 |
 |
|
| Bulk Builder | Magnetic Tunnel Junction Builder |
Table of Contents
The Atomic Manipulator is a general tool used to
modify system configurations in existing NanoLanguage script
define the geometry of new molecules, bulks, and two-probe systems
Three separate cases for using the Atomic Manipulator will be described in detail in the following sections:
Like most tools in Virtual NanoLab (VNL), the Atomic Manipulator is opened either by double-clicking its icon, or by dropping a configuration on the tool icon, either in form of a NanoLanguage script or a drag-and-drop from another tool. Configurations can also be dropped on an open Atomic Manipulator. In this case, the behavior differs depending on the type of the dropped configuration, and the system type presently contained in the tool:
When dropping a molecule on a Atomic Manipulator that contains a two-probe system, or vice versa “dropping a two-probe on a molecule”, the dropped configuration is added to the system. If a two-probe was dropped, the system is changed to two-probe, with the preexisting molecule added to the central region.
If a nanotube is dropped on a Atomic Manipulator that contains a two-probe system, the nanotube is added to the central region.
In all other cases, the dropped configuration will be ignored (because it makes no sense to combine the two system types). Otherwise, the present contents of the tool would be destroyed.
Some concepts mentioned here, like “central region”, will be defined and explained in the following sections.
When you are finished setting up your system, you can either
save the configuration to the file system using Save or Save As, or
drag-and-drop from the icon
at the bottom of the window directly to another tool to
transfer the configuration.
Note, when NanoLanguage scripts are dropped on the Atomic Manipulator, only the system configuration is imported; when you click the button Save, the original file will not be overwritten. The Atomic Manipulator can also be cleared and put in an unused state by left-clicking the New button.
![[Important]](images/icons/important.png){kind=icon}
|
Important |
|---|---|
|
When saving the configuration as a NanoLanguage script, information about bonds and surfaces is lost. Since this information is usually not crucial in any of the subsequent tools, it is usually not a problem even though some features in VNL will not work properly. If you want to keep this information, you should instead save the configuration as a VNL file using Save As. |
In addition to creating new molecules and two probes using existing molecules as templates, the Atomic Manipulator can also be used to build molecules from scratch by inserting atoms in the relevant positions.
|
|
Important |
|---|---|
|
We strongly suggest that you use the Molecular Builder tool for constructing and defining new molecules. |
For complicated molecules, it is often convenient to define sub-parts, such as side-groups or rings, of the system, manage the internal coordinates of each part separately, and finally put them all together (by rotating and translating the parts) without distorting the internal configuration. This work flow is implemented in the Atomic Manipulator.
For generality, the sub-parts in the Atomic Manipulator are referred to as “molecules”. When the configuration finally is created, the entire system consists of one single composite molecule. Molecules are inserted or deleted by
right-clicking the light-gray area either to the left (see Figure 28) or in the preview window.
Then select Insert or
Delete depending on the operation you desire.
Each new molecule is by default given the name “New Molecule” displayed
in the left panel. To rename the molecule,
right-click the light-gray area of the molecule and
select Rename.
Atoms are inserted
into the molecules by right-clicking the white area
belonging to the molecule in the left panel, and choosing Insert
Atom from the context menu. To delete
an atom, right-click on the corresponding line and select
Delete Atom from the context menu. Hydrogen is the
default element for new atoms; to change to another
element, simply choose a different element from the drop-down list. Similar to the
functionality of the Nanoscope, by
right-clicking an atom and choosing
Properties from the context menu, you can change the
configuration parameters of the
displayed plot.
The final positions of the atoms in a molecule (this also includes molecules that are part of the central region in a two-probe system) depend on the origin of the molecule and the internal (relative) coordinates of the atoms with respect to this origin, as well as the orientation of the molecule.
The internal coordinates of an atom are entered in the table in the left panel, and should be given in units of Ångström (1 Å = 10-10 m). By default, new atoms are inserted at the origin of the molecule.
