FASHION FOR THE FLIRT MODE
In the laboratory studies using an atomic force microscope, contact and resonance modes are the most widely used allowing acquisition of scans in relatively short times. On the other hand, their disadvantage is the difficulty of controlling the force impact, which can lead to irreversible changes in the morphology of the sample.
This is particularly critical when scanning soft objects such as bacteria, cells, and polymers. In the contact mode this interaction occurs also due to frictional forces which magnitude is difficult to control when scanning the sample surface. In the resonance mode the influence of frictional forces is smoothed out by vertical motion of the probe; however, high quality of the cantilever oscillations does not allow controlling its force at sudden differences in topography of the observed surface. In order to overcome these obstacles, we have developed a mode of delicate and gentle surface scanning, which employs a light touch to the surface only. We have named this mode as flirt mode.
FASHION FOR THE FLIRT MODE
I.V.Yaminsky1, 2, 3, Doct. of Sci. (Physics and Mathematics), Prof. of Lomonosov Moscow State University, Physical and Chemical departments, Director of Advanced Technologies Center, Director of Energy Efficient Technologies, ORCID: 0000-0001-8731-3947 / email@example.com
A.I.Akhmetova1, 2, 3, Engineer of A.N. Belozersky Institute of Physico-Chemical Biology, Leading Specialist of Advanced Technologies Center and of Energy Efficient Technologies,
D.V.Kornilov1, 2, Engineer, ORCID:0000-0003-1989-668X
Abstract. In the laboratory studies using an atomic force microscope, contact and resonance modes are the most widely used allowing acquisition of scans in relatively short times. On the other hand, their disadvantage is the difficulty of controlling the force impact, which can lead to irreversible changes in the morphology of the sample. This is particularly critical when scanning soft objects such as bacteria, cells, and polymers. In the contact mode this interaction occurs also due to frictional forces which magnitude is difficult to control when scanning the sample surface. In the resonance mode the influence of frictional forces is smoothed out by vertical motion of the probe; however, high quality of the cantilever oscillations does not allow controlling its force at sudden differences in topography of the observed surface. In order to overcome these obstacles, we have developed a mode of delicate and gentle surface scanning, which employs a light touch to the surface only. We have named this mode as flirt mode.
Keywords: atomic force microscope, FemtoScan X, piezo manipulator, contact and resonance mode, flirt mode, scanning probe microscopy
For citation: I.V. Yaminsky, A.I. Akhmetova, D.V. Kornilov. Fashion for the Flirt Mode. NANOINDUSTRY. 2022. V. 15, no. 3–4. PP. 178–185. https://doi.org/10.22184/1993-8578.2022.15.3–4.178.185
The name of this mode in English is flirt mode. Flirt mode is implemented in the fast atomic force microscope FemtoScan X using a programmable logic device (PLD). The deliberate choice is accomplished with the aid of a PLD, not a microcontroller. Analysis of the situation shows that the complexity of control and data transfer algorithms actually imposes the use of a processor as a central component of the scanning probe microscope control unit. Hence, the seemingly obvious solution is to use a microcontroller as a hardware platform. Thanks to their processor-based architecture, ease of programming and affordability, microcontrollers have found widespread use in industry and mass-production.
However, the microcontroller executes a program sequentially, in a "one instruction per clock cycle" mode, and cannot handle multiple signal connections simultaneously. This fact, together with the need to handle a large number of devices, creates a "bottleneck" in a microcontroller-based system, exemplified by the need to run several processes; some of them can only be executed when a certain condition is met while the others can be executed in parallel. It reduces the speed of the system and, hence, the quality of the scanning results.
Fortunately, programmable logic devices (PLDs) are available for complex tasks requiring high performance, parallel execution and multi-channel capability, such as the microscope control and data processing development. They allow of forming, directly from logic cells (LUTs) on one crystal, both a processor which is carrying out realization of high-level control algorithms of microscope modules, data processing, and low-level modules necessary to form the control signals of DAC, ADC, a frequency generator and other devices. This approach offloads the CPU, parallels the execution of microscope electronics tasks, and reduces the number of external signal connections and components of the device, thus increasing system performance. The greater programming flexibility and lack of a fixed command structure, like in microcontrollers, allows for more complex signal processing with the PLD, while the ability to re-flash enables expansion without replacing the CPU, saving significant amount of money. Thus, despite the higher cost and the need to initiate the time-consuming programming process, the positive aspects mentioned above make a PLD-based system the better solution in the long run.
That was the way we first chose to develop a FemtoScan X scanning probe microscope and, later, to implement a delicate mode of scanning – the flirting mode.
DESCRIPTION OF THE FLIRT MODE OPERATION
The FemtoScan X scanning probe microscope  makes use of an optical circuit to track the cantilever position during scanning. The laser beam, after reflection from the cantilever surface, falls on a photodiode, which signal corresponds to the cantilever bend.
Firstly, let us consider a process of bringing the cantilever to a sample surface. In the ideal case at first, when the probe is retracted from the sample, a signal from the photodiode that corresponds to the cantilever deflection (we will call it deflection) remains constant (Fig.1). Then, when a cantilever is lowered, in the simplest case there is a peak or pit corresponding to the Van der Waals interaction between the needle and the sample. Then the signal begins to rise linearly with the cantilever bending longitudinally. Correspondingly, at the reverse stroke the pulls away and the deflection signal reaches its stationary value.
