[2] Field of the Invention
The present invention is directed to scanning probe microscopes (SPMs), including atomic force microscopes (AFMs), and more particularly, to using Peak Force Tapping® (PFT) (Peak Force Tapping is a registered trademark of Bruker Nano, Inc.) mode of AFM operation, as described in the above-identified applications, to measure sample properties using PFT mode and at least one of electrical, thermal, microwave and optical sample excitation, for example.
[3] Description of Related Art
Scanning probe microscopes (SPMs), such as the atomic force microscope (AFM), are devices which typically employ a probe having a tip and which cause the tip to interact with the surface of a sample with low forces to characterize the surface down to atomic dimensions. Generally, the probe is introduced to a surface of a sample to detect changes in the characteristics of a sample. By providing relative scanning movement between the tip and the sample, surface characteristic data can be acquired over a particular region of the sample, and a corresponding map of the sample can be generated.
A typical AFM system is shown schematically in
Notably, scanner 24 often comprises a piezoelectric stack (often referred to herein as a “piezo stack”) or piezoelectric tube that is used to generate relative motion between the measuring probe and the sample surface. A piezo stack is a device that moves in one or more directions based on voltages applied to electrodes disposed on the stack. Piezo stacks are often used in combination with mechanical flexures that serve to guide, constrain, and/or amplify the motion of the piezo stacks. Additionally, flexures are used to increase the stiffness of actuator in one or more axis, as described in application Ser. No. 11/687,304, filed Mar. 16, 2007, entitled “Fast-Scanning SPM Scanner and Method of Operating Same.” Actuators may be coupled to the probe, the sample, or both. Most typically, an actuator assembly is provided in the form of an XY-actuator that drives the probe or sample in a horizontal, or XY-plane and a Z-actuator that moves the probe or sample in a vertical or Z-direction.
In a common configuration, probe 17 is often coupled to an oscillating actuator or drive 16 that is used to drive probe 12 to oscillate at or near a resonant frequency of cantilever 15. Alternative arrangements measure the deflection, torsion, or other characteristic of cantilever 15. Probe 17 is often a microfabricated cantilever with an integrated tip 17.
Commonly, an electronic signal is applied from an AC signal source 18 under control of an SPM controller 20 to cause actuator 16 (or alternatively scanner 24) to drive the probe 12 to oscillate. The probe-sample interaction is typically controlled via feedback by controller 20. Notably, the actuator 16 may be coupled to the scanner 24 and probe 12 but may be formed integrally with the cantilever 15 of probe 12 as part of a self-actuated cantilever/probe.
Often, a selected probe 12 is oscillated and brought into contact with sample 22 as sample characteristics are monitored by detecting changes in one or more characteristics of the oscillation of probe 12, as described above. In this regard, a deflection detection apparatus 25 is typically employed to direct a beam towards the backside of probe 12, the beam then being reflected towards a detector 26, such as a four quadrant photodetector. The deflection detector is often an optical lever system such as described in Hansma et al. U.S. Pat. No. RE 34,489, but may be some other deflection detector such as strain gauges, capacitance sensors, etc. The sensing light source of apparatus 25 is typically a laser, often a visible or infrared laser diode. The sensing light beam can also be generated by other light sources, for example a He—Ne or other laser source, a superluminescent diode (SLD), an LED, an optical fiber, or any other light source that can be focused to a small spot. As the beam translates across detector 26, appropriate signals are processed by a signal processing block 28 (e.g., to determine the RMS deflection of probe 12). The interaction signal (e.g., deflection) is then transmitted to controller 20, which processes the signals to determine changes in the oscillation of probe 12. In general, controller 20 determines an error at Block 30, then generates control signals (e.g., using a PI gain control Block 32) to maintain a relatively constant interaction between the tip and sample (or deflection of the lever 15), typically to maintain a setpoint characteristic of the oscillation of probe 12. The control signals are typically amplified by a high voltage amplifier 34 prior to, for example, driving scanner 24. For example, controller 20 is often used to maintain the oscillation amplitude at a setpoint value, As, to insure a generally constant force between the tip and sample. Alternatively, a setpoint phase or frequency may be used. Controller 20 is also referred to generally as feedback where the control effort is to maintain a constant target value defined by setpoint.
A workstation 40 is also provided, in the controller 20 and/or in a separate controller or system of connected or stand-alone controllers, that receives the collected data from the controller and manipulates the data obtained during scanning to perform data manipulation operations such as point selection, curve fitting, and distance determining operations. The workstation can store the resulting information in memory, use it for additional calculations, and/or display it on a suitable monitor, and/or transmit it to another computer or device by wire or wirelessly. The memory may comprise any computer readable data storage medium, examples including, but not limited to, a computer RAM, hard disk, network storage, a flash drive, or a CD ROM.
AFMs may be designed to operate in a variety of modes, including contact mode and oscillating mode. Operation is accomplished by moving the sample and/or the probe assembly up and down relatively perpendicular to the surface of the sample in response to a deflection of the cantilever of the probe assembly as it is scanned across the surface. Scanning typically occurs in an “x-y” plane that is at least generally parallel to the surface of the sample, and the vertical movement occurs in the “z” direction that is perpendicular to the x-y plane. Note that many samples have roughness, curvature, and tilt that deviate from a flat plane, hence the use of the term “generally parallel.” In this way, the data associated with this vertical motion can be stored and then used to construct an image of the sample surface corresponding to the sample characteristic being measured, e.g., surface topography. In one practical mode of AFM operation, known as TappingMode™ AFM (TappingMode™ is a trademark of the present assignee), the tip is oscillated at or near a resonant frequency of the associated cantilever of the probe, or harmonic thereof. A feedback loop attempts to keep the amplitude of this oscillation constant to minimize the “tracking force,” i.e., the force resulting from tip/sample interaction, typically by controlling tip-sample separation. Alternative feedback arrangements keep the phase or oscillation frequency constant. As in contact mode, these feedback signals are then collected, stored and used as data to characterize the sample.
Regardless of their mode of operation, AFMs can obtain resolution down to the atomic level on a wide variety of insulating or conductive surfaces in air, liquid or vacuum by using piezoelectric scanners, optical lever deflection detectors, and very small cantilevers fabricated using photolithographic techniques. Because of their resolution and versatility, AFMs are important measurement devices in many diverse fields ranging from semiconductor manufacturing to biological research. Note that “SPM” and the acronyms for the specific types of SPMs, may be used herein to refer to either the microscope apparatus or the associated technique, e.g., “atomic force microscopy.”
As with most measuring devices, AFMs often require a trade-off between resolution and acquisition speed. That is, some currently available AFMs can scan a surface with sub-angstrom resolution. These scanners are capable of scanning only relatively small sample areas, and even then, at only relatively low scan rates. Traditional commercial AFMs usually require a total scan time typically taking several minutes to cover an area of several microns at high resolution (e.g. 512×512 pixels) and low tracking force. The practical limit of AFM scan speed is a result of the maximum speed at which the AFM can be scanned while maintaining a tracking force that is low enough not to damage or cause minimal damage to the tip and/or sample. Great strides have been made in this area in which SPM has achieved video scan rates with high resolution for small samples and small scan sizes.
Nonetheless, given current limitations associated with known modes of operation, including both TappingMode™ AFM and contact mode, improvements have been desired. Again, in contact mode, lateral scanning of the tip creates large forces between the tip and sample that can compromise both. And when imaging soft samples such as biological samples and polymers, the surface can be destroyed, rendering the measurement useless, or at least deformed severely, thereby significantly compromising resolution. Note that “imaging” is used herein to indicate obtaining SPM data at multiple points of a sample surface, typically by providing relative scanning motion between the sample and probe and correspondingly interacting the sample and probe.
TappingMode™ AFM is a lower force technique and is the most widely used mode of AFM operation to map sample surfaces, especially for delicate samples. The typical force of the tip on the sample is about a few nN to tens of nN. Again, by oscillating the tip, rather than dragging the tip, the shear forces are minimized. That said, TappingMode™ AFM suffers from a drawback in that it is difficult to control the normal force acting on the sample surface. The user typically tries to select a setpoint that is only a small variation from the free air deflection/amplitude of the probe in order to minimize tip-sample interaction forces to get the best reproduction of the sample profile. The dilemma, especially for soft samples, is that if the imaging force is too low, the tip will not track the sample properly (i.e., maintain interaction with the sample during the scan), while if too high, damage/deformation of the sample may lead to an image that does not accurately reflect surface topography. Overall, the better this force can be controlled (i.e., the lower it can be maintained) the less chance of sample and/or tip damage, and thus resolution can be improved.
A review of the tip-sample forces in each of these modes provides insight in to the limitations of each. When a probe interacts with the surface through TappingMode™ AFM or Jumping Mode™ (see, e.g., U.S. Pat. Nos. 5,229,606, 5,266,801 and 5,415,027, the entirety of which are incorporated by reference herein), the tip touches the surface periodically.
