI from its center in an aspect-ratio
preserving manner.
• to reshape a polygon or free hand ROI, move the mouse over its edge and single click
the left mouse button. Moveable circles will appear at the vertices. To reposition one of
the them, move the mouse over the circle, press the left button and drag the mouse. To
hide the circles, single – click the left button on the line.
• to translate a ROI, move the mouse over it, press the left mouse button and drag the
mouse. The ROI will follow the mouse. Individual parts of a ROI composed of multiple
disjoint regions (ex. after merging multiple ROIs) will move independently, unless the Ctrl
key is held down.
• to remove a ROI, move the mouse over its edge and double click the left mouse button.
• to merge multiple ROIs, go to ROI → Merge ROIs. A dialog will appear, which will let you
select which ROIs should be merged and whether individual ROIs should be deleted after
merging.
88
• to subtract ROIs, go to ROI → Subtract ROIs. A dialog will appear, which will let you
select the operands and whether individual ROIs should be deleted after subtraction.
This action is unavailable if there are no overlapping ROIs drawn on the currently
displayed image.
• to create a new ROI representing the convex hull of another ROI, go to ROI → Convex
hull
You can also draw ROIs which includes whole image, except for a hole (Fig. 33 B). The shape of
the hole depends on the current ROI mode. For example, when drawing holes is switched on
and the Rectangular ROI mode is on, you can select a ROI which includes the whole image
except for a rectangle. To draw ROIs with a hole, go to ROI → Draw ROI hole.
By default, ROIs are identified with consecutive natural numbers. You can also specify a name
for the ROI. Move the mouse over ROI’s label, then double - click left mouse. A dialog will
appear, which will lets you specify a new name.
Descriptive statistics (Fig. 64) are calculated for the area inside all ROIs taken together and for
each ROI separately. For the list of statistics calculated by AtomicJ, see the subsection
Descriptive statistics in Working with the results of curve processing.
The descriptive statistics are instantly updated when you move or reshape ROI, so you can use
ROI to probe the differences between parts of the sample.
Fig. 64. Descriptive statistics calculated for image ROIs. The statistics are grouped in several tables, each
for a different image channel, under its own tab. Each table contains statistics for all ROIs taken
together) and for ROI batch separately. The statistics are instantly updated when you reshape or
translate one of the ROIs.
89
• to display descriptive statistics for ROIs, go to Data → Statistics ROIs, or use the
shortcut Ctrl + Shift + T.
Image dialog lets you generate both static and live histograms for the area inside all ROIs. Static
histograms are not updated when you translate or modify ROIs. Live histograms are instantly
updated in response to every modification to ROIs. For more information about histograms, see
section F.3. Creating histograms.
• to generate static histograms for ROIs, click the toolbar icon , or go to Data →
Draw ROI histograms, or use the shortcut Ctrl + Shift + D.
• to display histograms already generated in the current session, you click the toolbar
icon , or go to Data → Show histograms, or use the shortcut Ctrl + H.
Histograms for ROIs are stored in the same dialog as the histograms for the whole image.
• to display Live ROI histogram, you can use the shortcut Ctrl + Shift + H, or go to Data (Alt
+ R) → Live ROI.
Image dialog lets you generate both static and live box-plots for the area inside ROIs. Static box
plots are not updated when you translate or modify ROIs. Live box plots are instantly updated in
response to every modification to ROIs. For more information about box plots, see section F.5.
Creating box plots.
• to generate static box plots for ROIs, click the toolbar icon or go to Data → Draw
ROI box plots.
• to display histograms already generated in the current session, you click the toolbar
icon , or go to Data → Show box plots.
• to display Live ROI box plots go to Data (Alt + R) → Live ROI box plots.
90
J.9. ROI Shape Factors dialog
The Shape factors dialog (Fig. 65) provides detailed information about the geometry of the
selected regions of interest. It contains the values of shape factors, like area or perimeter,
calculated for each ROI. It can be accessed from all dialogs that display charts supporting regions
of interest (ex. Image Dialog and Map Dialog). To display it, go to ROI → Show ROI shape factors
or use the shortcut Ctrl + Shift + U.
Fig. 65. Dialog containing shape factors for ROIs. The shape factors are instantly updated when you
reshape or translate one of the ROIs.
• ROI – the name of the ROI. The same name is used as a ROI label in the chart.
• Area – area in square microns.
• Centroid X – the first coordinate of the geometric center
• Centroid Y – the second coordinate of the geometric center
• Perimeter
• Box width – the width of the smallest rectangle that completely encloses the roi and
whose sides are parallel to the axes
• Box height – the height of the smallest rectangle that completely encloses the roi ad
whose sides are parallel to the axes
• Feret min – the shortest Feret (or caliper) dimeter
• Feret max – the longest Feret (or caliper) dimeter
• Circularity index – a measure of circularity, calculated according to the equation:
4 A
f =
P2
91
The minimal and maximal Feret diameters are calculated with the rotating calipers algorithm
(the version of Preparata and Shamos - 1985). This algorithm requires that we first calculate the
convex hull of the ROI shape. AtomicJ calculates t with the Graham’s scan algorithm.
Regions of Interest (ROIs) are supported by images, maps, stacks and stack cross-sections. In all
cases, their style can be edited with the ROI style dialog (Fig. 66).
• to display ROI style dialog, click ROI in the main menu bar, the click ROI style in the
drop-down menu.
The ROI style dialog allows you to set different style for finished ROIs and the ROIs that are
being drawn. The ROIs get instantly updated when you introduce changes to their style, which
makes it easier to find the optimal style parameters.
92
• Batch apply (Alt + B) applies this ROI style to all charts of the same type in the same
dialog. Charts of the same type are usually grouped under the same tab in a chart dialog.
