As the size of electronic devices continues to shrink, characterization methods capable of precisely probing localized properties become increasingly important. Scanning probe microscopy techniques can examine local phenomena, and conductive atomic force microscopy can, in particular, study local electromechanical properties. Such techniques have already played a valuable role in the development of nanoelectronics, but their capabilities remain relatively limited compared with the probe stations typically used to examine electronic devices. Here, we discuss the potential of conductive atomic force microscopy in nanoelectronics. We explore possible characterization strategies, enhanced electronics for the technique and improved multiprobe approaches. We also propose a multiprobe scanning probe microscopy system that combines different types of probes and could allow multiple nanofabrication and characterization experiments to be carried out simultaneously and under vacuum conditions.
Over the past few decades, the size of electronic devices has been continuously reduced — according to Moore’s law1 — leading to larger integration densities, higher computation speeds and lower power consumptions2. To analyse the performance of such devices, a probe station connected to a semiconductor parameter analyser (SPA) is typically used3. Here, the devices are exposed to electrical stresses (ramped voltage stresses (RVS) and constant voltage stresses (CVS), for example), and their most critical parameters and/or figures of merit (subthreshold slope in field-effect transistors (FETs) and switching voltages/currents in memristors, for example) can be extracted4. However, as fabricating devices can be complex and expensive, simplified test structures are often used5. In such cases, the tips of the probe station are not placed in direct contact with the surface of the material or device under test. This is because the tip–sample contact force cannot be accurately controlled and could easily lead to damage of the sample or device. Moreover, estimating the real tip–sample contact area is laborious and inaccurate, and can produce incorrect estimations of the electrical properties of the material or device (such as sheet resistance or tunnelling current density). Therefore, in a probe station, the electrical stresses are always applied by placing the tips on metallic pads or device terminals, and the signals registered apply to the entire area of the test structure or device. Therefore, it is impossible to assess which locations make more significant contributions to the electrical signals measured. As the size of the devices is scaled down to the nanoscale, the role played by spatial fluctuations, which are related to non-idealities in the devices (for example, thickness fluctuations, lattice distortions and grain boundaries in polycrystalline materials), becomes increasingly significant.
In the early 1990s — when the channel of FETs was reduced to <1 μm — the conductive atomic force microscope (CAFM) was developed by modifying a standard atomic force microscope (AFM)6. An AFM is a tool that allows the contact force (FC) between an ultrasharp tip and a sample to be accurately measured (sensitivity <1 nN)7. To do this, the tip is located at the end of a cantilever and the quantification of FC is achieved by detecting the deflection of the cantilever using an optical or a piezoresistive system8,9. As FC can be converted into distance by using the spring constant of the cantilever and Hooke’s law, AFMs have been readily employed to measure the surface topography of nanomaterials (over areas from 10–4 to 104 μm2) with very high spatial resolution (<0.1 nm vertically and <1 nm laterally)8.
If the ultrasharp tip used is conductive and if equipment to apply/measure electrical stress is connected to it, the CAFM approach can be also used to test the local electrical properties of a material. The main advantage of CAFM compared with a probe station is that the effective area across which the electrons can flow from tip to sample and vice versa (denoted as Aeff) can be as small as 1 nm2 (ref. 10). However, the electronic capabilities of standard CAFMs are much more limited than that of any probe station due to the use of modest voltage sources and current-to-voltage preamplifiers8. The reason is that CAFMs have always been understood (even by the manufacturers) as a mere extension of the AFM methods that can be used to collect basic electrical information about samples (that is, current maps under CVS, and RVS with very limited voltage, current and time resolution), rather than as an independent nanoelectronic characterization tool. However, as the size of electronic devices is scaled down to a few nanometres, interest in applying more sophisticated electrical stresses at the nanoscale has rapidly grown, and prototypes have been developed to enhance the electrical capabilities of CAFMs11. Nevertheless, many other improvements are still required to explore novel nanoelectronic phenomena. In this Perspective, we first examine the history and current status of CAFM, and then explore future setups, experiments and capabilities of the technique, discussing, in particular, their potential impact in the development of nanoelectronics.
