MAGNETIC FORCE MICROSCOPY
Principles of the Method :
Magnetic force microscopy has become one of the most widespread tools for studying the magnetic structure of ferromagnetic samples and superconductors (Rugar and Hansma, 1990; Sarid, 1991). The technique is based on the forces between a very small ferromagnetic tip attached to a flexible cantilever and the inhomogeneous stray magnetic field immediately outside a sample of interest. As the magnetic tip is scanned over a magnetic sample, these minute forces are sensed in any of a variety of ways to provide maps related to the magnetic field above the sample.
In the implementation of an MFM, there are a very large number of choices to be made that influence, in a fundamental way, how the observed image should be related to the magnetic field. For example, an MFM in which the tip is magnetically relatively soft (i.e., the tip magnetization is modified by its interaction with the sample) provides a very different image of the magnetic field than one which utilizes a tip that is magnetically relatively hard (i.e., the tip magnetization does not change in response to the magnetic field of the sample). Similarly, the distribution of magnetic moments in the tip can have a pronounced qualitative effect on the imaging. Whereas a point-like magnetic particle on the tip may best be modeled as a simple magnetic dipole, a tip that is longer or more like a needle may be best modeled as a magnetic monopole.
Furthermore, theMFMmay sense either the net deflection of the tip in the field of the sample or may sense changes in either the amplitude, frequency, or phase of the vibrational motion of a tip oscillating resonantly in the field above the sample. One direct result of the complex and multifaceted interaction between the tip and sample is that it is very difficult to determine the field distribution from an MFM image.
Because of the huge number of possible implementations implied by these (and many other) available choices,
and further because the state-of-the-art for magnetic force microscopy is currently being developed very rapidly in numerous laboratories, it is not possible in the current context to provide a thorough or complete report on the status of magnetic force microscopy. Instead, we will present a general description of an MFM in one form readily available commercially.
Rather than directly sensing the deflection of the flexible cantilever due to magnetic forces acting on the magnetic cantilever tip, MFMs typically detect changes in the resonant vibrational frequency of the cantilever. In
this mode of operation, the cantilever is electrically driven to oscillate, with the driving frequency controlled to track very precisely the resonant vibrational frequency of the cantilever. When they are in close proximity, the scanning tip and sample surface generally experience an attractive net force (e.g., from van der Waals interactions). In addition to this attractive force, there will be a force of magnetic origin, which will be either attractive or repulsive depending on the relative orientation of the tip magnetization and the magnetic field gradients above the sample. Where the magnetic force is also attractive, the cantilever is deflected further towards the surface. This effectively stiffens the cantilever, raising its natural resonant frequency. If the magnetic force is repulsive, the cantilever is deflected less strongly towards the sample, effectively softening the cantilever and lowering its resonant frequency. Because the resonant frequency of the cantilever can be determined with very high precision,that frequency provides a convenient means by which to monitor local variations in the magnetic field gradients. However, because the magnetic forces between tip and
sample are very small, one general problem that all MFMs must address is the separation of the image contrast that arises from magnetic forces from the image contrast that is due to other (stronger) short-range physical forces (Schoenenberger et al., 1990).
Figure 1 shows, as an inset, a magnified region of a test pattern written on magnetic storage media as imaged with a commercial magnetic force microscope. The written bits are clearly visible and highlight the fact that the MFM is primarily sensitive to gradients in the magnetic field.
Regions where the magnetization changes from left to right (or right to left) are visible as white (or dark) lines.
This contrast would reverse with a magnetization reversal of the MFM tip.
Practical Aspects of the Method
There are several practical aspects to MFM imaging that should be noted. First, it is generally not known with confidence what underlying contrast mechanism gives rise to an MFM image. The signal is generally proportional to spatial derivatives of the stray field above the sample,but which spatial derivative in which direction depends on the such factors as the details of the magnetic moment distribution within the tip, tip/sample interactions, operating mode of the MFM signal detection and control electronics,vibrational amplitude of the cantilever, and lift height of the MFM scan relative to the AFM topographic scan. As a consequence, quantitative interpretation of MFM images is generally very difficult (Hug et al.,1998). Even qualitative interpretation can sometimes be very uncertain. Because the MFM can often be configured
to give contrast that is proportional to the magnetic field as well as to various spatial derivatives, one is often uncertain whether observed magnetic contrast is due to a magnetic domain or to the domain wall between two domains.