• Save as defaults (Alt + S) saves this measurement style to a OS-dependent backing store.
In case of MS Windows, the backing store is simply the Windows Registry. From now on,
this style will be used to as the default style to draw ROIs on the charts of the same type
as the currently modified chart.
• Reset to defaults (Alt + R) resets the style of ROIs to the default style (see above).
• Undo (Alt + U) reverts the introduced changes.
The Image dialog lets you modify the whole images and the selected Regions of Interest. Height
profiles, descriptive statistics and the Live Histogram are instantly updated whenever you
modify the images.
• to ensure that the values in the image starts from 0, you can either:
This option finds the lowest value in the image and subtracts it from all values in the
image.
• to correct the substrate tilt, first draw a ROI (see subsection J.5. Creating and modifying
Regions of Interest (ROIs)) around the substrate in the image (or multiple ROIs), then
either:
This options fits a plane to the inside of all selected ROIs taken together and subtracts it
from the whole image.
• to threshold the parts of image, go to Process → Threshold . The Set data thresholds
dialog (Fig. 68) will be displayed. It lets you specify whether the thresholding should be
applied to the whole image, the area outside or the area inside of ROIs, set the lower
93
and upper thresholds, and specify the values that will replace the pixels smaller than the
lower threshold and greater than the upper threshold.
Fig. 68. Dialog for thresholding images. Thresholding can be applied either to the whole image,
or to the area inside or outside the selected ROIs.
• to add or subtract a plane, go to Process → Add function. The Add function dialog (Fig.
69) will be displayed. It lets you specify whether the plane should be added to the whole
image, or to the area outside or inside the selected ROIs. Here you can also set the
parameters in plane equation.
Fig. 69. Dialog for adding (or subtracting) plane to images. The plane can be added to the whole
image, or the area inside or outsides the selected ROIs
94
Fig. 70. Dialog for removing
the polynomial background.
. Polynomial can be fitted to
the whole image, or the
area inside or outside ROIs.
Its degree can be set
independently for x and y.
• to replace parts of the image inside or outside ROIs with a place, go to Process →
Replace. The Replace data dialog (Fig. 71) will be displayed. It lts you specify whether the
plane should replace the outsides or insides of ROIs and set the plane equation
parameters
Fig. 71. Dialog for replacing insides or outsides of ROIs with a plane.
• to convolve the image with an arbitrary kernel, go to Process → Convolve. The Convolve
dialog (Fig.) will popup. It lets you specify the components of the convolution matrix and
whether it should be normalized. You can apply the convolution to the whole image, or
only to its selected parts, outside or inside ROIs.
95
N. Working with maps
The maps generated from the processed force – volume recordings are stored in the Map
dialog. It stores all maps that were generated during the current session, unless they have been
cleared. The map dialog contains a list of force – volume recordings (Fig. 72 d) that allows you to
navigate between maps for different recordings.
The map dialog pops up immediately after processing of force curves is finished, if any of the
processed batches contained a force-volume recording. It contains a central tabbed pane and
each of the generated maps is located under its own tab. Seven different maps are always are
generated:
Depending on the contact model used for force map processing, additional maps may be
generated:
The Map dialog resembles the Image Dialog. All actions that can be performed on images
can also be performed on maps. In addition, the map dialog allows recalculating of force
curves and for creating stacks.
96
K.2. Recalculating force curves
The map dialog lets you recalculate force curves in the current force-volume recording. You can
recalculate all curves, curves within ROIs or just a single curve. Thanks to this, you can use
different processing settings for different parts of a map, or to recalculate manually the force
curves that turned out to be poorly processed with the automatic method.
To recalculate all force curves, go to Curves (Alt + U) → Recalculate (Ctrl + R). A dialog will
popup, where you can specify whether to recalculate all curves, or the curves inside or outside a
selected ROI. You can also specify whether to update the old maps, or to keep them intact and
create new ones with the recalculate results. The old charts for individual curves and the old
entries in the Result dialog can be deleted or kept along the recalculated ones.
To recalculate a single curve, move the mouse over the position in a map that you want to
recalculate, then click the right mouse button. Click the Recalculate curve item in the context
menu.
Fig. 72. The Map dialog stores the maps generated in the current session. It contains a list of processed
files with force-volume recordings (d) and a central panel (c), displaying each map under its own tab.
Maps can be manipulated and analyzed through action accessible from the main menu bar (a), toolbar
(b) and a set of buttons (c).
97
K.3. Navigating to processing results
The map dialog allows you to find charts showing the force curve, force – indentation curve and
pointwise modulus – indentation curve, as well as the numerical results of processing that
correspond to the selected position on the map.
The Graphical results dialog will be shown, displaying the charts corresponding to the
clicked position, if they are accessible. If the charts cannot be found (either they
were not generated or they have been deleted), another dialog will pop up, asking
you whether you want to preview the force curve.
• to go to the processing results, the mouse over the position of interest, then click the
right mouse button. Click the Find results in the context menu. Unless the results have
been cleared, The Results dialog will be shown, with the rows corresponding to the
clicked position selected.
Force – volume recordings contain information about the changes in mechanical properties of
the sample across its depth. This information cannot be displayed in a single map, but you can
still extract this information in the form of stacks.
Stacks consist of series of frames and can be thought of as movies. Each frame is a map showing
the spatial distribution of some quantity, calculated either for a particular indentation depth
(Fig. 73 A) or tip displacement (Fig. 73 B). You can generate stacks for the following quantities:
• force that produced the current indentation or tip displacement
• pointwise modulus, calculated for the current indention depth or tip displacement
• stiffening, which is the difference between Young’s modulus calculated from the fit
to force – indentation curve and the pointwise modulus, calculated for the current
indentation depth or tip displacement.
