This investigation is devoted to a method of deposition of catalytic nickel nanoparticles of controllable sizes, surface density, and shape on a smooth substrate. The nickel nanoparticles are intended for use as catalytic centers for a low-temperature plasma-enhanced chemical vapor deposition (PECVD) synthesis of various carbon nanostructures, viz. carbon nanotubes (CNT), carbon nanofibers (CNF), and carbon nanospheres (CNS). The prepared nanoparticles can also be used as highly sensitive elements of detectors as well as magnetic storage medium.
Within the framework of the project construction and control methods of a miniature robot-nanopositioner of “walker” type (noninertial) are investigated. The positioner under development is free of its predecessor's drawbacks such as: x, y, z position instabilities, undesired rotation in the sample plane and sticking, all due to microroughnesses and pollutions in the contact area of electrostatic supports. The positioner can be operated in both “feet up” and “feet down” configurations. The suggested positioner has built-in capabilities for compensating its own thermal drift and creep along x, y, and z directions that is especially important in the environments with strong temperature gradients. Although the robot-nanopositioner is capable of autonomously routing while moving by a bearing surface, it is not a completely autonomous device, it still requires an external power supply and control wires. The robot-nanopositioner is absolutely noncritical to roughness of the surface by which it travels. Another advantage of the suggested positioner is its full compatibility with the feature-oriented scanning (FOS) approach. Both ambient and UHV arrangements will be possible. The positioner preserves performance characteristics upon the tilt, contains no magnetic materials. Due to a low weight and small dimensions, the proposed nanopositioner may be incorporated in many instruments like SPM, SEM, FIB, AES, SIMS, etc. This nanopositioner is mounted instead of a sample and the sample itself is mounted on top of the nanopositioner.
A sort of feature-oriented scanning (FOS) called distributed calibration being implemented on a standard surface makes it possible to determine the local calibration coefficients (LCCs) Kx, Ky, Kz for each point of the scanner movement space. Application of FOS excludes in situ the negative influence of thermal drift, creep and hysteresis on the obtained results. The sensitivity of LCCs to errors in determination of position coordinates of surface features forming the local calibration structure (LCS) is eliminated by performing multiple repeated measurements followed by building a regression surfaces. There are no principle restrictions on the number of repeated LCS measurements.
By using the collected calibration database, all spatial distortions induced by nonlinearity, nonorthogonality and parasitic crosstalk couplings of X, Y, Z manipulators of the scanner may be corrected in one operation. To provide high precision of spatial measurements in nanometer range, the calibration is carried out using natural standards – stable reconstructions of atomic surfaces of crystals.
Counter-scanning (CS) is a method for measuring surface topography with a scanning probe microscope enabling correction of raster distortions resulted from drift of the microscope probe relative to the surface being measured. Two surface scans, viz. direct scan and counter one, are obtained during CS. The counter scan starts in the point where the direct scan ends. This point is called the coincidence point (CP). With the counter scan, the probe movement along the raster line and the probe movement from one raster line to the other raster line are carried out along the directions that are opposite to the movements in the direct scan. The obtained pair of images is called the counter-scanned images (CSIs).
To implement correction, it is required to find the same feature on both CSIs and to determine its lateral coordinates. Because of drift, the same topography points on the direct and the counter scans do not coincide with each other except for a single point – the coincidence point. Building a simple system of linear equations, linear transformation coefficients (LTCs) may be calculated for each CSI. Using the LTCs found, drift correction is carried out along x, y and z, then the corrected images are matched in CP and then the topography is averaged out in the overlap area of these images. It is supposed while composing the system of equations that the drift velocity is constant during the scanning time. Therefore, the slower is the change in drift velocity, the higher is the correction precision.
By reading the signal during the retrace sweep while scanning the direct and the counter images, another CSI pair may be obtained. Working with two CSI pairs allows to increase the accuracy of drift correction as well as to decrease the noise level in the output image. If the drift velocity is changing significantly during counter-scanning, then the developed nonlinear correction methods should be applied.
The less the scan sizes are, the less probable is a change in the drift velocity during counter-scanning. Therefore, CS method would reach a maximal effect when used with feature-oriented scanning (FOS) method, which operates with small-sized scans – segments and apertures. Moreover, as drift is correctable within the apertures, it allows increasing FOS productivity by refusing the skipping operation and building the surface of apertures instead of segments. This is, however, at the cost of some decrease in FOS measurement precision.
Application of CS allows:
(1) Easy detecting feature pairs on CSIs due to the existence of a CP whose neighborhood actually contains the same topography.