The simplest way to specify the origin is to type in the desired position in the three Origin fields in the left panel. Sometimes, however, it is easier to specify the origin implicitly by translating the molecule or explicitly specifying the position of a certain molecule constituent. Both operations are possible and may be executed by
right-clicking any atom of the molecule in the Preview window
choosing Translate from the context menu.
The orientation
of the molecule, i.e. how the
internal coordinate axes of the molecule are aligned relative to the global
coordinate axes (shown in the Preview window) can in
principle also be entered directly in the left panel. It is, however, rather
difficult to get an intuitive understanding for how the three orientation angles
(these are entered in the three fields under Orientation) are
related to the actual orientation, since they correspond to three successive
rotations (which are carried out in the order Y, X, and Z). For the same reason, it
really only is useful to type in the orientation directly if the rotation angles are
very simple, such as a 90 degrees rotation about the X axis. In more complicated
cases, it is much easier to rotate the
molecule into place using the Rotation tool from
the context menu.
We recommended that you always align the overall molecular geometry symmetrically with respect to the coordinate axes. If not, it becomes very complicated to plot quantities such as the electron density as a contour plane in the Nanoscope, since this plot requires the surface normal to be specified. In turn, the surface normal should be related to the molecular symmetry directions for the plot to be simple to interpret.
The view of the molecule in the Preview window to the right can be zoomed, rotated, and panned using the mouse in the same way as in other parts of the program. Note, however, that in order to rotate, zoom, or pan the camera, it is necessary to click on an empty part of the preview area that does not contain any atoms (i.e. the background). If you instead click on an atom and then move the mouse, the camera will be fixed while the corresponding molecule is translated or rotated. This can be used to position a group of atoms properly, but be careful not to destroy any alignment already set up, as it is almost impossible to obtain any accuracy in these operations.
A third method for positioning and orienting a molecule is to operate with the mouse in the graphical Preview window:
Use the middle mouse button to grab the molecule or hold down the Shift key.
Grab with the left mouse button and move the molecule to the desired position.
Grab the molecule with the left mouse button to rotate it.
If you miss the molecule and grab the background, the camera is panned or rotated instead. Since the system is three-dimensional, it requires some practice and patience to move and rotate a molecule into place.
The separator between the Preview window and the table of the
atomic coordinates can be moved to adjust the space shared by the Preview and the
table. The background color and other options
regarding the appearance of the Preview
window can be modified by right-clicking the
background and selecting Properties from the context
menu.
Click the button Save or Save As to save the configuration on the disk.
Dropping a bulk configuration on a closed Atomic Manipulator gives you the possibility of editing certain parameters. Exactly which parameters that can be edited depends on the fixed crystal symmetry of the configuration. More specifically, once created, a bulk configuration always belongs to the same Strukturbericht (see the Crystal Cupboard for more details). This implies that the lattice symmetry, the number of atoms in the primitive cell, as well as the position of cell atoms relative to the lattice vectors, are all fixed.
You are, however, free to change the lattice constants (the lengths a, b, c, and the angles α, β, and γ) on the Lattice tab, to the extent that they are not fixed by the lattice symmetry. In other situations, it is more convenient to edit the b/a and c/a ratios; the values of b and c are automatically updated when these fields are changed, and vice versa. Note that the only way to change the lattice parameter a is to edit the corresponding field itself. In addition, the Atomic Manipulator also displays the unit cell volume, and the lattice unit vectors A, B, and C.
Figure 29: The prototype PtS (Strukturbericht B17) found in the Crystal Cupboard has tetragonal symmetry. As a result, only the lattice constants a and c can be changed in the Atomic Manipulator window.
Also, the chemical element of the unique atoms in the basis can freely be chosen by clicking the Basis tab. Strukturberichts typically have several atoms in the basis, but only a few of them are in fact unique; the remaining ones can be generated from the unique ones by the space group symmetry operations. The positions of the atoms are fixed relative to the lattice vectors. Consequently, the table displays the positions of the atoms in Cartesian coordinates, thereby taking the actual values of the lattice parameters into account.