98
Fig. 73. Stacks can be generates for a
series of indentation depths (A) or tip
displacements (B).
Often we are not interested in the whole area in the map, but only in a particular region. This is
the case when the map contains not only your sample, but surrounding substrate, ex. mica. The
map dialog allows you to create stack only for the selected ROIs.
All types of stacks can be generated through the main menu of the Map dialog:
In each case, a dialog similar to the dialog in Fig. 74 will appear. Here you can specify the
settings necessary for generating the stack:
99
o the Maximum field lets you specify the largest values of indentation/tip
displacement. It will be used for the last frame.
o the Frame no field lets you specify the number of frames n. This number determines
the step between successive frames, which equals (Maximum – Minimum)/(n - 1).
Once you specified all the parameters, click the OK button (Alt + O).
Fig. 76. Series editor (see section H.1.a.) for maps. The General tab is displayed. Here you can specify
whether tooltips should be displayed, what shape should be used as a map pixel or whether to draw
guidelines (i.e. lines that shows the grid on which the maps is based).
100
Fig. 77. Series editor (see section H.1.a.) for maps. The Label tab is displayed. Here you can specify whether
labels should be displayed and what color and font should be used. You can also set the number format
used for labels. By default if the map is too large to display all labels clearly, the labels are hidden.
Fig. 78. A zoomed-in part of a Young's modulus map with labels visible. Labels are hidden by default if the
map is so large that the labels would overlap, but you can always switch them off for a smaller, zoomed-in
area, using the Live chart style dialog.
101
L. Working with Stacks
Stacks provide an insight into three-dimensional distribution of the mechanical properties of the
sample. Stacks are series of frames, showing force, pointwise modulus or stiffening calculated
for particular values of indentation or tip displacement. They can be generated from processed
force-volume recordings, through the Map dialog (see the section J. Working with maps).
Stacks are displayed in the Stack dialog (Fig. 79), resembling a movie player. It is equipped with
a menu bar, a central panel where the stack is displayed, a slider for manual changing of frames
and a set of buttons for playing the stack as a movie. While stack as a whole can be thought of
as a movie, individual frames are charts, similar to map charts, so most of the actions that can
be performed on other charts can be also performed on stacks.
102
L.2. Saving stacks to files
You can either save a whole stack as a movie file, or individual frames as image files.
In both cases, the Save dialog appears. See section I. Saving charts for details.
• to save all frames as separate image files, go to File (Alt + F) → Save all frames (Ctrl +
Shift + S)
• to save the whole stack as an AVI movie, go to File (Alt + F) → Save as movie. The Save
dialog appear. See section I. Saving charts for details.
Stacks can be played as a movie both forward and in reverse. The user can modify the frame
rate and the movie length. The stack dialog is equipped with an easily accessible set of buttons
(Fig. 79 d) for playing the movie. Additional actions can be accessed from the Movie menu (Alt +
M in the main menu bar):
103
1. click the button at the bottom of the stack dialog
2. go to Movie (Alt + M) → Next frame.
3. use the shortcut Ctrl + Right arrow.
• to change frame rate or movie length, go to Movie (Alt + M) → Edit parameters. The
movie parameters dialog (Fig. 80) will be displayed. Here you can specify the rate in
frames per seconds (fps) or the length in seconds. These parameters are interdependent –
when you change one, the other gets updated automatically.
104
L.4. Recalculating force curves
The Stack dialog allows you to recalculate curves in the force-volume recordings from which the
stack was generated. This is similar to recalculating curves from the Map dialog. Now, however,
you can judge whether recalculation is necessary from the quality of stack. You can recalculate
all curves, curves within ROIs or a single curve. You can also use different processing settings for
different parts of the stack, or to recalculate manually the force curves that turned out to be
poorly processed with the automatic method.
• to recalculate all force curves, go to Curves (Alt + U) → Recalculate (Ctrl + R). This opens
The Processing assistant, allowing you to recalculate the force-volume recordings
corresponding to the stack. For instruction on how to process force curves, see section E.
Force curve processing. After processing is over, recalculated maps will be added to the
Map dialog. Now you can generate a new stack – for details, see section K. Working with
maps.
• to recalculate the curves within ROIs, go to Curves (Alt + U) → Recalculate ROIs (Ctrl +
Shift + R)
• to recalculate a single curve, move the mouse over the position in the stack that you
want to recalculate, then click the right mouse button. A context menu will appear. Click
the Recalculate curve item.
The Stack dialog allows you to find charts showing the force curve, force – indentation curve
and pointwise modulus – indentation curve, as well as the numerical results of processing that
correspond to the selected position on the current frame of the stack.
1. move the mouse over the position of interest, then click the right mouse button.
A context menu will appear. Click the Find graphs item.
2. move the mouse over the position of interest, then double click the left mouse
button.
The Graphical results dialog will be shown, displaying the charts corresponding to the
clicked position, if they are accessible. If the charts cannot be found (either they
105
were not generated or they have been deleted), another dialog will pop up, asking
you whether you want to preview the force curve.
• to go to the processing results, the mouse over the position of interest, then click the
right mouse button. A context menu will appear. Click the Find results item. Unless the
results have been cleared, The Results dialog (Fig. 21) will be shown, with the rows
corresponding to the clicked position selected.
Style of stacks can be customized using the same menu options and dialogs that are available
for image and map charts. You can use The Live chart style dialog, the Color gradient dialog or
the Range histogram dialog, all of which can be displayed from the Chart (Alt + C) menu in the
main menu bar – for details go to section H. Customizing style of charts. Here it is important to
note that:
1. the changes in the style of the stack are always applied to all frames, so you can modify
the whole stack with a single click.