(2) Increasing the drift correction precision due to a greater difference in position coordinates of the features that compose a pair.
(3) Ensuring a preset measurement error within the bounds of a certain image area in accordance with the actual change in drift velocity that occurred during the CS.
(4) Decreasing the creep produced by the piezoscanner during FOS by means of inducing a counter creep while scanning topography segments/apertures.
Feature-oriented scanning (FOS) is a method intended for high-precision measurement of nanotopography as well as other surface properties and characteristics on a scanning probe microscope using features (objects) of the surface as reference points of the microscope probe. With this method, during successive passings from one surface feature to another one located nearby, the relative distance between the features and the topography of their neighborhoods called surface segments are measured. Such approach permits to scan the required area of the surface by parts and then reconstruct the whole image from the obtained fragments.
The surface segment mentioned is a square scan of feature neighborhood covering the feature with certain margins. The sizes of the segments are adjusted in such a way that the segments of neighboring features partially overlap each other so that the topography of the measured surface can be reconstructed later with no gaps. Scanning in the segment is carried out in counter manner thus permitting correction of the raster distortions produced by drift of the microscope probe relative to the measured surface.
Making real-time program recognition of a feature in the segment, one may determine the drift-induced displacement of this feature relative to the segment center. The displacement found is compensated by the corresponding change in the probe position. Repeating periodically the segment scan, the segment recognition and correction of the probe position, it is possible to implement tracking the feature with the probe also known as attachment of the microscope probe to the surface feature.
Any topography elements that look like hill or pit in the wide sense may be taken as surface features. Some examples of surface features (objects) are atoms, interstices, molecules, clusters, grains, nanoparticles, crystallites, quantum dots, nanoislets, nanopillars, pores, short nanowires/nanorods/nanotubes, elements of chains, bacteria, viruses, cells, organelles, etc. The only restriction imposed on the used feature is a requirement of commensurability of its lateral sizes. In other words, extends of the features should be comparable in different directions in the lateral plane. Otherwise, the feature can not be localized in a small-sized segment. Unsuitable surfaces, for instance, are a defectless surface of a one-dimensional diffraction grating, a surface of integrated circuit with a great number of long conducting wires, and so forth.
On the whole, FOS of a surface is carried out by the following steps:
(1) Detecting and catching the nearest (current) surface feature.
(2) Scanning the current feature neighborhood called an aperture that contains several neighboring features.
(3) Selecting among the neighboring features the next one in accordance with a certain rule of connection.
(4) Implementing the skipping (see below) between the current and the next features.
(5) Moving the microscope probe to the position of the next feature which becomes current after that.
The pointed out sequence of operations is repeated until the set surface area is scanned out (theoretically, there are not restrictions on sizes of the area).
Skipping is a basic measurement operation in FOS intended for accurate determination of relative coordinates of the neighboring features and acquisition of topography segments. The operation of feature skipping consists in moving the probe from the current feature position to the next feature position, scanning-recognizing the segment of the next feature, calculating the “forward” difference between the current and the next feature coordinates, moving the probe back to the position of the current feature, scanning-recognizing its segment and calculating the “backward” difference between the current and the next feature coordinates.
The relative coordinates of the next feature are calculated as a half-sum of the obtained forward and backward differences that allows us to exclude the influence of drift on measurement of the distance between features. Thus, by specifying a large number of skipping cycles, the measurement precision can be considerably improved by means of measurement averagings. The microscope resolution can also be improved this way provided that a sharp probe is available. The number of the FOS-executed averagings is estimated as hundreds of thousands or even some millions. Theoretically, the number of averagings is unlimited. In practice, however, the number of averagings is mainly restricted by a long-term stability of the investigated surface, experimental environment and characteristics of the probe microscope.
Generally, FOS is based on the following principles: localization of measurements; operating with separate surface features; movement at short distances from one feature to another located in the vicinity; measurement of relative distances; multiple iterations of the measurements; ceaseless probe attachments to the surface features; continuous monitoring of drift velocity; hierarchically organized counter movements.
It should be noted that unlike the conventional scanning, the trajectory of probe movement in FOS is not defined beforehand. The trajectory is being built dynamically during FOS execution, and preset is only the law of feature connection, in a general form.
The advantage of FOS method as compared to the regular scanning is direct operating with surface features which are mostly the subject of research or technology. It should be emphasized that the FOS topography measurement method proves to be ineffective when scanning surfaces with small number of features or no features at all. The fact is that FOS on such surfaces will attempt to reveal the rare features by gradually increasing the number of points in the aperture and on failure it will try to change the scale of the measurements by gradually decreasing/increasing the scan step. The increase in aperture size is equal to degeneration of FOS method into the conventional scanning. Adjustment of aperture size and search for suitable scale require additional time.