In reality, the situation is somewhat more complicated. Firstly, the signal from the photodiode can fluctuate during the feeding process due to thermal movement of a cantilever, noise in the electrical circuit or other factors (Fig.2). Secondly, the sloping parts of the power curves may not coincide due to plastic deformation of the sample. Third, when the probe detaches from the surface, oscillations can be observed, which can be additionally analyzed, but it is also possible to ignore them by quickly lifting the probe in order to save scanning time and wait for some time in the upper point of the movement trajectory until the oscillations subside already after retraction. Thus, the scanning mode must be designed to take all these features into account.
Let us describe the scanning process in the flirt mode (Fig.3). Further, for the sake of convenience, it will be described in the sample reference system, i.e. as if the probe is moving, however, it is necessary to remember that in FemtoScan X scanning probe microscope it is actually the sample that is moving. This is done in order not to disturb the precise alignment of the laser beam. At the beginning of scanning a probe is smoothly brought to the first point on the sample surface by means of a piezo manipulator (it is supposed that the sample is in the height range accessible to the piezo manipulator, and in this case the height difference of its surface points is not too big and allows only scanning of the frame by means of the piezo manipulator). To reduce influence of different signal noise from the photodiode when measuring it during the approach process, averaging over several values is used.
When the signal from the photodiode reaches a pre-selected threshold value, the vertical position of the sample is recorded. In doing so, we can stop the approach process at any point, e.g. during Van der Waals interaction or somewhere earlier on the curve if electrostatic forces are acting in the system, and we can also control a given level of probe-sample interaction force, which is, for example, unattainable in the resonance mode, as the cantilever oscillates under mechanical resonance conditions. Thus, the first surface point is recorded and afterwards the probe is moved to height h. When retracting, we can analyse cantilever oscillations that occur when the probe pulls away from the surface. After retracting, the probe moves in the X–Y plane to the next position in the frame at which a new approach to the sample will begin. It is possible to set a waiting time at this point so that the cantilever oscillations have time to fade before beginning to move to a new point. This sequence of the approach-release-displacement is repeated at all subsequent points in the frame until the entire frame has been shot.
Thus, as compared to the contact and resonance modes, the flirt mode provides more precise control of the forces acting on the sampl, because the operator himself determines at which part of the force curve the probe-sample contact should be fixed. This feature of the developed mode is extremely useful for scanning soft objects, such as bacteria and blood cells, to avoid their deformations and irreversible changes in morphology of the studied objects.
Let us mention some analogues of the developed mode. These include PeakForce Tapping™, where the sample contact is handled by a feedback loop, and ringing mode  where cantilever oscillations after detachment are analyzed. Because of the more complex data processing, the average scanning time in these modes is slightly longer than in the flirt mode.
FLIRT MODE PARAMETERS
Let us consider the scanning parameters specific for the frirt mode (Fig.4, Fig.5). The detailed description of the flirt mode parameters is brought together in Table 1.
SCANNING IN FLIRT MODE
A sample of iodine crystals on the mica surface was chosen to demonstrate the capabilities and benefits of the flirt mode. Iodine was chosen as the brittle material to compare the effects of the cantilever in different modes. The comparison was made for the contact mode and the flirt mode. For the contact mode the values of the feedback links were chosen to give the best scanning result. It can be seen from the scanning results that in the contact mode most crystals were shifted by the probe during scanning, while in the flirt mode they remained in place (Fig.6).
The deflection signal is an indicative criterion for judging quality of the scan. Under perfect feedback this signal should tend to zero. Deviations in this signal indicate errors in a feedback loop. Magnitude of the feedback signal can be estimated from such parameters as range and root-mean-square deviation. From a comparison of the data in Figures 6c and 6d, it can be seen that the overall magnitude of the deflection signal for the contact mode is about 10 times greater than for the flirt mode. The RMS deviations are also very different: in the contact mode RMS = 1.3 × 10–2 V and in the flirt mode RMS = 3.5 × 10–3 V.
This paper contains just one possible illustrative example where the flirt mode has an advantage for studying the nanometre-sized objects. The main advantage of the flirt mode becomes evident in the study of soft matter – living cells and tissues, polymer gels, jelly-like materials and many other various systems and objects.
This publication presents the simplest implementation of the flirt mode. Its other versions may additionally include various modulation techniques, measurements at multiple frequencies, statistical analysis of power curves by a set of parameters, and much more.
The flirt mode, as well as the scanning probe microscopy itself, is being improved continuously. Certainly, it is not just a fashion to use the flirt mode but a sustained interest in it and its successful application in wildlife research.
The study was completed with the financial support of the RFBR, the London Royal Society No. 21-58-10005, RFBR, Project No. 20-32-90036, and from the Foundation for the Promotion of Innovation, Project No. 71108, Agreement 0071108, and with the support of Endor LLC, Moscow.
PEER REVIEW INFO
Editorial board thanks the anonymous reviewer(s) for their contribution to the peer review of this work. It is also grateful for their consent to publish papers on the journal’s website and SEL eLibrary eLIBRARY.RU. ■
Declaration of Competing Interest. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.