As the tip departs the surface after δT, an attractive force will develop a capillary meniscus, exhibiting a maximum adhesion force Fa_max right before the meniscus is broken away. The tip then enters into a non-interactive region and continues to a maximum departure position.
In the interaction free zone, when the probe is farther from the surface, the interaction force is zero or sufficiently near zero to form a baseline, as indicated in
In prior known modes disclosed in TappingMode™ AFM and JumpingMode™ AFM, the amplitude Amax or RMS of the tip oscillation amplitude is used as the feedback control parameter. An example of such feedback control apparatus is shown in
In conventional control, typically implemented using a gain control feedback loop, positioning actuators and a cantilever response detection component (quadrant photodetector, for example), the AFM uses detected probe deflection or an RMS signal corresponding to cantilever (i.e., probe) motion as an indication of the tip-surface interaction and uses the feedback loop to maintain constant or RMS deflection.
Yet a major limitation of conventional AFM is its inability to acquire quantitative mechanical property information simultaneously with the high-resolution imaging. AFM has been primarily focused on topographical imaging. Little progress has been made in achieving quantitative mechanical mapping, including elasticity, plasticity, and work of adhesion.
Moreover, TappingMode™ control uses amplitude or phase of the measured deflection signal to control tip-surface interaction using feedback. Notably, both amplitude and phase are average properties of the probe/tip oscillation using at least one cycle of interaction. More specifically, the average pertains to probe/sample interactions occurring in all the positions in the tip trajectory (
In addition, it is important to note that TappingMode™ AFM was created to overcome what is known as the stick-in condition that occurs when probe touches the sample intermittently. As the probe touches the sample, capillary force will tend to catch the tip and prevent it from releasing. The amplitude of probe oscillation in TappingMode™ will drop to zero, thereby causing feedback oscillation. This problem was overcome when using TappingMode™ by using probes having a certain stiffness, usually 10 N/m (Newton/meter) to 60 N/m, with a nominal value of 40 N/m, while operating the TappingMode™ AFM at an oscillation amplitude higher than about 10 nm peak-to-peak. Under these conditions, as the probe touches surface, the kinetic energy of the tapping probe converts to enough static elastic energy to overcome the capillary force, assuring steady amplitude in each cycle. One drawback of this mode is that the kinetic energy stored in the probe is also proportional to the cantilever spring constant. When employing a lower spring constant cantilever, such as 1 N/m, TappingMode™ is impossible when measuring many materials because the cantilever cannot overcome the capillary adhesion forces using its own resonance oscillation energy. Consequently, most TappingMode™ applications are only possible when one uses a stiff cantilever generally know in the art as a lever.
In an alternate mode of operating an SPM, known as the pulsed-force mode or PFM (see, e.g., U.S. Pat. No. 6,880,386 and U.S. Pat. No. 7,129,486), the amplitude of the oscillation of the probe is adjusted so the tip goes in and out of contact during each cycle. In this mode, control is provided by monitoring tip-sample interaction forces. It operates based on properties associated with a force curve, another common measurement made in the AFM field to measure material properties at a particular location. Force measurements are common, and can be mapped over an entire sample to create what is known as a force-volume image.
In PFM, by analyzing the shape of the force-distance curve, and using the data to control the forces acting between the tip and the sample, the amount of data acquired is lessened compared to other modes of SPM operation. Importantly, PFM typically needs to operate at Fr_i (discussed below) or the peak pulse force, which substantially exceeds the adhesion induced deflection, as well as coupling induced deflections. As a result, a high repulsive force is needed as a control reference. Such high force could damage the sample or the tip, and thus prevent acquisition of high resolution images. Moreover, PFM has other limitations, particularly with respect to operating speed and resolution limitations, and thus, though it has been implemented to image soft samples, it has not been more widely adopted for all types of AFM imaging applications. In addition, imaging in a fluid environment presents a further challenge to PFM since viscous force in fluid produces large deflection even when the cantilever probe is not interacting with the sample.
More particularly, a main reason why imaging speed is limited in standard PFM AFM is illustrated in
In addition to the above-noted issues, setup and operation of the relatively complex and versatile AFM can be time consuming and tricky, especially for a novice AFM operator and/or a scientist or engineer not familiar with complex metrology equipment. For example, setup and operating parameter values typically depend on factors such as the type of sample material, including whether it is hard or soft, conductive or non-conductive, and organic, synthetic or biological in nature, among other things.
In other measurement techniques such as scanning electron microscopy (SEM), a sample can readily be mounted in the instrument and a good image obtained with little user training or expertise. However, AFM is often the preferred technique given its ability to make a wide range of measurements including multidimensional topography and mechanical properties (elasticity, etc.). Nonetheless, AFM most often requires expert knowledge of the tool and the measurements to be made. In this regard, the user needs to locate a position of interest, introduce the tip of the probe to the sample (by moving either the sample or the probe). Then, once a measurement scan is initiated, the user needs to make sure the tip tracks the sample, typically by maintaining a stable feedback loop.
Moreover, once a measurement has been made, interpreting the data obtained is often a challenge. In general, these can be time consuming tasks that most often require the knowledge and experience of a physicist or electronics engineer, with the limitations attendant to relying on human judgment. Importantly, because AFM has the potential for wide applicability, it would be advantageous if the AFM did not rely so heavily on an expert's ability to perform. For example, given its ability to obtain unmatched material property measurements, including maps of samples, biologists and material science experts would more widely employ AFM if it were easier to use. In this regard, ease of use would be aided if the AFM and/or method of operation could minimize or eliminate the challenges associated with both a) maintaining feedback stability while making and preparing to make measurements and b) interpreting the data obtained.
To address these issues, the fundamental challenges presented by AFM and its currently preferred operating modes were considered. Initially, with respect to maintaining stability in known AFM modes, controller adjustment is critical. In most current commercial systems, the user must control both the set-point as well as the gain (I (integral) and P (proportional)). With respect to the set-point, control depends on the mode. In contact mode, the instrument attempts to maintain constant contact force between the tip and sample, which is relatively straightforward. However, in the most widely used mode of AFM operation, oscillating mode or TappingMode™ AFM described above, controlling the set-point (tapping amplitude or phase) is complicated because, most fundamentally, there is no straightforward relationship between the set-point and the tip-sample forces. The same set-point change can indicate either high or low tip-sample interaction force, with cantilever dynamics (fundamental resonant frequency, etc.) being highly influential, including with respect to imaging in varying environments (fluid v. atmosphere, for instance).
Stable and optimal feedback also requires applying appropriate gains. Generally, feedback will become unstable under high gain, and will have reduced tracking capability under low gain. P and I gain are adjusted with the user typically employing trial and error to make sure the feedback remains stable, while also providing sufficient tracking capability. However, in TappingMode™ AFM, the feedback dynamics are greatly influenced by set-point, i.e., the same gain may exhibit different feedback stability under different amplitude set-point. Because the gains do not operate independently, the process of gain optimization is particularly complicated.
Stable feedback also requires applying appropriate gain when a deviation in the oscillation from the set-point is detected. The gain must be adjusted to return oscillation back to the setpoint. P and I gain are adjusted with the user typically employing trial and error to make sure the feedback remains stable. And because the gains do not operate independently, the challenge is particularly complicated.
In response to the desire in the metrology field to have an AFM system that maintains stable feedback with less expert user participation, solutions have been proposed. Nonetheless, each has significant limitations.
In Rifai and Youcef-Toumi, entitled “On automating atomic force microscopes: An adaptive control approach,” as well as in Schitter et al., entitled “Fast contact-mode atomic force microscopy on biological specimen by model-based control,” higher order or model-based controllers are employed over a standard P/I controller. Such controllers are difficult to design and are inherently imperfect. Importantly, such controllers require information related to system dynamics prior to operation. Though they can be effective when operating the AFM in contact mode, they typically have difficulty working when the AFM is operated in TappingMode™ given that, as suggested above, system dynamics change with varying set-point.
In Astrom and Hagglund, a standard P/I controller is employed, but the tuning required for stable operation is automated. Astrom and Hagglund employ simple regulators using specifications on phase and amplitude margins. In this approach, the target system is most typically large plants with slow time response. In particular, the time scale of the response is usually minutes to hours. This characteristic is essentially in direct contrast to an AFM system in which response time is milliseconds and the Q of the response is high (low energy dissipation). In other words, automatic tuning of the controller as taught by Astrom and Hagglund (using simple regulators with slow response times) would not work for most AFM applications.