Strictly speaking, there are no absolutely smooth surfaces (even atomically flat surfaces have finite corrugation); the problem is usually formulated as follows: whether the features on the surface correspond to the measurement scale that the researcher/technologist is currently interested in as well as if the features are contrast and stable enough to serve as reliable places of microscope probe attachment.
Scanning probe microscopy is the major area of FOS method application. Since the FOS method implies recognition of the image scanned, the topography features should be understood in the broad sense. They refers not only to pure topography, but also to physical inhomogeneities such as magnetization domains, places of localized electric charge, and so forth. Therefore, the method may be used with any instrument of the SPM family: a scanning tunneling microscope (STM), an atomic-force microscope (AFM), a magnetic-force microscope (MFM), an electric-force microscope (EFM), a near-field scanning optical microscope (NSOM), etc. Moreover, FOS can be used in a scanning electron microscope (SEM), a laser scanning confocal microscope (LSCM), an optical and mechanical profilometers. Generally speaking, FOS method may be used with any device that has a probe (mechanical tip, focused light beam, focused electron beam, focused ion beam, etc.), a scanning system (a scanner) and a unit that registers the results of the probe interacting with the measured surface (a detector).
Beside the scanning probe microscopy, FOS may be applied to the bottom-up nanofabrication. The bottom-up nanofabrication implies elementwise assembling of nanodevices. Nanoclusters, nanoparticles, molecules or even single atoms may be used as construction blocks. The apparatus able to implement assemblage of nanodevices is called nanoassembler. In the future, the nanoassembler will have an array of specialized probes – one part of them is intended for technological operations and the other for analytical measurements and checking. FOS is able to take control over the nanoassembler making the nanofabrication completely automatical. The operator should only formulate the task in general terms, which should then act according to the principle “run and forget”. At present, SPM is used as the nanoassembler, which allows testing various measurement techniques, particular technological modes and operations.
Feature-oriented positioning (FOP) is a method of precise movement of the scanning microscope probe across the surface under investigation. With this method, surface features are used as reference points (points of probe attachment). Actually, FOP is a simplified variant of feature-oriented scanning (FOS). With FOP, no topographical image of a surface area is acquired. Instead, a probe movement by surface features is only carried out from the start point A (neighborhood of the start feature) to the destination point B (neighborhood of the destination feature) along some route that goes through intermediate features of the surface.
To be distinguished are a “blind” FOP and a FOP by known feature “map”. The blind FOP is when the coordinates of features used for probe movement are unknown in advance. FOP by known feature map is when the relative coordinates of all features are known, for example, in case they were obtained during preliminary FOS. Probe movement by a navigation structure is a combination of the pointed above methods. With this method, a general topology of the structure is known a priori and also approximately (because of technological uncertainty and surface instability) known are: sizes of the elements, their shape, the distance between neighboring elements, movement direction. At the same time, the current probe position within the structure is not required to be known precisely, and it is also admissible for the probe to get into a defective area on the structure, etc.
Besides providing high-precision movement of the microscope probe within the field of the fine positioner, application of FOP allows us for high-precision placement of the fine positioner field in an arbitrary-large field of a coarse positioner. When the fine positioner reaches the edge of its movement range, it is carrying out repeated probe attachments to the current feature. At the same time, the coarse positioner starts to move the probe relative to the sample surface slowly by small steps. The fine positioner “feels” this movement in the attachment cycles by a displacement of the current feature. The movement direction of the coarse positioner is chosen so that to make the fine positioner follow the current feature towards the opposite edge of its range. The coarse positioner continues to move until the fine positioner reaches the opposite edge of its range. After that, the fine positioner may proceed moving along the previous direction.
The described method of allocating the fine positioner field within the coarse positioner field permits both scanning and technological operations at large surface areas with the accuracy of the fine positioner. In both cases, the movement length is only restricted by the range of the coarse positioner. The advantage of the suggested approach is that no strict demands are actually made of the precision and linearity of the coarse positioner.
Beside scanning probe microscopy, surface tunneling and force spectroscopies, FOP method may be used in bottom-up nanofabrication to implement high-precision movement of the nanolithograph/nanoassembler probe along the sample surface during nanotechnological operations. Moreover, once made along some route FOP may be then exactly repeated the required number of times. With multiprobe instruments, FOP approach allows to apply any number of specialized technological and/or analytical probes successively to any surface feature (object) or to any point of its neighborhood. That enables complex nanofabrication consisting of a large number of technological, measuring and checking operations.