2. stacks are updated instantly when you modify their style, even if they are being played
as a movie. This makes it easier to choose the best style, as you see at once what is the
effect of the changes you are introducing.
3. by default, the stack title gets updated when you change the frame or play the stack as
a movie. The current level of the stacking quantity (for example force) is automatically
set as the title. You can override this behavior, using the Title editor of the Live chart
dialog (Ctrl + L) (see section H.1.d.). It contains an additional tab, named Updating,
where you can also set the number format used for automatic titles.
106
L.7. Drawing histograms for a stack
Histograms can help you to analyze the distribution of the quantity shown in the stack. You can
find out how distribution changes between frames using the Live histogram dialog. The live
histogram is instantly updated when you change the frames manually or the movie is playing.
Alternatively, you can generate a histogram for a single frame:
• to draw a histogram for the current frame, go to Data → Draw frame histogram
(Ctrl + D). The histogram assistant will appear, where you can specify how the histogram
should be drawn. For details about the Histogram assistant, see section F.3. Creating
histograms.
• to view the histograms already generated for the frames, go to Data (Alt + C) → Frame
histograms (Ctrl + H)
• to draw a live histogram, go to Data → Live histogram (Ctrl + Shift + L). If the live
histogram has not been already drawn, the Histogram assistant will be displayed to
collect drawing settings.
Descriptive statistics are calculated for each frame separately and are stored in a separate
dialog. Statistics for all frames are displayed in the same table, so you can easily compare them
(Fig. 81). See section F.7. Descriptive statistics for details about the descriptive statistics
calculated by AtomicJ. To display descriptive statistics for each frame, go to Data → Statistics
(Ctrl + T).
107
L.9. Displaying raw stack data
Each frame of a stack is a visual representation of an array of numbers. You may want to view or
save this raw data, for example to analyze it statistically or to plot it using another application.
For more details on displaying raw data, see section D.2. Raw data dialog. To display raw stack
data, go to Data →Raw data (Ctrl + W).
Cross-sections can help you to analyze the stack, revealing spatial distribution of the visualized
quantity in the direction perpendicular to the sample surface. Stack cross-sections are two-
dimensional arrays of values, with rows corresponding to the frames in the stack. Those arrays
are displayed as images. You can control the interpolation method used to create the cross-
section, which is especially important if the cross-section is not cut along the pixels.
The simplest cross-sections are horizontal or vertical, cut along the lines formed by pixels in the
stack frames. You can create at once a series of such cross-section, cut at regular intervals.
Sometimes, however, you will need greater flexibility, for example to make cross-section
perpendicular the direction of filaments or fibers present in the sample. The stack dialog allows
you to create flexible cross-section cut along an arbitrary line.
The cross-sections are dynamic entities, tightly coupled to their parent stacks. User-interactive
chart elements - lines and markers, are drawn both in the stack and cross-section charts. They
help you to find out how the futures in the stack correspond to the features visible in the cross-
section.
In the cross-section chart, the following interactive elements are drawn by default:
1. a horizontal line (white dashed in Fig. 82 and 84), marking the Y-coordinate of the
current frame in the stack. You can move it with the mouse to change the frame in the
stack. It moves when you change frames in the stack itself.
2. at least one vertical line (white solid in Fig. 82 and 84), acting as a moveable marker. This
marker is coupled with a corresponding triangular marker in the stack– if you move one,
the other also moves.
In the stack, the user interactive elements are drawn when cross-sections dialogs are visible or –
in case of flexible cross-sections – when the Extract profile mode is on:
108
1. cross-section lines (white solid in Fig. 85) showing where the cross-sections were cut.
You can move these lines with the mouse - the cross-sections are then updated
accordingly.
2. triangular markers (white triangles in Fig. 85) that can be moved with the mouse along
the section lines. Their position is coupled to the position of the vertical lines in the
cross-section chart. You can delete the marker or add new markers with mouse.
• to add a new marker in the Stack dialog, move the mouse over the section line and then
click the right mouse button. In the Stack dialog, a new triangular marker will be added.
In the Cross-section dialog, a new vertical line will be drawn in the position
corresponding to the marker.
• to add a new marker in the Cross-section dialog, go to Markers (Alt + K) → Add marker
(Alt + A). If this item is selected, the mouse cursor appears as a cross over the cross-
section plot and you can add markers by clicking the left mouse button. Once you added
markers, deselect the Add marker item.
• to move a marker in the Stack dialog, move the mouse over the marker (small triangle),
press the left mouse button and drag the mouse. The marker will move along the section
line.
• to move a marker in the Cross-section dialog, move the mouse over a vertical line, press
the left mouse button and drag the mouse.
• to remove a marker in the Stack dialog¸ move the mouse over the marker (small
triangle), then double click the left mouse button.
Series of horizontal and vertical cross-sections are displayed in separate Cross-section dialogs.
You can think of a series of cross-sections as a movie. Successive cross-sections, taken at regular
intervals, are then frames. The Cross - sections dialog resembles a movie player. It is equipped
with a menu bar (Fig. 82 a), a central panel where the current cross-section is displayed (b), and
a slider together with a set of buttons for playing the series of cross-sections as a movie (c). For
details on how to play a movie, see L.3. Playing stack as a movie in the section Working with
stacks.
109
• to display a stack cross-section, go to Sections (Alt + E) in the menu bar, then Horizontal
slices (Alt + H) or Vertical slices (Alt + V). A cross-section dialog will be displayed
(horizontal shown in Fig. 82, vertical in Fig. 83).
Fig. 82. The Horizontal section dialog. The central panel displays the current frame of the cross-section
movie (b). Cross-sections can be manipulated and analyzed through action accessible from the main
menu bar (a) or played as a movie with the help of dedicated controls (c, b).