In another system, disclosed in Rice et al. (U.S. Pat. No. 7,513,142), the system works to detect the onset of instability, and then makes a correction. However, because the time period between the onset of instability and out of control instability (i.e., instability of a magnitude requiring stopping and restarting the measurement process) is so short, it is difficult to implement control before having to stop the measurement process. As understood in the art, hysteresis is primarily responsible when the system is not able to respond quickly enough. Moreover, in this solution the system makes a judgment based on the measured oscillation. An acceptable noise amplitude is defined, and if that amplitude is exceeded, the system adjusts the gain. One main issue concerns the fact that the noise amplitude is so complicated, particularly when operating the AFM in TappingMode™, and when measuring certain types of samples. In TappingMode™ AFM, the oscillation is a non-linear representation of the interaction force between the tip and sample. Therefore, controlling the tapping amplitude, for instance, provides an indirect control of the tip-sample interaction force. This indirect control of the interaction force is susceptible to the effects of variables such as oscillation harmonics and system oscillation, including from the piezo actuator itself and the mechanical components of the AFM. It is these TappingMode™ dynamics that make it extremely difficult to develop a robust control algorithm, particularly when imaging may occur in varying environments.
As a result, though this system does not require user input to make a judgment, its ability to decipher the measured oscillation and modify the control when the system is about to become unstable is limited. Again, in TappingMode™ AFM, system dynamics depend on both set-point (e.g., amplitude or phase) and gain, which severely complicate the ability to develop a control algorithm that can accommodate instabilities.
In sum, while past attempts have been made with AFMs to automatically adjust gain, this method also has not proven particularly effective. Known methods may not be able to handle both sample topography and operating parameters, such as setpoint, actuator hysteresis and tip shape, which can unpredictably and adversely impact any attempt to maintain stability through gain adjustment. As a result, automatic gain adjustment is largely ineffective.
Again, this is not surprising in view of the numerous scan parameters that must be taken into account in AFM setup and operation, along with those that can require adjustment during AFM operation. For example, a user may need to adjust such scan control parameters as setpoint, scan speed, proportional gain, integral gain, drive frequency, drive amplitude, and other parameters. Without great care, considerable experience, and sometimes a little luck, tip, cantilever or sample damage can occur, poor or unusable results can be obtained, and, in instances where everything appears to be operating well, operational inefficiencies can be so great that scanning time is nowhere near optimal, which is particularly problematic for high throughput applications such as those in the semiconductor industry.
At present, if the value of any one of the several manually selected control parameters is not at or within a reasonable range of its optimum, poor performance and unacceptable data will likely result. In addition, relatively complex interdependencies existing between certain AFM parameters often make setup a trial and error procedure, even for the most experienced AFM operators.
In performing AFM setup, the values for several control parameters must be set along with feedback loop gains for different operational modes and other instances where setting up such gains is required. Setup must take into account and configure for parameters such as scan size, pixels per line, number of scan lines, scan rate, tip scanning speed, digital-to-analog (D/A) resolution, Z-center position, i.e., Z-center voltage or the center of the Z piezo operation range, tip wear control, and sample damage minimization.
When an AFM is set-up to operate in an oscillatory mode, such as TappingMode™, setup must include choosing an amplitude and setpoint associated with the oscillation. Moreover, initial values for integral gain (I-gain), and proportional gain (P-gain) are also manually set. Selecting gain values can be tricky because it typically depends on factors such as the nature of the oscillatory mode being employed, sample topography, the hardness and/or roughness or any other mechanical characteristics of the sample and medium in which it is located, as well as other factors. For example, where gain is set too low, system response tends to be relatively slow, which can result in the tip not tracking the sample surface. Where gain is set too high, the feedback loop can start oscillating or backfeeding upon itself, which can undesirably add considerable noise to the sample image being generated.
In addition, the gain setup may be fine initially, only to be unsuitable later once some other factor, such as topography, changes. For instance, where the sample is relatively rough, gain typically should be set higher in order to image such high featured topography with any resulting increase in feedback oscillation noise being tolerable. Where the sample is relatively smooth or flat, gain should be set lower to minimize noise. By keeping noise low with low gain, better resolution of flat areas is achieved, thereby enabling the AFM to better image its finer details. However, as understood in the field, excessive noise can adversely affect imaging along flatter areas of the sample where an initially high gain setting ends up being too high when the sample flattens out. Conversely, an initial low gain setting frequently impedes imaging of higher features of the sample producing an image with such higher features being either distorted or missing.
These setup considerations become even more problematic when operating in TappingMode™ because the highest useable gains typically depend on cantilever dynamics. Cantilever dynamics are a function of the free air tapping amplitude and set-point and thus tuning the gains is very difficult, especially for the novice user. Indeed, factors such as cantilever dynamics and Z-actuator response speed can create such difficulty in setting the initial setpoint and gains, the operator often resorts to trial and error until the sample image starts to look good.
Unfortunately, because one can affect the other, trial and error can go on for a long time. For example, as setpoint is lowered, gain can be set higher and vice versa. However, while lower gains may permit a lower setpoint to be used, which typically increases cantilever response, it also increases error generation rate, which can undesirably blur or otherwise distort the image being produced during scanning.
In the end, what often results is the operator setting some initial parameter values, gains and setpoint and then manually adjusting the value of each, one-by-one until feedback oscillation occurs and then backs off. While this process may work reasonably well for experienced AFM operators, it is inefficient, time consuming, and quite often, less than optimal. In addition, it does nothing to address the dynamic nature of AFM imaging, which often requires an operator to either change certain settings on the fly during operation or to observe the image, etc., and go back and re-scan those parts of the sample that are poorly imaged with adjusted parameter values. Once again, this process can be extremely slow.
As a result, the field of scanning probe microscopy was in need of what one might call a “point and shoot” solution for imaging and mechanical property measurement on a wide array of samples that preferably is easy to use, as well as capable of minimizing the forces generated by tip-sample interaction while also maintaining fast imaging speeds.
Moreover, given the limitations of popular AFM modes, including TappingMode™ in which the output is averaged, thus making certain physical property measurements impossible (or at the very least very limited using small tapping amplitudes and only flat samples to allow the tip to operate in a very narrow interaction region), a solution was desired that could provide the ability to measure a variety of physical properties over a wide range of samples.
The preferred embodiments take advantage of the new mode of AFM operation known as Peak Force Tapping Mode®, in its design of a control scheme that minimizes the need for a skilled and experienced user. PFT Mode essentially eliminates the need for the user to tune the gain while imaging. Moreover, PFT Mode enables further ease of use of an AFM by providing the ability to automatically control operating parameters such as the setpoint, Z-limit and scan rate. In this case, in particular, physical property measurements of a sample can be performed over a wide range of samples, using a variety of probe/sample excitation techniques (electrical, electromagnetic, optical, etc. excitation).
Fundamentally, the preferred embodiments are directed to an AFM that limits the need for an expert user and are realized by employing PFT Mode which operates to move the tip substantially perpendicularly to the sample surface to cause the tip to interact with the sample, and then depart from the sample. The feedback circuit uses instantaneous interaction force (e.g., substantially orthogonal to the sample surface) at any interaction point, preferably using the maximum repulsive force. This new mode of operation takes advantage of the instantaneous response of the probe upon tip-sample interaction (no need to wait for ringdown like prior techniques, the present technique determines a baseline or zero force reference and forcefully substantially instantaneously brings the tip back to the surface), using the feedback loop to maintain a steady state interaction, and to control tracking of the tip on the sample. By moving the tip perpendicularly to the sample surface, this mode shares the advantages of TappingMode™ AFM to at least substantially eliminate friction forces during raster scanning or other relative probe sample motion in the XY plane. In addition, the implementation of this mode minimizes parasitic coupling so that a far more sensitive force control than PFM and TappingMode™ AFM can be accomplished (at least three (3) orders magnitude). In doing so, the lowest force imaging (using alternating force) known in the AFM art is realized and directly controlled, thus allowing the AFM to provide improved high resolution images exceeding TappingMode™ AFM at speeds exceeding typical TappingMode™ AFM speeds (TappingMode™ bandwidth is below 1 kHz).
An added benefit of PFT mode is that each cycle of the vertical movement produces a force curve, or multiple force curves at each pixel, allowing simultaneous acquisition and mapping of height and mechanical property data. This method is therefore called Peak Force Tapping® (PFT) mode since it generates and analyzes each and every individual force curve, then measures and controls the AFM based on the corresponding peak interaction forces during each instance of the tip tapping on the sample, with imaging speed higher than TappingMode™ imaging speed.
In accordance with a first aspect of the invention, a method of operating a SPM includes generating relative motion between a probe and a sample and detecting motion of the probe. The method recovers, from the detected probe motion, a probe-sample interaction that is substantially independent of parasitic probe deflection (i.e., parasitic cantilever motion).
In another aspect of the invention, a method of operating a SPM includes generating an image while maintaining a maximum repulsive probe-sample interaction force of no more than about 10 pN during each cycle of substantially perpendicular cyclical movement of the tip relative to the sample. Such interaction force can be directly controlled and accurately calibrated.
According to another aspect of the invention, a method of operating an SPM includes generating an image for at least 1 hour with peak force of less than 5 nN, without user intervention, while maintaining an image resolution better than 5 nanometers regardless of environment, including ambient, gaseous, fluid and vacuum.