Figure 30: Even though the primitive unit cell of the prototype (PtS) (see Figure 29) contains a total of four atoms, only two unique atoms can have their chemical elements changed.
If the dropped “bulk” configuration instead is a nanotube, there are no editable lattice parameters, since the period length in the tube direction is given by the chirality (and the bond length). In other words, no additional degrees of freedom exists for a one-dimensional configuration. Still, in order to perform nanotube calculations, it is necessary to assign a three-dimensional periodicity to the tube. For nanotubes, it is possible to define a padding factor that may be set in the Atomic Manipulator. If the padding factor is zero, the unit cell in the transverse plane will be defined with the restriction that no direct matrix elements extend outside the unit cell. By assigning a positive value to the padding factor, the cell is extended further, while a negative value (which is not recommended) decreases the cell size. If the cell is too small, you will typically see a degeneracy breaking in the energy spectrum at symmetry points on the edge of the Brillouin zone. These are due to electrostatic interactions extending outside the unit cell.
It is also possible to edit the chemical element in the atomic basis of nanotubes, making it easy to generate, for example, Si or Au nanotubes.
The name of the configuration is shown in the upper left corner, and will be used
as the suggested file name for storing the configuration NanoLanguage script when
pressing Save As. The name can be changed by
right-clicking the left panel and choosing
Rename from the context menu.
On the context menu, you also find an option to cleave the configuration. This is used to turn the crystal into an electrode that can be used in defining a two-probe setup.
The Preview window to the right works exactly like the Nanoscope, except that no plots can be inserted or removed. You can therefore
The default view includes the primitive unit cell, spanned by the primitive lattice vectors, the atoms in the basis, and the coordinate axes. Using the usual mouse commands, the view can be rotated, zoomed, and panned. In addition, the preview can also be exported or printed.
The general definition of a two-probe system is a geometry where two semi-infinite periodic systems, or electrodes, are combined with a finite scattering region, called the central region. The interpretation of these three parts of the two-probe system are however very broad. The electrodes can be one-dimensional systems (atomic chains, nanotubes), or bulk materials (gold, silicon). The scattering region, in turn, can be nothing (a vacuum gap), a piece of a nanotube or conducting polymer, a molecule, or a slice of a bulk material.
The electrodes are created by cleaving a crystal or a nanotube, and the cleaved surfaces are then combined with the central region to form the two-probe system. VNL automatically takes care of most of the setup of the electrodes. Additional fine tuning, such as the positioning, rotation, and translation of molecules in the central region must be handled by the user.
![Finished setup of a two-probe system consisting of two 3x3 Au [111] electrodes with a DTB (dithiol-benzene) molecule positioned in the central region.](Figures/am2p.png)
Figure 31: Finished setup of a two-probe system consisting of two 3x3 Au [111] electrodes with a DTB (dithiol-benzene) molecule positioned in the central region.
When a two-probe system is present in the Atomic Manipulator (after dropping an existing two-probe on the tool or when a bulk is cleaved and then turned into a two-probe), the Atomic Manipulator window is extended with a Surface and Two-Probe tab. These are discussed in the subsequent two sections:
Two-probe parameters (the parameters of the central region)
The Preview window to the right works exactly like the Nanoscope except that no plots can be inserted or removed. A particularly useful feature when setting up two-probe systems is the possibility to hide separate parts of a plot, such as the bonds, the left surface region, the right electrode region. This will give you a better view of the particular parts of the system. Here is how to achieve these effects
Right-click the part you would like to hide, choose
Properties from the context menu, and deselect the
Visible option.
The default view includes the two electrodes and their units cells, the atoms in the central region, and the coordinate axes. The view supports the same functionality as the Nanoscope tool.
When the setup is complete, click the button Save or Save As to create a new two-probe configuration and save it on the disk. You also have the possibility of converting the two-probe system to a bulk by left-clicking Save Equivalent Bulk which then will create an equivalent bulk configuration and save it on the disk.