As with many scanned-tip microscopes, navigation on a sample can be difficult. The MFM always operates at
very high magnification, so it can be difficult to find specific isolated features for study. As mentioned above, the effects of surface topography on magnetic force microscopy can be pronounced. One common solution to this problem is to acquire the magnetic image in two steps. In the first step, the microscope is operated
as a conventional atomic force microscope (AFM; not discussed here) so as to determine in detail the topographic profile along one scan line. The microscope then retracts the tip and, using the previously determined line profile,rescans the same line with the magnetic tip at a small but constant height (typically 20 to 200 nm) above the sample. In this second scan, the longer-range magnetic forces still affect the cantilevered tip, but the effect of the shortrange forces is minimized. This effectively provides a map
only of variations in the local magnetic field, free of topographic contrast.
Even with such methods to compensate for the effects of surface topography, samples must be very smooth for these methods to be effective. Surfaces that have significant roughness or surface relief can be very difficult to image with the MFM. Sometimes the AFM prescan is not adequate because the surface relief is too great for the AFM tip positioning to be reliable. Even if the AFM topography scan is successful, other problems arise if the combination of tip vibrational amplitude and surface texturing is so large that the tip contacts the sample during the MFM scan.
One very useful feature of the MFM is that, because the MFM senses the stray magnetic field rather than the sample magnetization directly, it is easy to see the magnetic structure even through relatively thick nonmagnetic and even insulating overlayers.
Data Analysis and Initial Interpretation :
Significant image processing is generally required to aid in the interpretation of MFM images. The relevant imageprocessing steps are typically integrated into the commercial MFM instrument controllers. One example is the abovementioned subtraction of contrast due to surface topography. This subtraction, however, has the side effect that domain walls that run parallel to the scan direction can be much more difficult to image than walls running at a significant angle to the scan direction.
Sample Preparation
One of the very attractive features of the MFM is that minimal surface preparation is required. Samples that
are smooth and flat enough to image with an atomic force microscope can generally be studied with MFM as well.
Problems
As mentioned above, consideration must always be given to the extent to which the magnetic structure of the tip or sample is modified in response to the magnetic field of the other.
Principles of the Method :
Magnetic force microscopy has become one of the most widespread tools for studying the magnetic structure of ferromagnetic samples and superconductors (Rugar and Hansma, 1990; Sarid, 1991). The technique is based on the forces between a very small ferromagnetic tip attached to a flexible cantilever and the inhomogeneous stray magnetic field immediately outside a sample of interest. As the magnetic tip is scanned over a magnetic sample, these minute forces are sensed in any of a variety of ways to provide maps related to the magnetic field above the sample.
In the implementation of an MFM, there are a very large number of choices to be made that influence, in a fundamental way, how the observed image should be related to the magnetic field. For example, an MFM in which the tip is magnetically relatively soft (i.e., the tip magnetization is modified by its interaction with the sample) provides a very different image of the magnetic field than one which utilizes a tip that is magnetically relatively hard (i.e., the tip magnetization does not change in response to the magnetic field of the sample). Similarly, the distribution of magnetic moments in the tip can have a pronounced qualitative effect on the imaging. Whereas a point-like magnetic particle on the tip may best be modeled as a simple magnetic dipole, a tip that is longer or more like a needle may be best modeled as a magnetic monopole.
Furthermore, theMFMmay sense either the net deflection of the tip in the field of the sample or may sense changes in either the amplitude, frequency, or phase of the vibrational motion of a tip oscillating resonantly in the field above the sample. One direct result of the complex and multifaceted interaction between the tip and sample is that it is very difficult to determine the field distribution from an MFM image.
Because of the huge number of possible implementations implied by these (and many other) available choices,
and further because the state-of-the-art for magnetic force microscopy is currently being developed very rapidly in numerous laboratories, it is not possible in the current context to provide a thorough or complete report on the status of magnetic force microscopy. Instead, we will present a general description of an MFM in one form readily available commercially.