In another aspect of the invention, a method of operating an SPM includes generating at least one force-distance curve for each imaging pixel. The force-distance curve can be used to produce accurate measurement of one or more of Van der Waals adhesion, elasticity, work of adhesion of tip-sample interface, plasticity such as hardness and viscoelasticity.
According to another aspect of the invention, the Peak Force Tapping® method of operating an SPM includes using cantilevers with spring constants equal to about 0.01 N/m to 1000 N/m (which can enable the capability to map mechanical properties over a range from about 10 kPa to 100 GPa). This range of applicable cantilevers is several orders of magnitude wider than cantilevers generally applicable to ContactMode AFM (0.01-1 N/m) and TappingMode™ AFM (1 N/m-40 N/m).
A SPM configured in accordance with the invention could be used to scan a wide variety of samples, including patterned wafers, biological samples in ambient and fluid, polymers, thin films, and data storage device component.
According to a further aspect of the invention, a method of operating a SPM includes interacting a tip of a probe with a sample, then terminating the interaction, resulting in a decaying probe oscillation. Thereafter, the method repeats the interaction before ringdown of the decaying probe oscillation is substantially complete, and detects the motion of the probe.
In another aspect of the invention, a method of operating a scanning probe microscope (SPM) includes generating relative motion between a probe and a sample, and then detecting motion of the probe. In addition, the method includes recovering, from the detected probe motion, a substantially instantaneous force between the tip and sample. Preferably, the method also automatically controls the generating step to maintain a feedback setpoint.
In another aspect of the invention the control loop controls the interaction force at a pre-determined synchronization distance. Synchronization distance is defined as the time from the start of a modulation period to the time corresponding to the point chosen to control feedback. The instantaneous force occurring at this time point is used as the feedback control parameter, usually chosen as the point where the peak repulsive force occurs.
Again, TappingMode™ is complicated by a) indirect force control and b) cantilever resonance dynamics of multiple harmonics modes. Another major drawback is that neither amplitude nor phase of the probe oscillation during data acquisition has a monotonous relationship with tip-sample interaction force. As a result of these complications, subjective judgment must be employed in the feedback optimization process to acquire a desired image, which often means that the user must be an AFM expert to achieve a high quality image, with minimal interaction force, and with the best stabilized (most robust) feedback. The synchronized peak force control of the preferred embodiments (PFT Mode) eliminates the complications due to cantilever dynamics, as well as the complications induced by cantilever resonance and its harmonics. Also, for the first time, PFT Mode uses interaction force directly as the feedback control parameter. Even in contact mode AFM, constant drift of the cantilever deflection due to thermal or other system factors makes accurate force control impossible. In Peak Force Tapping® Mode, the system re-establishes the non-interacting baseline by moving the probe far from the sample in each interaction period. This process allows accurate determination of the interaction force every time the probe interacts with the sample. Through direct force control and elimination of the complications due to cantilever dynamics, the criteria required to achieve the highest quality images became monotonous. As a result, automation of the control loop can be implemented by designing an appropriate computer program. The subjective judgment of an expert user, based on her past experience of, for example, imaging a similar sample, to optimize feedback performance is also eliminated.
In yet another aspect of the invention, the automatically controlling step includes automatically determining a minimum interaction force required for control based on the noise background of the system. It is this minimum interaction force that can be used as the set-point in the control feedback loop.
In yet another aspect of the invention, the automatically controlling step includes determining feedback instability within less than 5 tip-sample interaction periods (for example, 2.5 ms), about 100 times faster than an expert's visual judgment.
In yet another aspect of the invention, the automatically controlling step includes automatically controlling a gain in a corresponding feedback loop.
In a further aspect of the invention, the method includes automatic Z limit control, and preferably automatic scan rate control.
These and other features and advantages of the invention will become apparent to those skilled in the art from the following detailed description and the accompanying drawings. It should be understood, however, that the detailed description and specific examples, while indicating preferred embodiments of the present invention, are given by way of illustration and not of limitation. Many changes and modifications may be made within the scope of the present invention without departing from the spirit thereof, and the invention includes all such modifications.
Preferred exemplary embodiments of the invention are illustrated in the accompanying drawings in which like reference numerals represent like parts throughout, and in which:
The preferred embodiments are directed to a Peak Force Tapping® (PFT) Mode of AFM operation in which the interaction force between the probe (tip) and sample is monitored and used to control tip-sample separation at very low forces, yet without compromising scanning speed. The techniques described herein provide high resolution by maintaining probe tip-sample forces low, and realize essentially real-time property mapping of sample surfaces. The preferred embodiments are inherently stable and thus facilitate long-term force control while maintaining the ability to acquire high integrity data (improved resolution). Moreover, because tuning is not required, unlike conventional TappingMode™ AFM, the AFM setup is faster and easier than with other AFM modes. The key concepts driving the PFT Mode are illustrated graphically and discussed herein.
Practically, there were three major issues to be resolved before AFM control using instantaneous interaction force could be implemented. These issues were 1) accommodation of deflection background due to coupling; 2) determination of a baseline; and 3) determination of the instantaneous force, as defined herein.
In
With further reference to
It is important to differentiate cantilever deflection and tip-sample interaction force. While cantilever deflection is used to gauge the tip-sample interaction force, not all the deflection represents tip-sample interaction force; namely, parasitic forces contribute to cantilever deflection. For example, as shown in
In known force control systems, the control is based on a maximum force occurring in a period. Hence the repulsive force must be higher than any of the parasitic contributions to deflection for true tip-sample interaction to be differentiated from parasitic forces and historically used by the feedback loop. This force differentiation requirement required a relatively high imaging force that could damage the tip and/or the sample, thereby preventing the system from achieving high resolution.
In a preferred embodiment, the RMS or constant deflection is replaced by an instantaneous interaction force Fr_i, determined according to
δFr=Fr_i−Fbaseline Equation (1)
Fbaseline is the interaction force when the probe is not contacting the sample. It should be zero. In AFM, the force is usually represented by cantilever deflection. In this case, Fbaseline corresponds to the cantilever deflection when the tip is not interacting with the surface. Fr_i is the interaction force when the tip is at close proximate contact with the surface. A synchronization algorithm is used to align the start time of each drive period, so that the region δT (
The synchronizing distance or Sync Distance can be precisely controlled. For example, if the tip oscillation period T is 100 μs, when the synchronizing distance is 48 μs, the interaction force occurring at the 48th μs will be used as the feedback control parameter. The feedback loop will try to maintain an instantaneous interaction force Fr_i (i=48 μs) at the 48th μs from the start of the period. In more general applications, any point of interaction force within the interaction region δT can be used for feedback. δT can also extends beyond the marked region in
To achieve an accurate measurement of the baseline, multiple deflection data points are gathered when the tip is not interacting with the sample and used to generate an averaged baseline level. Again, the non-interaction region (greatest separation/highest distance) can be determined by the Sync Distance because this region should be around the half cycle of the modulation period after the peak force position. The Sync Distance also determines the feedback force operating point, and the actual force is measured by δFr. δFr can be either negative or positive.
Due to adverse affects of drift (e.g., thermal) on the deflection signal, the corresponding force Fr_i may vary over time. The relative force δFr (relative to baseline determination) preferably is used for feedback control instead of Fr_i because it is a more accurate reflection of tip-surface interaction. This relative value removes the adverse influences due to system drift on cantilever deflection.
δFr also represents a controllable force by the feedback loop such that δFr remains constant over time at various positions as the tip scans across the sample.
In
More particularly,
Turning to
It becomes clear that the minimum controllable force employable when performing parasitic background subtraction is lessened greatly (by, for example, three (3) orders of magnitude), allowing the preferred embodiments to control tip-sample separation so the probe-sample interaction forces are reduced to the pN range. The way in which this subtraction may be accomplished in the hardware is described further below with respect to
Overall, it is primarily this ability to detect such small forces, and to use such forces as a control parameter in an SPM feedback loop, that allows an SPM operating according to the present invention to image a sample using what is referred to herein as “instantaneous force control.” Instantaneous force control using real-time force detection offers improved control, thus improving image resolution and minimizing the chance for sample damage. In this context, real-time or instantaneous force detection implies that essentially each point of the varying force illustrated, for example, in
Yet another benefit in the peak force tapping control is that it does not need to be operated near the cantilever resonance frequency. Such operation can substantially eliminate cantilever delay due to transient resonance response, rendering instantaneous interaction control possible.
Turning next to
For measurement with very sensitive force detection, very soft cantilevers (spring constant 0.01 N/m to 0.3 N/m) are typically used. These levers have lower resonance frequency and very long ringdown time. More importantly, the adhesion induced oscillation (snap out of contact) is much stronger, as shown in
Steady state again means a constant maximum force or a constant minimum force, or a combination of the characteristics of the interaction force curve shape in each cycle of the probe/sample relative motion.