The horizontal dashed line marks the Y-coordinate of the current frame in the sectioned stack.
110
L.12. Flexible cross-sections
Flexible cross-sections are extracted along arbitrary section lines that can be drawn on any stack
frame. To generate a cross-section, you have to first draw the section line. Please consult the
G.3. Extract profile mode subsection in Mouse actions on charts. The cross-sections are
displayed in the Flexible slices dialog (Fig. 84). The dialog contains a separate tab for each
section you have made. The tab name is the section number, the same that is displayed in the
Stack dialog next to the section line (Fig. 85). If you generate a new cross-section, a new tab is
added, if you delete a section, the corresponding tab is removed.
Flexible cross-sections are dynamic. When you translate, change the angle or length of the
section line in Stack dialog, the corresponding cross-sections in the Flexible dialog are
immediately updated. Just like horizontal or vertical sections, the flexible cross-sections are
tightly coupled to their parent stacks. The same user-interactive chart elements - lines and
markers - are draw both in the stack and cross-section chart.
• to generate a new flexible cross-section, go to Sections (Alt + T) in the menu bar of the
Stack dialog, then select the Line section, Polyline section or Free hand section in the
drop-down menu. This will turn on the Extract profile mode, in which you can draw the
cross-section line with the mouse. Please refer to the section Mouse actions on charts
and stacks for more details. A cross-section dialog will be displayed (Fig. 84).
• to translate the section profile in the Stack Dialog, move the mouse over it, press the
left mouse button and drag the mouse. The line will follow the mouse and the cross-
section will be instantly updated.
• to modify the section profile, move the mouse over it and single click the left mouse
button. Moveable circles will appear at its ends of line profiles and nodes of the
polyline/free hand profiles . To reposition one of the endpoints or nodes, move the
mouse over the circle, press the left button and drag the mouse. To hide the circles,
single – click the left button on the line. The cross-section will be instantly updated.
• to remove the cross-section, move the mouse over the cross-section line and double
click the left mouse button.
111
• to view the length, angle and position of the section lines, go to Sections (Alt + T) →
Section geometry (Alt + G) in the Stack dialog. A dialog with geometrical parameters of
section lines will be displayed.
112
L.13. Selecting interpolation method for stack cross-sections
Creating a stack cross-section may require interpolation to find values corresponding to the
positions between the nodes in the grid of force-volume data. Interpolation is almost always
necessary for flexible sections. Because horizontal and vertical sections are cut along the nodes
of the force volume data (i.e. along the pixels in stack frames), interpolation is not used as long
as the number of points included in the cross-section equals the size of grid. AtomicJ supports
three interpolation methods:
• nearest-neighbor interpolation
• bilinear interpolation
• bicubic spline interpolation
You can also specify the number of points calculated for the cross-section, i.e. its resolution. For
horizontal and vertical sections the default number of points is equal to the size of the grid. For
flexible cross-sections, the default value is always 200. The choice of interpolation method
influence the visual appearance of the cross-section.
• to change the interpolation method, use the Section interpolation dialog (Fig. 86). To
display it, you can either:
1. click Interpolation (Alt + I) in the main menu bar of the cross-section dialog, then
click Change section interpolation (Alt + S) in the drop-down menu.
2. use shortcut Ctrl + I.
The section interpolation dialog (Fig. 86) allows you to select the method of interpolation and
the number of points that make up the cross-section. The images of cross-section are
immediately updated to reflect the introduced changes. The dialog is equipped with three
buttons.
1. The Apply to all button (Alt + A) applies the interpolation to all flexible cross-
sections. This button is disabled for horizontal and vertical cross-sections.
2. The Reset button (Alt + R) discards all the changes you have made.
3. The Close button (Alt + C)
113
L.14. Overlaying map profiles on stack cross-sections
Features present in the stack cross-section may be correlated with the features visible in the 2D
maps, for example sample topography (i.e. position of contact point) or Young’s modulus. To
find out if there is any correlation, you may wish to overlay the stack cross-section with map
profiles. Cross-section dialogs allows you to choose which profiles should be overlaid and which
interpolation method should be used to generate them.
114
M. Appendix
Vista, Windows 7:
Windows XP:
The relation between indentation and force depends on the shape of the tip and the thickness
of the sample. AtomicJ supports several tip shapes – Sphere, Cone, Pyramid, Power-shaped,
Hyperboloid, Power-shaped, Blunt cone, Blunt pyramid, Truncated (flat-ended) cone and
Truncated pyramid. For sphere and cone it also implements corrections for finite thickness of
the sample. All equations assume that the material behaves as linear elastic (i.e. that the
indentations are relatively small).
a b
115
Fig. 89. Tip shapes: a. Paraboloid that approximates sphere in the Hertz’s equation. R – radius of
curvature at the apex. b. Hyperboloid. R – radius of curvature at the apex, θ – half angle
between the asymptotes.
c d
e
Fig. 89 (cont.). Tip shapes. c. Cone. θ –
half angle. d. Blunt cone (cone capped
by a sphere). R- radius of curvature at
the apex, b – tip radius at the level of
transition between the capping sphere
and the cone, θ – half angle. If the
transition from sphere to cone is
smooth, then b = R Cos . e.
Truncated cone. b – truncation radius, θ
– half angle
P – load (force)
E – Young’s modulus;
ν – Poisson’s ratio
δ – depth of indentation;
a – contact radius
h – sample thickness
116
1. Sphere (Hertz). An approximation of the sphere by a paraboloid. It is accurate when the
contact radius a is much smaller than the sphere radius R, which means that √𝑅 𝛿 ≪ 𝑅.