History and status of CAFM
The first electrical measurement using an AFM was carried out in 1993, and consisted of analysing the reliability of a 12-nm-thick SiO2 film (grown on a p-type Si substrate) by applying a RVS at different locations, using a standard Si tip coated with 100 nm Ti connected to a source meter6. The measurements showed that the electrical strength of the SiO2 film could vary by a factor of up to 2.7 over an area of 0.14 μm2. By analysing the parameters measured during the RVS at different locations, the first CAFM map was also constructed6. Currently, all standard CAFMs can be used to collect topographic and current maps simultaneously and independently: that is, the topographic map is drawn by measuring the deflection of the cantilever at each location (pixel) of the sample (image), and the current map is built by recording the current using a current-to-voltage preamplifier12. The decoupling of the topography and current signals is a major advantage of CAFM compared with scanning tunnelling microscopy (STM), a tool that measures the current flowing from tip to sample (and vice versa) depending on the distance between them13 — if the current varies from one location to another, it is unclear whether this is related to a change in topography or resistivity (or both). The fact that in the CAFM the tip is in physical contact with the sample also simplifies the setup; a STM typically requires operation in an ultrahigh vacuum chamber, but CAFM does not.
Following this initial report6, CAFM studies concentrated on studying the reliability of ultrathin dielectrics (especially high-dielectric constant (k) materials) and several related phenomena, including tunnelling current14, trap-assisted tunnelling15, stress-induced leakage current16, polycrystallization17, effect of doping18, irradiation19, dielectric breakdown20 and resistive switching21, were analysed. In CAFM studies, the tip is usually placed on a dielectric material, otherwise the high lateral resolution of the technique is lost because electrons far from the contact spot are also collected22. However, over the past 25 years, the CAFM has been also employed to analyse many other nanomaterials (for example, nanowires23, nanotubes24, quantum dots25, two-dimensional (2D) materials26, molecular junctions27 and living cells28) and various electronic phenomena (for example, ferroelectricity29, photoelectricity30, flexoelectricity31 and piezoelectricity32,33) at the nanoscale.
Unfortunately, many reports did not provide enough data to demonstrate their claims statistically, which in CAFM science is essential because the currents registered by the CAFM tip depend on many experimental factors. These include the value of FC (ref. 34), the environmental conditions (air, dry nitrogen, vacuum (<10–5 torr), ultrahigh vacuum (<10–10 torr))10,20, the radius of the tip at the apex35 and the conductivity of the tip14. A valuable example of CAFM characterization can be found in ref. 14, in which the authors determined the value of Aeff by collecting and fitting more than 7,200 current versus voltage (I–V) curves. By using such a large number of I–V curves and fittings, they were able to obtain a smaller uncertainty in the value reported, compare the effect of the apex radius and metallic coating of the tip, and even detect some defective tips provided by the manufacturer.
Another critical problem of CAFM studies is the degradation of the tips due to the high current densities (up to 108 A cm−2) and lateral frictions, which can lead to lower (false) currents as the experiments proceed36. Fortunately, (B-doped) diamond-coated Si tips, solid (B-doped) diamond tips and solid metallic tips provide a longer lifetime, but they can also degrade due to particle adhesion and sharpness loss — and diamond tips can easily damage (scratch) the surface of most samples37. In any case, the industry of ultrasharp conductive tips for CAFMs has developed quickly, and today it is possible to select the best tip for each type of experiment (for example, spectroscopic measurements or lateral scans). In parallel, CAFM instrumentation also experienced notable progress, which has allowed fundamental limitations to be solved. For example, the development of CAFMs working in intermittent contact has allowed soft materials (such as conductive polymers and membranes) to be measured in a reliable manner, as the absence of lateral frictions helps to preserve even the softest samples38,39 — this also significantly helps to maintain the conductivity of the CAFM tips. Figure 1 shows a timeline of the main breakthroughs in CAFM, from its invention until today.
Novel characterization strategies
Beyond the established capabilities of CAFM (Fig. 1), many new CAFM experiments are possible. In particular, the CAFM has a unique ability to apply mechanical and electrical stresses simultaneously with very high precision, which could be of value in flexible electronics.