Another major advantage of the present techniques is the ability to determine the baseline with high amplitude oscillatory data. Since the resonance frequency of the cantilever is known, in an alternative embodiment, the average can be determined in the non-interacting zone by analyzing an integer multiple of cycles of the cantilever resonance frequency. The integer cycle averaging can effectively remove the oscillatory deflection data, yielding a constant baseline.
Notably, cantilever resonance frequency can also be determined by known techniques such as frequency sweep and thermal tune.
Turning next to
Turning to
Taking the above further, as shown in
Where Di representing data in the ith cycle. The averaged signal with a signal to noise ratio improved by a factor of √N, thereby reducing the minimum controllable force (can use narrow lock-in bandwidth), is shown on
Turning next to
During operation, probe deflection is measured by bouncing a light beam “L” off the back of the probe and toward a detector 114, such as a four quadrant photodetector. The deflection signal is then transmitted to an analog to digital converter 103. The digitized signal is used for maintaining the tip-sample force low while operating the AFM at high speed.
In the embodiment shown in
Sync Distance calculator 135 determines the phase shift between the deflection and the Z modulation DDS (Block 127) that provides the drive and synchronization control in the form of a time delay. Peak force or repulsive force gate position generator 129 generates the timing signal for force detector 126, with the help of the synchronization marker and synchronization time distance. Force detector 126 analyzes the output of summation circuit 125 by either identifying the repulsive peak force or averaged repulsive force within the gated region illustrated in
Turning to
Motion of the probe is then detected; in particular, probe deflection is detected and transmitted to the converter for further processing. In Block 306, the method then operates to recover probe-sample interaction as described above, preferably performing hydrodynamic background subtraction using either lock-in amplification, or more preferably, synchronous averaging of the deflection. After filtering the output in Block 308 (e.g., selecting a number of cycles of ringdown to process), the method detects the force (peak force detection/gated averaging), preferably using the repulsive region of the force curve, in Block 310. In Block 312, the force is then compared to the setpoint force, set according to the user's desired interaction force. The Z-actuator responds to the control signals in Block 316 to adjust tip-sample separation and maintain the setpoint force, with the control signals being used to generate an image of the sample.
Turning to
To the contrary, as shown in
Algorithms
To assure accurate subtraction of the background, two schemes have been developed, as shown in
In
In
The background calculated according to
Background subtractions are preferably also executed during engagement of the probe with the sample surface, as shown in
The difference between the two engage methods is that the “normal” engage 600 in
In
The instantaneous interaction force can be determined by using the force δFr calculated by Equation (1), in which Fr_i can be an instant value at the Sync Distance. As illustrated in
Feedback is then used to maintain δFr and Fr_i at the preset value in Block 812. The XY scanner is also enabled, Block 814, to reposition the probe relative to the sample and eventually generate a topographic image, as well as one or more mechanical images indicative of, for example, elasticity, adhesion, and energy dissipation.
In
One important application of the instantaneous force controlled imaging is in deep trench measurement. When TappingMode™ AFM is used to image deep trenches (aspect ratio of about 3:1 or more, with the most difficult trenches to image having sub-100 nm width, typically 10 nm-100 nm) the strong attractive force at the side walls can cause amplitude change, resulting in a false measurement of the trench depth. Using direct repulsive force as feedback, the feedback only responds to z-change when the tip is in contact with the sample. As a result, the force controlled feedback can measure deep trenches much more reliably than TappingMode™ AFM.
Referring finally to
In
Advantages—PFT Mode
In sum, the benefits of PFT Mode AFM operation are numerous. Given the inherently stable long term force control, drift-free sample imaging can be achieved along with simultaneous height, stiffness, adhesion, elasticity and plasticity mechanical property measurements at TappingMode™ speeds. Because the technique is not impacted by DC drift (PFT mode creates its own reference every few hundred microseconds), steady operation is not compromised even without an expert operator. This allows the SPM to run for hours, even days (large samples-long time) without substantially compromising image integrity, particularly useful for in-process measurements, like crystal growth and monitoring polymer phase change, which can take several minutes or hours. Notably, a Peak Force Tapping® image can be generated at an operating bandwidth greater than 2 kHz. TappingMode™ bandwidth is about 1 kHz, primarily because cantilever dynamics control speed, e.g., it takes at least several milliseconds to stabilize to return to resonance (greater the amplitude error, the slower). The disclosed embodiments can also eliminate phase interpretation problems because it independently measures elasticity, adhesion, energy dissipation, etc. All these factors contribute to the phase of cantilever oscillation.
Moreover, PFT Mode is insensitive to cantilever dynamics because there is no need to wait for complete cantilever ringdown once the probe releases from the sample. This allows for high speed imaging in vacuum and also allows for arbitrary choice among cantilever options. This difference allows mapping over several orders of magnitude of interaction force, while repulsive force resolution can be used to produce artifact free cellular imaging.
The fact that PFT Mode does not have to operate at the resonance frequency of the probe offers a major advantage when imaging in fluid. Due to various parasitic coupling forces in fluid, cantilever tuning is a key issue in obtaining a fluid image. PFT Mode completely removes the need to tune the cantilever (baseline averaging, background subtraction, etc.). Furthermore, the range of force control and the ability to use a cantilever having a much wider spring constant range (typically, greater than 0.3 N/m for TappingMode™ AFM only, while PFT Mode can use cantilevers having spring constants as low as 0.01 N/m) gives imaging control much more room for biological sample imaging.
Again, this is due to the fact that PFT Mode does not depend on the oscillation energy stored in the cantilever to overcome capillary adhesion forces. Because the technique utilizes an external actuation element (of the feedback circuit, preferably triggering on peak force), the mechanism to overcome the capillary forces is far more powerful than in TappingMode™ wherein the static elastic energy of the cantilever itself (fed by the kinetic energy of the oscillating probe) pulls the tip away from the sample in overcoming the capillary forces. As a result, there is virtually no limitation on the cantilever spring constant to operate stably in presence of a capillary layer. PFT Mode therefore enables stable tapping control operation using a cantilever having a spring constant at least as low as 0.01 N/m.
Yet another benefit of the peak force tapping control is the ability to use cantilevers from 0.01 N/m to 1000 N/m in one mode of AFM operation. It enables high resolution mechanical property mapping of the broadest range of materials on a single instrument from 10 kPa to 100 GPa in elastic modulus.
In addition, given essentially instantaneous force feedback, tip crashing is virtually eliminated. Also, because the deflection is hydrodynamically corrected, no tuning is typically required, and therefore fast, ready setup by virtually any user can be accomplished.
When compared to existing modes of AFM operation, the low force high speed imaging provided by PFT Mode in combination with the low average tracking force and the virtual elimination of lateral forces on the tip, provide a significant advance in high speed imaging over a wide variety of samples. For example, single molecule elasticity can be measured, as well as narrow DNA samples in fluid (e.g., 2 nm wide DNA). By comparison, when imaging DNA in fluid, TappingMode™ AFM has at least a 2 nm lower resolution. Moreover, measuring DNA stiffness in fluid is challenging with TappingMode™ AFM because it does not have property quantification capacity, it primarily is only able to provide relative mechanical property measurements (for example, by looking at contrast in phase images). With the present technique, property measuring down to the molecular level can be achieved.
In the end, PFT Mode can acquire data as good as or better (a resolution [e.g., less than a 100 nm, and more preferably less than about 1 nm laterally], etc.) than that acquired in TappingMode™ AFM without damaging the tip and/or the sample. The technique provides significant speed improvement over other known force feedback techniques, and does so without requiring the use of a small lever. In fact, a rather large lever (>60 μm long) can be operated at sub-resonance in PFT Mode so that the lever response has a bandwidth far beyond that achievable when using a so-called small cantilever (>10 kHz).
Of course, an additional benefit of the present preferred embodiments is that a force curve is generated with every pixel so that the image provides information beyond a typical TappingMode™ AFM image. With every pixel, the user can obtain quantitative information regarding stiffness, adhesion, elasticity, plasticity, etc. And again, because baseline tip-sample separation is re-zeroed with every pixel, drift is minimized so that a large improvement in productivity and image reliability is realized.
In review, the present PFT Mode provides very low force imaging to provide very high resolution using real time property mapping (i.e., instantaneous force control). The force control is inherently stable (essentially drift free), over a term sufficiently long to image a sample with minimal or no user intervention. The system allows faster, simpler set-up because no tuning is required (baseline averaging and hydrodynamic background correction). Moreover, precise control over force basically eliminates tip crash, while the technique/system also essentially eliminates lateral force on the sample surface. The system is also insensitive to cantilever dynamics by not having to wait for probe ringdown before interacting the probe with the sample once it releases from the sample. And, as discussed, a wide range of cantilevers are available to the user to obtain simultaneous measurements of height, stiffness, adhesion, elasticity and plasticity at TappingMode™ AFM speeds (>2 kHz). The present SPM can image samples such as 2 nm wide DNA in fluid with these characteristics, as well make improved mechanical property measurements such as single molecule elasticity.