3
4E
P= R δ 2
(
3 1 − ν2 )
2. Sphere (Sneddon). Unlike Hertz’s, Sneddon’s solution (Sneddon 1965) does not require that
√𝑅 𝛿 ≪ 𝑅.
E R2 + a2 R+a
P= 2
Log −a R
( )
1− ν 2 R−a
a R+a
δ= Log
2 R−a
4E 3
2α 4α 2 2 8 3 4π 2 3 16α 3 3π 2 4
P= R δ 1 −
2
χ + 2 χ − 3 α + βχ + 4 α + βχ
(
3 1 − ν2 ) π π π 15 π 5
R
where =
h
Coefficients α and β depends on whether the sample is adherent to the substrate or not.
For adherent sample:
3 − 2ν
α = −0.347
1− ν
5 − 2ν
β = 0.056
1− ν
117
Ea 3 2 ξ π 1 ξ R Cot θ
P= ( )
ξ + 1 − ξ + ArcTan − where ξ =
2
( 2
)
1 − ν R 2 2 2ξ 2 a
a2 π 1 ξ
δ= ξ + ArcTan −
2R 2 2ξ 2
2E Tan θ
P= δ2
(
π 1− ν 2
)
8E Tan[𝜃] 2 2Tan[𝜃]𝛿 2 2
𝛿2
𝑃= δ [1+ 𝛼 + 16𝛼 Tan[𝜃] ]
3𝜋 𝜋2ℎ ℎ2
7. Power-shaped (Galin 1946). The tip is modeled as a solid obtained by revolving a power
function along the y axis. The tip is thus axisymmetric and its profile is given by the power
function f(r) = Arλ.
2
E A 2
2 a +1
2
P=
1 − 2 + 1 ( )
2
2
A 2 −2
= 2 a
( )
where Γ is Euler’s gamma function. Hertz’s equation for sphere and Sneddon’s equation for
cone are special cases of the equation for the power-shaped tip, with λ = 2, A = 1/(2R) and λ = 1,
A = 1/Tan[θ], respectively. For example, the equation with λ = 2, A = 0.5 corresponds to the
Hertz’s model with R = 1 μm. In the AtomicJ interface, A is termed the factor and λ is termed the
exponent.
118
8. Blunt cone (Briscoe et al. 1994). The Hertz solution for the sphere is applied for small
indentations, for which the contact radius a is smaller than the transition radius b. This is the
b2
case as long as . For larger indentations:
R
a2 b a
3 1
a 2 − b2
P=
2E
aδ − − ArcSin − (
+ a 2 − b2 ) 2
b
+
(
1 − ν2 ) 2 Tan
θ 2
a 3 R 2 Tan θ 3 R
a π b a 2 2
( ) − a
1/2
δ= − ArcSin − a − b
Tan θ 2 a R
9. Truncated cone (Briscoe et al. 1994). Truncated cone can be treated as a special case of
blunted cone, for which the radius of curvature at the apex is infinite. Substituting R = ∞ into
the equations for blunt cone, we get:
a2 b
1
P=
2E
aδ − − ArcSin + a − b
2
(2
) 2
b
(
1 − ν2 ) 2 Tan
θ 2
a 2 Tan θ
a π b
δ= − ArcSin
Tan θ 2 a
1.4906 ETan[ ] 2
P= δ
2 (1 − ν 2 )
11. Blunt pyramid, regular, four-sided (Rico et al. 2005). The Hertz solution for the sphere is
applied for small indentations, for which the contact radius a is smaller than the transition
b2
radius b. This is the case as long as . For larger indentations
R
2E 21/2 a 2 b a
3 1
21/2 b a 2 − b2
P= aδ − − ArcSin − + a 2 − b2 ( ) 2
+
(
1 − ν
2
) π Tan θ 2 a 3R π Tan θ 3R
23/2 a π b a 2 2
( ) − a
1/2
δ= − ArcSin − a − b
π Tan θ 2 a R
119
12. Truncated pyramid (Rico et al. 2005). Truncated pyramid can be treated as a special case of
blunted pyramid, for which the radius of curvature at the apex is infinite. Substituting R = ∞ into
the equations for blunt pyramid, we get:
2E 21/2 a 2 b
1
21/2 b
P= aδ − − ArcSin + a − b
2 2
( ) 2
(
1 − ν
2
) π Tan θ 2 a π Tan θ
23/2 a π b
δ= − ArcSin
π Tan θ 2 a
HYPERELASTIC MODELS
13. Sphere, Fung’s hyperelastic model (Fung 1979) .The equation for force – indentation depth
relation was derived by Lin et al (2009), who found that the equation for sphere’s contact radius
a
from Hertz’z model (used here) holds for hyperelastic materials as long as 0.4 .
R
20 E0 a 5 − 15 Ra 4 + 75 R 2 a 3 a 3 − 15Ra 2
P= exp b 3
(
9 1 − ν2 ) 5 Ra − 50 R a + 125 R
2 2 3
25 R a − 125 R
2
a = R
14. Sphere, Ogden’s hyperelastic model (Ogden 1972).The equation for force – indentation
depth for single-term Ogden’s model relation was derived by Lin et al (2009):
− /2 −1 −1
40 E0 a 2 a a
P= 1 − 0.2 − 1 − 0.2
(
9 1 − ν 2 ) R R
a = R
ADHESIVE CONTACT
15. Derjaguin-Muller-Toporov (DMT) (Derjaguin et al. 1975). DMT model was derived for
indentation with a sphere (approximated by a paraboloid) in the presence of adhesion forces.
We implemented the most commonly referenced form of the DMT model, in which the
additional load due to adhesive forces equals 2πγR and is independent of indentation. This
120
formulation is due to Maugis (1992). DMT model gives good approximation of the load –
indentation relationship only for stiff samples, small values of tip radius R and small surface
energy γ.