A CAFM has, for example, been used to evaluate the mechanical strength of grains and grain boundaries in polycrystalline HfO2 films40, showing that nanoscale electrical phenomena, such as resistive switching, only take place at mechanically weak locations (that is, the grain boundaries). Other studies have used the CAFM to induce pressure-related conductivity changes in the materials, which are related to flexoelectric31 and piezoelectric33 effects (among others). However, though a degree of deformation may have been induced in the materials during these experiments, the samples do not correspond to real flexible devices. An early illustration of the real potential of the CAFM to measure flexible electronic devices can instead be found in ref. 23, in which the current generated by a piezoelectric ZnO nanowire was quantified under bending provoked by the CAFM tip.
Until now, the performance of flexible and electronic devices has been tested by applying bending stresses at the device level (using a piezo stage), and measuring the current signals driven by these devices before and after the tests41,42 or during them43 (using a source meter or a SPA). However, the amount of mechanical stress that can be applied using this setup is very limited, and so far, bending stresses with radiuses down to few millimetres have been applied41,42,43,44, which may not be enough to test the materials under critical conditions. Moreover, device-level tests can mask several local phenomena driven by nanoscale features of the samples, such as lattice distortions and domains. In this context, using the CAFM would allow the application of local bending down to approximately 2 nm — that is, the minimum radius of the tip — and at the same time monitor the electrical properties of the materials.
How the conductance of a conductive filament across a dielectric changes under normal compressive strain applied with the tip of an AFM has, for example, been observed45. However, in that work, the current signal was collected with an external source meter: that is, the tip, which was not conductive, was only used to apply the mechanical stress. Using the CAFM tip to apply a mechanical stress and simultaneously collect electrical information about the samples should allow new nanoscale phenomena to be investigated40.
Another possibility in flexible optoelectronics is to combine the CAFM with other techniques to observe directly the bending/deflection of samples. For example, recently, an AFM integrated inside the chamber of a scanning electron microscope (SEM) was used to observe in situ the bending of graphene sheets under a mechanical strain applied with the tip of the AFM46. But again, this experiment did not collect any electrical signals. Nevertheless, integrating a CAFM in the chamber of an SEM could be used to analyse several other aspects of optoelectronic devices, such as material thinness and bending under lateral strains and its effect on the performance of FETs and memristors (among others).
One of the main advantages of such a setup is that SEM images can be helpful in locating the CAFM tip at the regions of interest on the sample, as well as observing at which locations the currents are generated. For instance, a CAFM integrated into an SEM has been used to analyse the origin of the currents generated in forest-like vertical arrays of ZnO nanowires, and it was observed that the currents are related not only to piezoelectric potential but also to contact and triboelectric potentials47. This unique observation was possible because the position of the CAFM tip and the bending of the nanowires were monitored using the SEM and correlated with the currents collected with the CAFM tip. Some AFMs can be also complemented with other optical techniques to obtain additional information about the devices9. For example, the combination of Raman spectroscopy and CAFM to study FETs with a channel made of graphene or other 2D materials should allow the carrier mobility to be linked with channel quality (for example, thickness, crystallinity). These types of experiment are easier to carry out with CAFMs that sense the bending of the cantilever instead, using piezoresistive systems, as they provide a clear (laser-free) interface above the tip.
Another promising CAFM-based characterization strategy is to collect current maps at different vertical heights and construct a 4D image of the material or device under test48. Standard CAFM current maps include three dimensions: two lateral distances (x and y) and current scale as the third dimension; a 4D image consists of taking CAFM current maps at different depths of the materials and assembling them together48. Such an approach is possible due to the ability of the CAFM to scratch the surface of a sample (and remove material) when a high enough FC is applied and when a tip with a high enough stiffness is used.
The etching ability of AFMs has been known for more than 20 years49, but the collection of current maps at specific heights is a relatively recent development50,51. However, this technique is still in its early stage and some fundamental limitations still need to be resolved. It was recently demonstrated that the CAFM-assisted etching process severely modifies the electrical properties of some materials (Si, TiO2 and Al2O3) even when applying a FC below the minimum required to induce etching52. This can easily result in false characterization of the features of interest within the sample if they are not stable enough. The developers of this technique have also claimed that it can be used to characterize local features narrower than 10 nm with subnanometre vertical resolution48. This however is a potential exaggeration given the large instability of the etching rate as the experiment proceeds and the increase in roughness at the etched region. Other groups have reported substantial difficulties when monitoring features with diameters below 30 nm (refs. 52,53,54), which appears more reasonable.