![[Note]](images/icons/note.png){kind=icon}
|
See also |
|---|---|
When a periodic system, which can be either a bulk crystal or a nanotube, is open
in the Atomic Manipulator,
right-clicking either the left panel or the 3D preview
window will display an option to Cleave the bulk. When
the bulk is cleaved, it is transformed from a periodic system to a two-probe system
without any central region. The result is two separate surfaces. Two new tabs, called
Two-Probe and Surface appear in the Atomic
Manipulator; in this section we will focus on the surface parameters, whereas the two-probe parameters are
discussed separately.
Nanotube electrodes are described at the end of this section; for now, we focus on bulk crystals.
|
|
Note |
|---|---|
|
For some combinations of Miller indices and surface vectors, the surface cell may contain a large number of atoms. In this case, the Preview window will not update automatically, for example when surface vectors are edited. The purpose of this is to avoid performance slow-downs. To update the preview manually, press the Show button. |
A general surface in a crystal system is specified by three integer numbers, called the Miller indices. Conventionally, these are denoted h, k, and l. The three numbers are entered in the corresponding boxes at the top of the Surface tab which appears when the bulk has been cleaved.
The other important quantity to specify when setting up a surface is the two-dimensional surface cell. Generally, the surface cell is a two-dimensional unit cell for the surface plane, and is spanned by two surface unit vectors (called S1 and S2). Combining the surface cell with a unit vector in the transport direction SC, we obtain the three-dimensional electrode cell. The electrode cells are visualized as boxes around the atoms of the left and right electrodes, respectively. The part of each electrode that is contained in these boxes will be repeated periodically along the SC direction (which is always parallel to the Cartesian Z axis) to form semi-infinite electrodes.
It is important to recall that VNL uses periodic boundary conditions in the directions perpendicular to the transport direction in a two-probe system[1]. Consequently, any two-probe system effectively consists of two infinite surfaces linked together by an infinite number of central regions.
If the electrodes are true surfaces (as opposed to e.g. nanotubes), the surface cell determines the distance between points on the surface where the molecules in the central region can be attached. If the distances between these points are smaller than the interaction range between the molecules, the repeated copies of the central region will interact. This interaction, which may be both direct (basis set overlap) and electrostatic, is not desired if the purpose is to study the properties of individual molecules in the central region. In each case, it is therefore necessary to verify that results are independent of the surface cell size. This convergence process can be quite time-consuming; in practice, however, it is often possible to make a qualified guess and then make the cell a bit larger.
Of course, it is also possible to exploit the boundary conditions in order to set up and study an interface or the effect of an oxide layer between two metallic bulk contacts. An example of this is discussed in the ATK manual, but the corresponding system can also be set up in VNL.
![Defining a 3x3 gold (fcc) [111] surface in the Atomic Manipulator: First import a gold bulk configuration, cleave it, and specify the desired Miller indices as well as the 3x3 surface cell. For improved visibility, the surface atoms, the right electrode, and bonds have been removed from the plot.](Figures/amsurf.png)
Figure 32: Defining a 3x3 gold (fcc) [111] surface in the Atomic Manipulator: First import a gold bulk configuration, cleave it, and specify the desired Miller indices as well as the 3x3 surface cell. For improved visibility, the surface atoms, the right electrode, and bonds have been removed from the plot.
Once the Miller indices have been specified, VNL will automatically attempt to determine the smallest possible surface unit cell, specified in terms of the primitive surface vectors SA and SB. A slight complication arises from the numerical requirement that the electrodes must be periodic in the transport direction (the transport direction is always along the Z-axis). Therefore, for some Miller indices and crystal types, it is required that the surface cell contains more than one layer. Each atom with a unique Z coordinate defines a layer. A typical example is an fcc crystal cleaved along the [111] plane. In this case, the unit cell contains 3 atoms, corresponding to the stacking sequence ABC.
Note that the transport direction SC does not necessarily coincide with the C direction of the corresponding bulk system which, in turn, does not need to be parallel to the Cartesian Z axis. However, when considering the surface properties, the coordinate system is always oriented such that the Z axis is parallel to the SC direction, which is perpendicular to the cleaved surface.