4E R
P= 3/2 − 2 R
(
3 1− ν 2
)
16. Johnson-Kendall-Roberts (JKR) (Johnson et al. 1971). JKR model was derived for indentation
with a sphere (approximated by a paraboloid) in the presence of adhesion forces. JKR model is
a good approximation of the load – indentation relationship for soft samples, large value of
radius R and large energy of adhesion (Maugis 1992).
4Ea 3 Eγa 3
P= - 2 2π
(
3R 1-ν 2 ) 1-ν 2 ( )
a2 2 a
= −
R E
(1 − 2 )
17. Sphere, Maugis solution. Maugis (1995) derived the load – indentation relationship in the
presence of adhesion, which takes into account the true shape of a spherical tip (as opposed to
the JKR model, which uses parabolic approximation).
18. Hyperboloid, Sun-Akhremitchev-Walker (SAW). Sun et al (2004) derived the load – contact
radius and indentation – contact radius relations for indentation with a hyperboloidal tip in the
presence of adhesion.
121
𝑎 2
𝐸 𝐴 𝑎2 − 𝐴2 𝜋 (𝐴) − 1 2𝑎3 𝛾𝜋(1 − 𝜈 2 )
𝑃= ( 𝑎𝐴 + ( + ArcSin [ ]) − 2√ )
1 − 𝜈2 𝑅 2 2 𝑎 2 𝐸
(𝐴) + 1
( )
𝑎 2
𝑎𝐴 𝜋 (𝐴) − 1 2𝜋𝑎𝛾(1 − 𝜈 2 )
𝛿= ( + ArcSin [ ]) − √
2𝑅 2 𝑎 2 𝐸
(𝐴) + 1
where 𝐴 = 𝑅Cot[𝜃]
We have calculated the following piecewise approximations of load required to indent a thin
layer during development of AtomicJ, using numerical solutions of the Lebedev – Ufliand type
integral equations (Lebedev and Ufliand 1958, Pupyrev and Ufliand 1960), combined with the
asymptotic solutions for very thin layers (Jaffar 1989, Yang 2003). The relative error due to the
approximation of integral equation solutions is below 0.1%, so these piecewise equations can be
regarded as exact for practical purposes, as long as other assumptions, e.g. about the linear
elasticity of the sample, are satisfied.
19. Cone Lebedev - Chebyshev: Equations for conical tips feature a dimensionless indentation
depth, defined as
𝛿 tan 𝜃
𝜘=
ℎ
2E Tan[𝜃] 2
𝑃= δ (1 + 0.461𝜘 + 0.346ϰ2 + 0.0484ϰ3 ) 𝜘 ≤1
𝜋 (1 − 𝜈 2 )
E ℎ2
𝑃= (0.0859 + 0.103𝜘 − 0.0647ϰ2 + 1.057ϰ3 ) 1 < 𝜘 ≤ 5
Tan[𝜃] (1 − 𝜈 2 )
𝜋E Tan2 [𝜃] 3
𝑃= δ 5 < 𝜘
3 (1 − 𝜈 2 )ℎ
For a bonded (adherent) incompressible (ν = 0.5) layer, we get
2E Tan[𝜃] 2
𝑃= δ (1 + 0.715𝜘 + 0.609ϰ2 + 0.735ϰ3 ) 𝜘 ≤ 0.9
𝜋 (1 − 𝜈 2 )
E ℎ2 2 3
𝜋 3 5 5
𝑃= (−0.265 + 1.225𝜘 − 1.651ϰ + 2.332ϰ + ( ) ϰ ) 0.9 < 𝜘 ≤ 89
Tan[𝜃] (1 − 𝜈 2 ) 80 2
𝜋 3 5 E Tan4 [𝜃] 5
𝑃= ( ) δ 89 < 𝜘
80 2 (1 − 𝜈 2 )ℎ3
122
20. Paraboloid Lebedev - Chebyshev: Equations for conical tips feature a dimensionless
indentation depth, defined as
𝛿𝑅
𝜘=
ℎ2
4E √𝑅 3 1
𝑃= δ2 (1 + 0.722 𝜘 2 + 0.822 ϰ) 𝜘 ≤ 0.5
3(1 − 𝜈 2 )
E ℎ3
𝑃= (−0.0633 + 0.260 𝜘 + π ϰ2 ) 0.5 < 𝜘 ≤ 450
𝑅(1 − 𝜈 2 )
𝜋ER
𝑃= δ2 450 < 𝜘
(1 − 𝜈 2 )ℎ
4E √𝑅 3 1 3
𝑃= 2
δ2 (1 + 1.105 𝜘 2 + 1.607 ϰ + 1.602 𝜘 2 ) 𝜘 ≤ 0.4
3(1 − 𝜈 )
E ℎ3 1 3 𝜋 3
𝑃= (0.616 − 3.114 𝜘 2 + 6.693 𝜘 − 7.170 𝜘 2 + 8.228 ϰ2 + ϰ ) 0.4 < 𝜘 ≤ 5400
𝑅(1 − 𝜈 2) 2
𝜋E 𝑅 2
𝑃= δ3 5400 < 𝜘
2(1 − 𝜈 2 )ℎ3
For a bonded compressible layer, the approximate equations for load are more complex, as in
this case the effect of the rigid substrate strongly depends on Poisson’s ratio of the sample.
AtomicJ also implements piecewise approximations of numerical solutions of integral equation
representations of the problem of indentation of a compressible thin layer with a conical or
paraboloidal tip, which can be found in the source code.
References:
123
5. Dimitriadis EK, Horkay F, Maresca J, Kachar B, Chadwick RS (2002) Determination of elastic
moduli of thin layers of soft material using the atomic force microscope. Biophys J 82:2798–
2810.