Rather than directly sensing the deflection of the flexible cantilever due to magnetic forces acting on the magnetic cantilever tip, MFMs typically detect changes in the resonant vibrational frequency of the cantilever. In
this mode of operation, the cantilever is electrically driven to oscillate, with the driving frequency controlled to track very precisely the resonant vibrational frequency of the cantilever. When they are in close proximity, the scanning tip and sample surface generally experience an attractive net force (e.g., from van der Waals interactions). In addition to this attractive force, there will be a force of magnetic origin, which will be either attractive or repulsive depending on the relative orientation of the tip magnetization and the magnetic field gradients above the sample. Where the magnetic force is also attractive, the cantilever is deflected further towards the surface. This effectively stiffens the cantilever, raising its natural resonant frequency. If the magnetic force is repulsive, the cantilever is deflected less strongly towards the sample, effectively softening the cantilever and lowering its resonant frequency. Because the resonant frequency of the cantilever can be determined with very high precision,that frequency provides a convenient means by which to monitor local variations in the magnetic field gradients. However, because the magnetic forces between tip and
sample are very small, one general problem that all MFMs must address is the separation of the image contrast that arises from magnetic forces from the image contrast that is due to other (stronger) short-range physical forces (Schoenenberger et al., 1990).
Figure 1 shows, as an inset, a magnified region of a test pattern written on magnetic storage media as imaged with a commercial magnetic force microscope. The written bits are clearly visible and highlight the fact that the MFM is primarily sensitive to gradients in the magnetic field.
Regions where the magnetization changes from left to right (or right to left) are visible as white (or dark) lines.
This contrast would reverse with a magnetization reversal of the MFM tip.
Practical Aspects of the Method
There are several practical aspects to MFM imaging that should be noted. First, it is generally not known with confidence what underlying contrast mechanism gives rise to an MFM image. The signal is generally proportional to spatial derivatives of the stray field above the sample,but which spatial derivative in which direction depends on the such factors as the details of the magnetic moment distribution within the tip, tip/sample interactions, operating mode of the MFM signal detection and control electronics,vibrational amplitude of the cantilever, and lift height of the MFM scan relative to the AFM topographic scan. As a consequence, quantitative interpretation of MFM images is generally very difficult (Hug et al.,1998). Even qualitative interpretation can sometimes be very uncertain. Because the MFM can often be configured
to give contrast that is proportional to the magnetic field as well as to various spatial derivatives, one is often uncertain whether observed magnetic contrast is due to a magnetic domain or to the domain wall between two domains.
As with many scanned-tip microscopes, navigation on a sample can be difficult. The MFM always operates at
very high magnification, so it can be difficult to find specific isolated features for study. As mentioned above, the effects of surface topography on magnetic force microscopy can be pronounced. One common solution to this problem is to acquire the magnetic image in two steps. In the first step, the microscope is operated
as a conventional atomic force microscope (AFM; not discussed here) so as to determine in detail the topographic profile along one scan line. The microscope then retracts the tip and, using the previously determined line profile,rescans the same line with the magnetic tip at a small but constant height (typically 20 to 200 nm) above the sample. In this second scan, the longer-range magnetic forces still affect the cantilevered tip, but the effect of the shortrange forces is minimized. This effectively provides a map
only of variations in the local magnetic field, free of topographic contrast.
Even with such methods to compensate for the effects of surface topography, samples must be very smooth for these methods to be effective. Surfaces that have significant roughness or surface relief can be very difficult to image with the MFM. Sometimes the AFM prescan is not adequate because the surface relief is too great for the AFM tip positioning to be reliable. Even if the AFM topography scan is successful, other problems arise if the combination of tip vibrational amplitude and surface texturing is so large that the tip contacts the sample during the MFM scan.
One very useful feature of the MFM is that, because the MFM senses the stray magnetic field rather than the sample magnetization directly, it is easy to see the magnetic structure even through relatively thick nonmagnetic and even insulating overlayers.
Data Analysis and Initial Interpretation :
Significant image processing is generally required to aid in the interpretation of MFM images. The relevant imageprocessing steps are typically integrated into the commercial MFM instrument controllers. One example is the abovementioned subtraction of contrast due to surface topography. This subtraction, however, has the side effect that domain walls that run parallel to the scan direction can be much more difficult to image than walls running at a significant angle to the scan direction.
Sample Preparation
One of the very attractive features of the MFM is that minimal surface preparation is required. Samples that
are smooth and flat enough to image with an atomic force microscope can generally be studied with MFM as well.
Problems
As mentioned above, consideration must always be given to the extent to which the magnetic structure of the tip or sample is modified in response to the magnetic field of the other.
Posts
No comments:
Post a Comment