PFT Mode—Ease of Use
The preferred embodiments of the present invention use PFT Mode to facilitate use of an AFM by a novice substantially on par with that of an expert user. In contrast to TappingMode™ AFM, which operates by controlling tip-sample interaction based on deviations from, for example, a setpoint amplitude or phase of probe oscillation as the tip interacts with the sample (representing a complex relationship relative to tip-sample forces), PFT Mode controls tip-sample interaction based on tip-sample interaction forces at each point along a cycle of probe modulation in oscillating mode. This direct control of the interaction force simplifies the control and allows the preferred embodiments to minimize the effects of complicating variables, including oscillation harmonics and system oscillation, and thus maintain stability.
Turning to
An AFM 1100 operable in PFT Mode to minimize the skill required to operate the AFM is shown schematically in
To facilitate stability, and thus minimize the need for an expert user, the gain is automatically tuned using a gain control circuit 1123. The control signal from PI Controller 1124 used to control the Z piezo 1110 is also transmitted to a block 1128 that re-samples the height data at a position corresponding to, preferably, the peak force (see block 1120). An oscillation detection algorithm 1130 is then employed to determine whether there is oscillation in the height data, i.e., whether there is an onset of instability. If the system is about to oscillate and become unstable, high frequency noise will be detected. The way in which algorithm 1130 determines the amount of noise is described in further detail below in connection with
In operation, the Z Scan control signal output by DAC 1126, and optimized by gain control circuit 1123, is combined with the output of the Z offset DAC 1136 (described further below) and the oscillating drive for PFT Mode provided by a Z modulation DDS (direct digital synthesizer) 1138 at summing circuit 1139. With this automatic gain scheduling active, the need of expert user tuning of the gains during AFM operation is eliminated.
One of the critical elements in automated adjustment of feedback gains is the ability to determine instability onset quickly and accurately. This determination is often complicated by unknown topography which may be misinterpreted as the instability induced noise in the gain controller. Turning to
In this regard, referring back to
H Diff(i)=(H(i−1)+H(i+1)−2*H(i))/2 Equation 3
The absolute value of this difference |H Diff(i)| which ideally does not reflect sample topology (and should not), is obtained in Block 1146. This step operates essentially like an oscillation detector. Then, in Block 1148, a moving average may be determined. Determining a moving average is only required for those samples that exhibit significant changes in topology such that the topology might not be filtered out in a given sample used for the height difference calculation. Such samples include, for example, a silicon grating with sharp steps. In such cases, the rapid change in topology is typically short lived. Therefore, by determining a moving average of height differences, which are computed over relatively long periods of time, such spikes should be filtered from the oscillation analysis. More particularly, by comparing the amplitude of the spikes with the moving average data, the spikes are filtered out. And because problematic oscillation noise typically lasts much longer than topology changes, the associated amplitude data tends to be similar to previous moving averaged data.
Continuing with method 1140, in Block 1149, if the absolute value of the difference obtained in Block 1136 is less than some multiple of the moving average, for example, four (4) times the moving average value computed in Block 1148, the output of oscillation detection algorithm 1140 is |H Diff(i)|. If the absolute value of the difference is greater than the multiple, then the output of algorithm 1140 is the moving average value. The RMS value of this quantity is then determined in Block 1150. It is this value that is compared to the “Noise Tolerance Margin” by summing circuit 1152, described above in connection with
A particular implementation of AFM operation using PFT Mode is illustrated in
Once the parameters are initialized, scan size is set to a small value in Block 1514. A small scan (10 nm, for example) is performed at low gain to determine an initial peak force setpoint and gain to provide a setpoint reference. For all AFM imaging, minimizing the peak tip-sample interaction force generally leads to improved tip life and sample integrity. The system can determine the minimum set-point based on the knowledge of the base noise in the system. For example, if the force detection noise, when the tip is not interacting with the sample, is 100 pN, the setpoint may be set at 300 pN, allowing enough S/N range for system control. In Block 1516, the engage is verified, and in Block 1518, the system modifies the initial gain and setpoint in an attempt to optimize the same. The optimization is an iterative process including:
Once the gain and force set-point are determined at small scan size in Block 1520, the system restores the user-input scan size in Block 1522 and begins AFM operation to acquire sample data.
In Block 1524, the system determines whether the algorithm is adjusting the gain or setpoint. If either gain or set-point is not being adjusted by the algorithm, the default gain/set-point value is restored in Block 1526. The system then enters a monitoring loop (Monitoring Mode) in Block 1528. Monitoring Mode determines whether oscillation exceeds the threshold. If so, gain can be adjusted (decreased). If not, gain can be increased for better tracking. Monitoring Mode also operates to detect a parachuting event. If a parachuting event is detected as described above, the setpoint may be increased for optimal performance. Setpoint increase is implemented, preferably, by 5% increments each time (and optionally verifying steps 1-3 outlined above). The above continues until the scan of the user-defined sample scan size is complete.
In sum, the above-described feedback control is able to maintain a substantially identical peak interaction force in each modulation period of probe oscillation/tip-sample interaction. The method automatically determines a setpoint associated with the peak interaction force based on noise background, and automatically determines feedback gain according to an oscillation magnitude of the instability. By doing so, the AFM can be used by a novice without making system adjustments during data acquisition.
The scan rate may also be automatically adjusted for optimization using PFT Mode. Turning to
PFT Mode also enables automatic Z limit control, further facilitating ease of use of this AFM. The Z limit parameter defines the dynamic range of the Z piezo actuator. Preferably, the probe is centered in this range. Z limit influences image resolution in Z, and is sample dependent. For example, if the sample is flat, then the noise amplitude is comparable to peak resolution, and resolution therefore should be reduced to minimize the chance that noise appears in the acquired image. Previously, this was done manually by the user by determining whether or not the sample was flat. In PFT Mode, control of the Z limit parameter is automated. In this regard, turning to
In addition to automation, PFT Mode is useful for maximizing the ability to insure quality imaging and obtain mechanical property measurements of the sample, at every scan location (e.g., pixel) of the sample. For instance, PFT Mode can be used to perform tip radius monitoring. One major setback to obtaining high quality images is the difficulty of the user to detect when the sharp probe tip has been compromised. The tip may be compromised by being contaminated (imaging in fluid, imaging an oily sample, etc.), or the physical structure may be compromised during imaging, due to wear (dull tip), for example. A compromised tip can be identified by reviewing a force curve obtained at a sample location.
An indication of tip health is contamination. Such contamination is determined by analyzing the shaded area “w” in schematic plot 1803 in
If the tip is functionalized through certain chemical bonds, such as Poly(ethylene glycol) (PEG) or dendron, the work of adhesion is then purposely introduced. In this case, the functionalized bonds only generate significant work of adhesion when the tip interacts with samples that exhibit particular interaction, at the molecular sites, that generates a bond, for example, to Poly(ethylene glycol) (PEG) or dendron. By monitoring this interaction, the adhesion map may become a chemical or biochemical recognition map.
One can also apply electric, optical, magnetic or thermal perturbation or excitation that is synchronized to the contact point D in schematic graph 1802 of
Advantages—PFT Mode and Ease of Use
In sum, PFT Mode provides several operational advantages that enable AFMs to be operated by non-expert users. When considering ease of use, several imaging factors must be accounted for to minimize the need for an expert user. First, stability of the feedback must be maintained and, with the above-described automatic gain tuning/scheduling enabled by PFT Mode, stability is realized without any expert being present to manually adjust the gains. Next, to obtain quality images, the AFM must track the sample surface. By basing control on the instantaneous tip-sample interaction force, the setpoint force can be selected for optimum tracking with minimum error. Scan rate and auto Z-Limit control, as described above, also work to minimize the need for an expert when operating the AFM without compromising imaging speed or the ability to obtain high quality images.
In contrast to known oscillatory modes of AFM operation such as TappingMode™, PFT Mode operates in an entirely different dynamic regime. Oscillation mode setpoint is, typically, an amplitude or phase of the oscillation, a parameter that has a highly complex relationship with the interaction and the force between tip and sample. As discussed herein, PFT Mode considers each point of tip oscillation as the tip interacts with the sample surface and uses the corresponding force information in its feedback scheme. This allows the preferred embodiments to operate without user controlled feedback, with no user adjustments being required during imaging (auto-minimization of the error signal). PFT Mode also provides intermittent contact with the sample (and its understood benefits) with tuning (only requiring a simple pre-image routine—
Moreover, by providing a force curve at every pixel, the user is able to obtain deterministic data (e.g., adhesion) at a reasonable speed and at a certain resolution, and can do so while imaging. This is all enabled by feeding back directly on force (tip-sample), which allows responses based on a single interaction between the tip and sample (representing a linear transfer function—direct contrast to known oscillating modes).
Notably, all the above concepts can be employed in the electrical context as well (e.g., STM) whereby the instrument feeds back on current.
Also, because of the complex nature of the feedback, the data obtained in conventional oscillating modes typically requires complex indirect interpretation. PFT Mode allows direct interpretation of the data given that it is force curve based rather than tapping “envelope” based.