Nevertheless, this technique has shown an extraordinary ability to characterize different nanomaterials in four dimensions, and it could play an important role in the development of nanoelectronics. For example, the approach has been used to characterize 2.2-μm-thick polycrystalline CdTe-based solar cells55, in which the grain boundaries were wide enough (>100 nm) to being reliably monitored; the resulting images show unique 4D information about the samples (Fig. 2).
Enhancing the electronics of CAFM
Despite the progress achieved with CAFM, the electrical capabilities of this technique are still a major limitation, and most CAFMs cannot apply high voltages (>10 V), measure currents over several orders of magnitude (from picoampere to milliampere), apply fast pulsed voltage stresses (PVS) and use variable current limitations. To solve these problems, one solution is to connect the SPA directly to the CAFM tip. Using this setup, the size of the dielectric breakdown spot in thin SiO2 films depending on the current limitations used during RVS has been analysed20,56, and random telegraph noise signals in different dielectrics (HfO2, hexagonal boron nitride) have been monitored by applying CVS and collecting current versus time (I–t) curves with high temporal resolution57,58, which allowed charge trapping and de-trapping mechanisms in the materials to be understood. Similarly, the degradation of thin SiO2 and high-k dielectric films has been analysed when applying PVS with frequencies up to 10 KHz (refs. 59,60).
The main problem of this setup is, however, the synchronization of the SPA with the movement of the CAFM tip. In brief, using the SPA to register currents during current maps would require starting (and finishing) a new measurement at each pixel. However, the SPA provides poor control on the time from one measurement to another, especially when low currents down to 1 pA are registered. Alternatively, currents have been registered using an external source during non-conductive AFM tip movement61,62, and the data obtained used to construct current images. However, the authors did not disclose details on this setup and methodology, plus the currents registered were in the microampere regime (which is quite large and does not fit most CAFM experiments). Consequently, the method has not been adopted by other researchers, and the solution most commonly used has been to connect the SPA to the CAFM tip to perform spectroscopic (static) measurements.
Recently, some manufacturers have started to offer enhanced electronics by including a high-voltage source, a logarithmic current-to-voltage preamplifier and a current limitation tunable via software39,63. However, so far, no manufacturer has offered the option of measuring I–t curves with high temporal resolution and/or high-frequency PVS. The most significant advancement in this direction is the use of microwaves with frequencies up to 3 GHz (ref. 64). This setup provides the advantage that the electrical circuit does not need to be closed: that is, the material to be analysed can be placed on an insulating substrate because the measurement is based on microwave reflection. However, it requires shielded tips, which are less sharp and more expensive.
The ideal features of a single-tip CAFM for advanced nanoelectronics research are listed in Supplementary Table 1; currently, no commercially available CAFMs can achieve all the listed features at the same time.
Despite their superior lateral resolution, most CAFMs still have a major limitation compared with probe stations: they only include one tip. Therefore, the electrical stresses can only be applied between two electrodes: the tip and the sample holder. As mentioned previously, one option could be to connect external electrical equipment to the samples. However, this requires additional sample preparation (for example, electrode pre-patterning, wire bonder connection), and equipment synchronization may also make the measurements more difficult. To solve this limitation, CAFM systems that include more than one tip have been developed9,65. The key to this setup is to use a piezoresistive system to detect the deflection of each tip; using one laser and one photodiode to detect the deflection of multiple tips would increase the complexity of the hardware. Multiprobe CAFM has already been used to provide a variety of new insights, including measuring the local sheet resistance of mechanically exfoliated graphene flakes66 and measuring the contact resistance of single-walled carbon nanotubes9.