By default, the surface vectors S1 and S2 (shown in the Preview) are set identical to the primitive surface vectors SA and SB. You can, however, specify S1 and S2 as an arbitrary linear combination of SA and SB, and thereby control both the shape and the size of the surface cell. For example, if you wish to have a rectangular surface cell in the [111] fcc case, set S1=2*SA+2B and S2=-SB.
Also remember, if any off-diagonal elements are introduced in the surface cell matrix (i.e. if S1 has contributions from SB, and S2 from SA), the calculations of the electrodes will become much slower. Therefore, unless there is particular reason to do otherwise, it is strongly recommended always to set S1 (S2) to an integer multiple of SA (SB). This is due to the fact that the underlying algorithms are designed to take advantage of the supercell construction; for more information, please refer to the ATK manual.
The simplest way to check if the surface cell is large enough is to save the two-probe configuration to a NanoLanguage script and drop it on the Nanoscope. When repeated along the SA/SB directions, the central regions should be well separated. For generality, the respective directions SA and SB are denoted as A and B in the Nanoscope.
The above discussion assumes that the central region consists of molecules (including pieces of nanotubes and linear chains). Another option is to have an interface two-probe system; in that case, the surface cell should simply match the interface structure.
In principle, the depth of the electrode cells can be specified arbitrarily in terms of the number of layers in the direction perpendicular to the surface plane, as long as you make sure to fulfill the two basic requirements for electrodes:
There should be no interactions between next-nearest neighbor cells.
The cell should be periodic in the transport direction.
The numerical models used in VNL assumes that the atoms inside the electrode cell only interact with atoms in the nearest repeated cell; all other interactions are simply neglected. Therefore, in order not to neglect non-zero interactions by mistake, there is a minimum length of the electrode cell in the SC direction.
There is therefore a minimal number of layers of the electrode that should be used to make sure all overlaps are accounted for. Continuing on the fcc [111] example, the minimal surface cell is 3 layers deep. Adding additional layers to fulfill the overlap requirement must be in increments of 3 (3,6,9,…) in order to meet the periodicity requirement stated above. For the opposite case, see the Figure 33.
Figure 33: For an atomic chain, the periodicity requirement is trivially fulfilled with just a single atom (layer) in the electrode cell. However, for Li, at least two atoms are needed to account for all first order interactions. To illustrate the electrode cell structure, all surface atoms have been hidden.
If the number of layers is different from 3,6,9,… for the [111] fcc case, you are forced to introduce a stacking fault, implying that periodicity of the electrode cell is broken. This may lead to incorrect results in the calculations, and should as far as possible be avoided. There are, however, some combinations of crystal symmetries and Miller indices (in particular in non-cubic systems) where it is impossible to find a proper surface cell which does not introduce a stacking fault. In those cases, either consider a different set of Miller indices, or set the number of layers by hand (the larger the better, usually), and hope for the best. VNL will still be able to handle the system, but be careful when interpreting the results.
By default, VNL suggests a number of layers that generates the smallest cell fulfilling the two basic requirements mentioned above. Usually, there is no reason to change this value.
To be precise, there are two types of interactions, so-called first and second order interactions or matrix elements. While it is crucial to include the first order interactions, second order interactions can often be neglected without compromising the accuracy of the calculation. This is helpful, since the cell may become very large if all second order interactions are included. This, in turn, causes the calculations to take more time with only minor influence on the results. The default number of layers suggested by VNL is therefore computed to keep the cell as small as possible while still making sure that only first order interactions are included. A note about neglecting second order matrix elements will therefore always appear by default. If the cell is made so small (by the user) that first order interactions also are neglected, a warning about too shallow electrodes appears.
The range of the interaction depends on the extension of the basis functions and pseudo-potentials of the atoms. Since the parameters for these quantities are unknown at the time when the geometry is defined in the Atomic Manipulator, default basis set parameters and pseudo-potentials will be used. Expert users may wish to modify these parameters later (in the NanoLanguage Scripter): In this case, carefully check that your assumptions regarding the interactions still are valid.