Another benefit of operating in PFT Mode includes the ability to image certain samples more effectively. For instance, in semiconductor applications, the inability of AFM to reliably image narrow trenches often causes users wanting to perform such measurements to select metrology instruments other than AFM. However, in PFT Mode there is no damping between the tip and sample (e.g., squeeze film damping), thus enabling confident measurement of high aspect ratio sample features.
In addition, PFT Mode is not subject to control parameter drift. For example, TappingMode™ AFM free air amplitude may change during imaging, causing change in the tip/sample force, and may result in loss of tip/sample interaction. Such drift prevents TappingMode™ AFM to perform long time stable imaging. With PFT Mode, a user can image for more than an hour (including overnight) versus less than an hour using conventional oscillating AFM modes.
Overall, in PFT Mode, there is a de-coupling of the cantilever response to environmental conditions. Imaging in vacuum (fluid) and atmosphere can be accomplished without affecting set-up thus making the instrument very easy to use. The oscillation frequency can be set independent of any cantilever resonance—greatly simplifying use in fluid. In particular, known intermittent contact modes require operation at resonance, while PFT Mode preferably operates at sub-resonance. This, again, is due to the ability to control based on ultra small instantaneous (not average) forces (about 1 μN to 1 pN). As a result, the AFM can also run feedback faster given that cantilever Q is irrelevant at subresonance (the transfer function is independent of the energy store in the cantilever at resonance). Finally, PFT Mode also allows use of cantilevers having sub 1-10 N/m spring constants, as discussed above.
Turning to
Interaction force measurement in
The time zone between p1 and p2 is the proximity interaction zone where the tip is interacting with the sample with detectable negative force (van der Waals) attraction, for example. Other examples include electric or magnetic forces. This region represents long range interaction force, usually exhibiting negative (or attractive) force. The same proximity interaction region occurs again between p4 and p5, where the minimum interaction force includes both, van der Waals force (or electric, magnetic forces), and capillary interaction at the tip of the probe. This region also exhibits negative (attractive) force in most cases.
The time zone between lines p2 and p3 is the contact interaction region. This time is measured by identifying the interaction force that occurs between the two transition points B′ and E. The contact region is usually accompanied by the rise of the interaction force to the more positive value as seen between points B′ and C. At point C, the Z control starts to retract the probe, but the probe remains in contact with the sample until point E, where the tip of the probe breaks free from the contact and capillary interactions.
Turning next to
In an alternative embodiment, the physical excitation signal can be heat applied to the sample or the probe, and the gated physical response can be the temperature change in the gated time zone between p2 and p4. The application of the heat or thermal excitation in
In still further embodiments, the excitation signal in
Notably, the best signal-to-noise ratio can typically be achieved using a gated average of the measured current. However, peak current or the average current in one-cycle can be used.
Turning to
Although the best mode contemplated by the inventors of carrying out the present invention is disclosed above, practice of the above invention is not limited thereto. It will be manifest that various additions, modifications and rearrangements of the features of the present invention may be made without deviating from the spirit and the scope of the underlying inventive concept.
[1] This application is a divisional of U.S. patent application Ser. No. 14/178,076, filed on Feb. 11, 2014 (and issued as U.S. Pat. No. 9,291,640 on Mar. 22, 2016), which in turn is a divisional of U.S. patent application Ser. No. 13/306,867, filed on Nov. 29, 2011 (and issued as U.S. Pat. No. 8,650,660 on Feb. 11, 2014), which in turn, claims priority under 35 USC § 1.119(e) to U.S. Provisional Patent Application Ser. No. 61/417,837, filed on Nov. 29, 2010, each entitled Method and Apparatus of Using Peak Force Tapping Mode to Measure Physical Properties of a Sample. U.S. patent application Ser. No. 13/306,867 is also a continuation-in-part of U.S. patent application Ser. No. 12/618,641, filed on Nov. 13, 2009 (and issued U.S. Pat. No. 8,739,309 on May 27, 2014), entitled Method and Apparatus of Operating a Scanning Probe Microscope, which in turn, claims priority under 35 USC § 1.119(e) to U.S. Provisional Patent Application Ser. No. 61/114,399, filed Nov. 13, 2008. The subject matter of these applications is hereby incorporated by reference in their entirety.
Number | Name | Date | Kind |
---|---|---|---|
4724318 | Bennig | Feb 1988 | A |
RE33387 | Binnig | Oct 1990 | E |
5060248 | Dumoulin | Oct 1991 | A |
5229606 | Elings et al. | Jul 1993 | A |
5266801 | Elings et al. | Nov 1993 | A |
5267471 | Abraham et al. | Dec 1993 | A |
5406832 | Gamble et al. | Apr 1995 | A |
5412980 | Elings et al. | May 1995 | A |
5415027 | Elings et al. | May 1995 | A |
5436448 | Hosaka | Jul 1995 | A |
5507179 | Gamble et al. | Apr 1996 | A |
5513518 | Lindsay | May 1996 | A |
5519212 | Elings et al. | May 1996 | A |
5538898 | Wickramasinghe | Jul 1996 | A |
5753814 | Han | May 1998 | A |
5955660 | Honma | Sep 1999 | A |
6005246 | Kitamura et al. | Dec 1999 | A |
6008489 | Elings et al. | Dec 1999 | A |
RE36488 | Elings et al. | Jan 2000 | E |
6134955 | Han et al. | Oct 2000 | A |
6196061 | Adderton | Mar 2001 | B1 |
6281495 | Kitamura | Aug 2001 | B1 |
6441371 | Ahn et al. | Aug 2002 | B1 |
6519221 | Manalis | Feb 2003 | B1 |
6608307 | Baur | Aug 2003 | B1 |
6690008 | Hantschel | Feb 2004 | B2 |
6880386 | Krotil et al. | Apr 2005 | B1 |
6894272 | Kranz et al. | May 2005 | B2 |
6906450 | Tamyo De Miguel et al. | Jun 2005 | B2 |
6935167 | Sahin | Aug 2005 | B1 |
7044007 | Struckmeier et al. | May 2006 | B2 |
7074340 | Lugstein et al. | Jul 2006 | B2 |
7129486 | Spizig et al. | Oct 2006 | B2 |
7156965 | Li et al. | Jan 2007 | B1 |
7249002 | Ben-Dov et al. | Jul 2007 | B1 |
7395698 | Degertekin | Jul 2008 | B2 |
7441447 | Degertekin | Oct 2008 | B2 |
7448798 | Wang | Nov 2008 | B1 |
7461543 | Degertekin | Dec 2008 | B2 |
7464583 | Kowalewski | Dec 2008 | B1 |
7478552 | Gotthard et al. | Jan 2009 | B2 |
7550963 | Xiang | Jun 2009 | B1 |
7552625 | Degertekin | Jun 2009 | B2 |
7555940 | Su | Jul 2009 | B2 |
7584653 | Su | Sep 2009 | B2 |
7596989 | Humphris et al. | Oct 2009 | B2 |
7617719 | Su | Nov 2009 | B2 |
7637149 | Degertekin | Dec 2009 | B2 |
7665350 | Giessibl | Feb 2010 | B2 |
7707873 | Degertekin | May 2010 | B2 |
7810166 | Struckmeier et al. | Oct 2010 | B2 |
8220318 | Degertekin | Jul 2012 | B2 |
8646109 | Hu et al. | Feb 2014 | B2 |
8650660 | Shi et al. | Feb 2014 | B2 |
20020084410 | Colbert | Jul 2002 | A1 |
20020096642 | Massie | Jul 2002 | A1 |
20020174714 | McWaid | Nov 2002 | A1 |
20040134264 | Massie | Jul 2004 | A1 |
20040134265 | Mancevski | Jul 2004 | A1 |
20050030054 | Chang | Feb 2005 | A1 |
20050212529 | Huang et al. | Sep 2005 | A1 |
20050266586 | Linder | Dec 2005 | A1 |
20060005634 | Schroeder et al. | Mar 2006 | A1 |
20060219905 | Gibson et al. | Oct 2006 | A1 |
20060260388 | Su | Nov 2006 | A1 |
20060283338 | Degertekin | Dec 2006 | A1 |
20070012094 | Degertekin | Jan 2007 | A1 |
20070024295 | Humphris et al. | Feb 2007 | A1 |