6. Fung YC, Fronek K, Patitucci P (1979) Pseudoelasticity of arteries and the choice of its
mathematical expression. Am J Physiol-Heart C 237:H620–H631
7. Galin LA (1946) Spatial contact problems of the theory of elasticity for punches of circular shape
in planar projection. J. Appl. Math. Mech. (PMM), 10: 425–448.
8. Gavara N and Chadwick RS (2012) Determination of the elastic moduli of thin samples and
adherent cells using conical AFM tips. Nat Nanotechnol. 7: 733–736.
9. Harding JW, Sneddon IN (1945) The elastic stresses produced by the indentation of the plane
surface of a semi-infinite elastic solid by a rigid punch. Proc Camb Philol Soc. 41:16.
10. Johnson KL, Kendall K, Roberts AD (1971) Surface energy and the contact of elastic solids. Proc.
R. Soc. Lond. A 324: 301-312.
11. Lebedev N, Ufliand I (1958). Axisymmetric contact problem for an elastic layer. Journal of
Applied Mathematics and Mechanics 22: 442:450.
12. Lin DC, Schreiber DI, Dimitriadis EK, Horkay F (2009) Spherical indentation of soft matter beyond
the Hertzian regime: numerical and experimental validation of hyperelastic models. Biomech
Model Mechanobiol 8:345–358.
13. Maugis D (1992) Adhesion of spheres: The JKR-DMT transition using a Dugdale model. J Colloid
Interface Sci 150: 243 – 269.
14. Maugis D (1995) Extension of the Johnson-Kendall-Roberts theory of the elastic contact of
spheres to large contact radii. Langmuir 11: 679 – 682.
15. Maugis D (1995) Extension of the Johnson-Kendall-Roberts theory of the elastic contact of
spheres to large contact radii. Langmuir 11: 679 – 682.
16. Ogden RW (1972) Large deformation isotropic elasticity—on the correlation of theory and
experiment for incompressible rubberlike solids. Proc R Soc Lond A Math Phys Sci 326: 565–584.
17. Rico F, Roca-Cusachs P, Gavara N, Farre R, Rotger M and Navajas D (2005) Probing Mechanical
Properties of Living Cells by Atomic Force Microscopy With Blunted Pyramidal Cantilever Tips.
Phys Rev E 72, 021914.
18. Sneddon IN (1965) The relation between load and penetration in the axisymmetric Boussinesq
problem for a punch of arbitrary profile. Int J Eng Sci 3: 47 - 57.
19. Sun Y, Akhremitchev B, Walker G (2004) Using the adhesive interaction between atomic force
microscopy tips ad polymer surfaces to measure the elastic modulus of compliant samples.
Langmuir 20: 5837 – 5845.
20. Pupyrev VA, Ufliand IS (1960) Certain Contact Problems for an Elastic Layer, J Appl Math Mech
24: 1015 – 1026.
21. Jaffar MJ (1989), Asymptotic Behaviour of Thin Elastic Layers Bonded and Unbonded to a Rigid
Foundation, Int J Mech Sci 31: 229 – 235.
22. Yang F (2003), Axisymmetric indentation of an incompressible elastic thin film, J Phys D Appl
Phys 36: 50 – 55.
124
AtomicJ supports plain text files, as well as files generated by microscopes from Agilent
Technologies, Bruker, JPK Instruments, Asylum Research, Park Systems, NT-MDT, Nanosurf and
AFMWorkshop. It also supports some file formats generated by the Nanotec Electronica WSxM
software.
If you use another microscope, please send me a few sample files generated by your
microscope, ideally both in binary and text format ([email protected]). If possible,
please attach print-screens showing how the files are displayed by the software that operates
your AFM. I will then add support for this type of files.
You can also convert your files to plain text (ASCII) files. AtomicJ can open comma-separated
(.csv) and text separated (.tsv) files. Their structure is very simple:
• Plain text files for force curves consist of two columns of numbers, separated by commas
(.csv) or tabs (.tsv) (Fig. 90). The left column contains values of scanner position (or
alternatively cantilever position), while the right on the corresponding deflections or
forces. Each column begins with the name of the quantity and a SI unit in brackets. You
will find curves saved in this format among the files available at
https://sourceforge.net/projects/jrobust/files/TestFiles/AFM Recordings.zip.
Fig. 90. A force curve saved in plain text format (.csv), here viewed in MS Excel. The file consists of two
columns - the left contains values of cantilver/scanner position, the right one the corresponding
deflections or forces.
• Plain text files for images (Fig. 91) should consist of a header and data for one or more
image channels. Header should contain eight fields:
125
1. Type – obligatory field, it should be always Image
2. XQuantity – the quantity on X-axis. Usually it is length in microns. The unit should
be written in brackets and only SI units will be recognized.
3. YQuantity – the quantity on Y-axis. Usually it is length in microns. The unit should
be written in brackets and only SI units will be recognized.
4. XCount- the number of pixels along the X-axis (i.e. the number of column in the
image).
5. YCount- the number of pixels along the Y-axis (i.e. the number of rows in the
image).
6. XLength – the length of the image along the X-axis.
7. YLength – the length of the image along the Y-axis.
8. Data – obligatory field, marking the header and beginning of the channel data
After the header you should specify channels, one by one. For each channel, you have
to include :
An image saved in plain text (CSV) format is included the among the files available at
https://sourceforge.net/projects/jrobust/files/TestFiles/AFM Recordings.zip.
Fig. 91. An image saved in plain text format (CSV), here viewed in MS Excel. The file consists of a header,
where the image dimensions and quantities on axes are specified, and data for channels.
126