When using nanotubes as electrodes, it is not possible to specify Miller indices for the surface, since there is only one way to cleave a nanotube (viz. to cut it in half). There is also no way to specify the surface unit vectors, since they are automatically given by the original unit cell in the transverse plane. The size of this cell can, however, be changed both before and after cleaving, by using the same field called Padding factor. Here, a positive value makes the cell bigger, whereas a negative makes it smaller (default is zero). For more details, see the section on editing nanotubes in the Atomic Manipulator.
The depth of the nanotube electrode unit cell is specified in terms of the number of nanotube periods, in order to maintain the periodicity mentioned above. The default number of repetitions is chosen in the same way as for bulk electrodes (see above) to make sure all first order matrix elements are accounted for.
The central region of a two-probe system consists of all the atoms not included in the electrode cells. In general, these atoms can be separated into three groups: left surface atoms, right surface atoms, and the rest. The surface atoms are automatically inserted by VNL according to the desired number of surface layers, while the remaining atoms are molecules or nanotube segments inserted and positioned by the user.
It is not possible to include a bulk crystal in the central region. To set up an interface, it is therefore necessary first to manually convert the corresponding crystal supercell into a molecular configuration. When using this approach to define interface structures, it is crucial to make sure that the supercell is commensurate with the electrode surface cell; otherwise, there will be unphysical overlaps of atoms or “holes” in the structure when the entire two-probe system is repeated along the surface of the electrodes.
When cleaving a bulk, the Atomic Manipulator automatically adds several “electrode atoms”; these, however, are not contained in the electrode cell. These atoms are in reality part of the central region, and are so-called surface atoms. The number of surface atoms is controlled by the Surface Layers parameters, which can be set separately for the left and right electrodes.
![For an [111] gold surface, the default values of two surface layers results in two Au atoms positioned on both the left and right surface.](Figures/amtwoprobe.png)
Figure 35: For an [111] gold surface, the default values of two surface layers results in two Au atoms positioned on both the left and right surface.
There is no strict rule for choosing the right number of surface layers but a good guess (at least for metallic electrodes) is to use values that give a surface depth comparable to the screening length.
To understand the “screening” concept, let us consider two alternative ways of viewing a two-probe system. We can either view it as a molecule “probed” by the addition of the electrodes, or consider the electrodes as being perturbed by the addition of the molecule. In fact, the two views are equivalent and mostly a matter of taste and interpretation. From a numerical perspective, however, VNL relies on the second version in the calculations, which is reflected in the fact that the electrodes are assumed to be unperturbed (i.e. identical to the bulk material from which they were once defined) a certain distance away from the central region.
Close to the central region, the electrodes are modified by the presence of the molecule and the “gap” in the electrode material. The number of surface layers specify the extension of this perturbation region, i.e. the part of the electrodes in which the electron density may differ from its bulk equivalent. If this region is not long enough, implying that the influence of the central region in reality extends deeper into the electrodes, the central region will not be completely screened. As a consequence, the results will not be accurate.
The recommended approach is to choose a moderate value for the number of surface layers, and then increase this systematically until your results converge.
Another parameter on the two-probe panel is the central
region width. This parameter can be adjusted by
dragging the right electrode (the one which is not placed at the origin of the
coordinate system) using either the middle mouse
button or the left mouse button while holding down
the Shift. Alternatively, a new value for the central region
width may be entered in the left panel, or you may use the
Translate command from the context menu to move the electrode in a controlled
fashion.
In essence, the central region width describes the empty space that should be left in-between the central region molecule and the electrodes. This width is equivalent to the distance between the two last electrode atoms facing the central region.
To insert molecules in the central region,
simply drop a molecular configuration on the open Atomic Manipulator window.
New molecules can also be added manually while working on a two-probe system by
right-clicking the left panel and choosing
Insert new molecule entry from the context menu.
Working with molecules in the two-probe configuration of the
Atomic Manipulator is completely analogous
to creating molecules when it comes to
positioning and orienting the molecule.