20070082459 | Faris | Apr 2007 | A1 |
20070084273 | Hare et al. | Apr 2007 | A1 |
20070089496 | Degertekin | Apr 2007 | A1 |
20070103697 | Degertekin | May 2007 | A1 |
20070107502 | Degertekin | May 2007 | A1 |
20070220958 | Gotthard | Sep 2007 | A1 |
20070295064 | Degertekin | Dec 2007 | A1 |
20080022759 | Su et al. | Jan 2008 | A1 |
20080049223 | Iyoki | Feb 2008 | A1 |
20080049236 | Iyoki | Feb 2008 | A1 |
20080127722 | Su | Jun 2008 | A1 |
20080257024 | Watanabe | Oct 2008 | A1 |
20080277582 | Shi | Nov 2008 | A1 |
20080295583 | Giessibl | Dec 2008 | A1 |
20080295584 | Cantrell | Dec 2008 | A1 |
20080307865 | Degertekin | Dec 2008 | A1 |
20090032706 | Prater | Feb 2009 | A1 |
20090261249 | Kobayashi | Oct 2009 | A1 |
20100039919 | Chou | Feb 2010 | A1 |
20100042356 | Kjoller | Feb 2010 | A1 |
20100045970 | Raschke | Feb 2010 | A1 |
20110170108 | Degertekin | Jul 2011 | A1 |
Number | Date | Country |
---|---|---|
1253291 | May 2000 | CN |
2591559 | Dec 2003 | CN |
H04254930 | Sep 1992 | JP |
H11-133038 | May 1999 | JP |
H11160333 | Jun 1999 | JP |
H11183414 | Jul 1999 | JP |
H11-352135 | Dec 1999 | JP |
2001108601 | Apr 2001 | JP |
2004170281 | Jun 2004 | JP |
2010071871 | Apr 2010 | JP |
20040106699 | Dec 2004 | KR |
2002048644 | Jun 2002 | WO |
2010032429 | Mar 2010 | WO |
2010065131 | Jun 2010 | WO |
Entry |
---|
Kwak et al, “Imaging stretched single DNA molecules by pulsed-force-mode atomic force microscopy”, Scient Direct (www.sciencedirect.com), DOI: 10.1016/S0304-3991(03)00049-4, (2003), pp. 249-255. |
Kruger et al, “Scanning Force Microscopy Based Rapid Force Curve Acquisition on Supported Lipid Bilayers: Experiments and Simulations Using Pulsed Force Mode”, ChemPhysChem (www.chemphyschem.org), DOI: 10.1002/cphc.200301059, (2004), pp. 989-997. |
Kresz et al, “Investigation of pulsed laser deposited crystalline PTFE thin layer with pulsed force mode AFM”, Science Direct (www.sciencedirect.com), DOI: 10.1016/j.tsf.2003.11.254, (2003), pp. 239-244. |
Shanmugham et al, “Polymer Nanowire Elastic Moduli Measured with Digital Pulsed Force Mode AFM”, American Chemical Society, DOI: 10.1021/Ia050538o, Jun. 11, 2005, pp. 10214-10218. |
Zhang et al, “Pulsed-Force-Mode AFM Studies of Polyphenylene Dendrimers on Self-Assembled Monolayers”, The Journal of Physical Chemistry, DOI: 10.1021/jp073388u, May 23, 2007, pp. 8142-8144. |
Jradi et al, “Analysis of photopolymerized acrylic films by AFM in pulsed force mode”, Journal of Microscopy, vol. 229, Pt 1 2008, pp. 151-161. |
Miyatani et al, “Mapping of electrical double-layer force between tip and sample surfaces in water with pulsed-force-mode atomic force microscopy”, American Institute of Physics (http://apl.aip.org/apl/copyright.jsp), vol. 71, No. 18, Nov. 3, 1997, pp. 2632-2634. |
Moreno-Herrero et al, “Characterization by Atomic Force Microscopy of Alzheimer Paired Helical Filaments under Physiological Conditions”, Biophysical Journal, vol. 86, Jan. 2004, pp. 517-525. |
Sotres et al, “Jumping mode AFM Imaging of biomolecules in the repulsive electrical double layer”, Science Direct (www.sciencedirect.com). DOI: 10.1016/j.ultramic.2001.01.020, Jan. 31, 2007, pp. 1207-1212. |
Kwak et al, “Topographic effects on adhesive force mapping of stretched DNA molecules by pulsed-force-mode atomic force microscopy”, Science Direct (www.sciencedirect.com), DOI: 10.1016/j.ultramic.2003.11.005, Nov. 24, 2003, pp. 179-186. |
Moreno-Herrero et al, “DNA height in scanning force microscopy”, Science Direct (www.sciencedirect.com), DOI: 10.1016/S0304-3991(03)00004-4, Nov. 22, 2002, pp. 167-174. |
Moreno-Herrero et al, “Jumping mode atomic force microscopy obtains reproducible images of Alzheimer paired helical filaments in liquids”, Science Direct (www.sciencedirect.com), DOI: 10.1016/j.eurpolymj.2004.01.018, Jan. 20, 2004, pp. 927-932. |
Moreno-Herrero et al, “Atomic force microscopy contact, tapping, and jumping modes for imaging biological samples in liquids”, The American Physical Society, Physical Review E 69, 031915, 2004, pp. 031915-1-031915-9. |
Jiao et al, “Accurate Height and Volume Measurements on Soft Samples with the Atomic Force Microscope”, American Chemical Society, DOI: 10.1021/Ia048650u, Oct. 8, 2004, pp. 10038-10045. |
Ven Der Werf et al, “Adhesion force imaging in air and liquid by adhesion mode atomic force microscopy”, American Institute of Physics, Appl. Phys. Lett. 65 (9), Aug. 29, 1994, pp. 1195-1197. |
Krotil et al, “Pulsed Force Mode: a New Method for the Investigation of Surface Properties”, Surface and Interface Analysis, vol. 27, 1999, pp. 336-340. |
Miyatani et al, “Surface charge mapping of solid surfaces in water by pulsed-force-mode atomic force microscopy”, Applied Physics A Materials Science & Processing, vol. A 66, 1998, pp. S349-S352. |
Maivaldt et al, “Using force modulation to image surface elasticities with the atomic force microscope”, Nanotechnology 2, Feb. 20, 1991, pp. 103-106. |
Marti et al, “Control electronics for atomic force microscopy”, American Institute of Physics, v. Bal Instrum 69 (8), Jun. 1988, No. 6, pp. 836-839. |
Sarid et al, “Driven nonlinear atomic force microscopy cantilevers: From noncontact to tapping modes of operation”, American Vacuum Society, J. Vac. Sci. Technol. B 14(2), Mar./Apr. 1996, pp. 864-867. |
Dr. Profos et al, “Handbuch der industriellen Messtechnik”, R. Oldenbourg Verlag Munchen Wien, Feb. 11, 1992, pp. 203-206. |
Marti et al, “Reibungsmikroskopie”, Aus der Wissenshaft, Phys. B1. 48 Nr. 12, 1992, pp. 1007-1009. |
Rosa et al, “The simultaneous measurement of elastic, electrostatic and adhesive properties by scanning force microscopy: pulsed-force mode operation”, Department of Experimental Physics, University of Ulm, Jul. 29, 1997, pp. 1-6. |
Radmacher et al, “Imaging Adhesion Forces and Elasticity of Lysozyme Adsorbed on Mica with the Atomic Force Microscope”, American Chemical Society, Jul. 11, 1994, pp. 3809-3814. |
Spatz et al, “Forces affecting the substrate in resonant tapping force microscopy”, University of Ulm, May 23, 1995, pp. 40-44. |
Radmacher et al, “Mapping Interaction Forces with the Atomic Force Microscope”, Department of Physics, University of California, Mar. 7, 1994, pp. 2159-2165. |
Martin et al, “Atomic force microscope-force mapping and profiling on a sub 100-A scale”, American Institute of Physics, J. Appl. Phys. 61 (10), May 15, 1987, pp. 4723-4729. |
Mizes et al, “Submicron probe of polymer adhesion with atomic force microscopy: Dependence on topography and material inhomogeneities”, American Institute of Physics, Appl. Phys. Lett. 59 (22), Nov. 25, 1991, pp. 2901-2903. |
Jacobs et al., “Practical Aspects of Kelvin Probe Force Microscopy”, Review of Scientific Instruments, AIP, Melville, NY, vol. 70, No. 3, Mar. 1, 1999, pp. 1756-1760. |
Gannepalli et al., “Thermal Noise Response Based Control of Tip-Sample Separation in AFM”, American Control Conference, 2004, Boston, MA Jun. 30, 2004, pp. 3122-3127. |
Jacobs et al., “Surface Potential Mapping: A Qualitative Material Contrast in SPM”, Ultramicroscopy, vol. 69, No. 1, Aug. 1, 1997, pp. 39-49. |
Devecchio et al., “Use of a Nanoscale Kelvin Probe for Detecting Wear Precursors”, Review of Scientific Instruments, vol. 69, No. 10, Oct. 1, 1998, pp. 3618-3624. |
Number | Date | Country | |
---|---|---|---|
20160274144 A1 | Sep 2016 | US |
Number | Date | Country | |
---|---|---|---|
61417837 | Nov 2010 | US | |
61114399 | Nov 2008 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 14178076 | Feb 2014 | US |
Child | 15077708 | US | |
Parent | 13306867 | Nov 2011 | US |
Child | 14178076 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 12618641 | Nov 2009 | US |
Child | 13306867 | US |