It is also possible to insert a nanotube by dropping a nanotube configuration on an open Atomic Manipulator window that already contains a two-probe system. This, for example, makes it possible to study carbon nanotubes positioned between metal surfaces.
Figure 36: Unfinished setup of a two-probe system with a (4,4) carbon nanotube in the central region. Like all armchair tubes, the (4,4) nanotube has a very short period length. In order for the tube to form a reasonable central region, it has been repeated several times along the tube axis.
A nanotube imported in this way can be translated just like a molecule, and is also given an origin enabling it to be positioned properly. On the contrary, the tube cannot be rotated, and therefore does not have any orientation. Finally, the tube can be repeated along the tube axis as shown in Figure 36.
By selecting Translate from the context menu
(right-clicking an atom in the
Preview window of the Atomic
Manipulator), you may either
translate a molecule (which may be a group in a real molecule or part of the central region in a two-probe system).
specify the corresponding translation by implicitly providing the position of the clicked atom.
In both cases, the translation is applied to the entire molecule and therefore modifies the origin of the molecule as displayed in the left panel of the Atomic Manipulator.
In the same way, it is also possible to translate the right electrode in a two-probe system (the position of the left electrode is always fixed), but only in the Z-direction. Such translations modify the width of the central region.
In the Translate dialog, you can either choose to specify the translation itself, or provide the new absolute coordinates of the right-clicked atom. A useful method for aligning a group is to superimpose an atom on the known position of another atom using the position command (find the position by hovering the mouse cursor over the relevant reference atom). Then specify its final relative position by the translate command. Note how the two sets of input fields are related:
Specifying a number in any field updates the associated field above or below. The translation is put into effect when you click the Apply button; to move the molecule by the same translation in several steps, simply push this button repeatedly. The Close button does not carry out any translation, it just closes the dialog.
An alternative way to translate a molecule is to
hold down the Shift button
click an atom in the molecule
move the mouse while pressing and holding the left mouse button
This can often be useful for a first, rough alignment, but is generally not accurate enough to obtain a good final positioning.
Molecules may be rotated by selecting Rotate from the
context menu that appears when right-clicking an atom in
the Preview window of the Atomic Manipulator. The rotation modifies the
orientation in the left panel of the Atomic
Manipulator. Therefore, rotating an object is the recommended way to obtain a
desired orientation, since the three
angles that yield the desired rotation often are far from obvious.
A rotation is defined by a rotation axis and a point in space, the rotation origin, which is the center of the rotation. Both are specified in the Rotation dialog. By default, the Rotation origin is the absolute position of the selected atom (i.e. the atom that was clicked to open the context menu), but any point in space can be specified. The rotation axis is defined as the unit vector directed from the rotation origin to the point specified by (x,y,z) under Rotation axis. These fields can only be edited when the Custom Axis option is selected; otherwise, just indicate the chosen Cartesian axis.
The rotation is put into effect once you click the Apply button; to rotate the molecule by the same angle in several steps, simply push this button repeatedly. The Close button does not carry out any rotation, it just closes the dialog.
An alternative method for rotating a molecule is to click an atom in the molecule, and move the mouse while holding down the left mouse button. This can often be useful for a first, rough alignment. In general, however, this approach is not accurate enough for obtaining a precise final positioning the molecule.
Instead of creating a regular two-probe from your two-probe system, you also have the option of converting the two-probe system to a bulk system by left-clicking Save Equivalent Bulk. This is useful for studying the properties of the equivalent bulk counterpart of a two-probe system.
In addition, since you cannot optimize a two probe system directly, this feature also provides a useful way to optimize the central region of the two probe and thereby capture the modification of the molecule structure due to the electrodes. Note, that this is done automatically by VNL when a two-probe is optimized.
[1] There are no periodic boundary conditions along the transport direction; if there were, we would not be able to apply a finite bias across the two-probe system.
| |
|
|
| Bulk Builder |
|
Magnetic Tunnel Junction Builder |