However, so far, the use of multiprobe CAFM has been relatively limited, and the number of reports using this technique is low. The main challenge with multiprobe CAFM is placing all the tips on the locations of interest on the sample. An optical microscope has been used to approach the tips roughly to the features of interest on the sample 67, and then a topographic map has been obtained with each tip to point them to the precise locations of interest66: here, specifically, each corner of a graphene sheet. However, if the tips are located very close to each other — which is necessary in many experiments — unwanted collisions can occur between them during the scans. Alternatively, a scan in which the sample is moved (and all tips are kept static) can be used to simultaneously collect a topographic map with each tip. However, if the aim is to monitor the same feature of a sample with different tips, the size of the scan should be larger than the distance between the tips, which in many cases could be too large (>10 μm) and incompatible with the planned experiment.
Another option is to integrate multiprobe CAFM into a SEM system, as has already been done for single-probe CAFM47. However, increasing the amount of CAFM tips inside the SEM chamber significantly increases the complexity and cost of the entire setup, and may result in lower CAFM performance (for example, lower lateral resolution, larger thermal drift). STM is, in fact, easier to integrate into a SEM or TEM68 (Fig. 3a) because it does not require FC to be accurately controlled. There is a previous report on integrating a multiprobe AFM into a SEM69, so that the position of the tips can be monitored with the SEM (Fig. 3c,d). However, this study did not carry out any electrical measurements.
Despite suggestions that multiprobe CAFM has been integrated into an SEM system68, the approach has not yet been clearly demonstrated. (The works discussed in ref. 68 employ STM and AFM probe tips, but they do not demonstrate current collection with FC control during tip location via SEM, or position–current correlation, as was shown with single-probe CAFM47.) Therefore, effective integration of multiprobe CAFM into a SEM remains a key short-term challenge for the field.
Furthermore, positioning multiple tips on the surface of a sample requires the use of customized tips to allow apex visualization (the cantilever should not hide the apex). Probe tips with carbon nanotubes attached have been used to provide smaller tip radius and clear apex visualization69, but this again increases the complexity of the whole experiment (due to the difficulty of attaching the carbon nanotube to the apex with controlled orientation and high interface conductivity). Consequently, additional efforts to simplify tip positioning in multiprobe CAFM are necessary. In this direction, the most feasible idea would be to scan the region of interest with only one master tip (when all other tips are far enough apart), and then point all the tips to the locations of interest in the topographic map recorded with the master tip. The approaching system should consider the shape of the tips (including the cantilever) to avoid collisions.
Multiprobe CAFM could also be used as a nanoscale probe station by placing the tips on the metallic electrodes of the devices. It has been demonstrated that when the tip of a CAFM is located on a metallic electrode, the system is able to collect all the current flowing across the entire area of the electrode, similar to what happens in a standard probe station70. Therefore, multiprobe CAFM could be used as a nanoscale probe station to test real devices without the need to pattern large pads; the pads normally used for device prototype testing can be as large as 100 μm × 100 μm. This approach could provide additional information about specific circuit-related phenomena affecting the performance of the devices, such as sneak path currents in cross-bar memristive arrays.
Multiprobe combined SPM
Each tip within a multiprobe SPM system could potentially be used to measure different electrical magnitudes simultaneously, and even to carry out nanodevice fabrication tasks. And if the entire system is placed inside a vacuum chamber, it could be used for in situ device fabrication and characterization within a perfectly clean environment. This potential system, for which we propose the name of multiprobe combined scanning probe microscopy (MPC-SPM), is schematically illustrated in Fig. 4. In fact, we may not be far from such a development, as all the required parts have been already developed and only integration issues remain to be solved.
In parallel to the development of CAFM, many other SPM-based characterization techniques relevant in nanoelectronics research have been developed. The most notable of these are: scanning capacitance microscopy (SCM), which allows the amount and polarity of dopants in a material to be measured71; Kelvin probe force microscopy (KPFM), which allows the contact potential difference between a probe tip and a sample surface to be measured72; scanning thermal microscopy (SThM), which can be used to detect the local temperature and thermal conductivity of a material at the nanoscale73; scanning photocurrent microscopy (SPCM), which is able to map the current signals generated by the illumination of a focused light beam74; scanning microwave impedance microscopy (sMIM), which is a near-field technique that allows the conductivity, dielectric constant, carriers density and doping of a material (even if this is buried) to be measured64,75; and scanning gate microscopy (SGM), which uses a conductive tip placed above the surface of the sample as a gate electrode (that is, apply an electrical field to change the conductivity of a part of the sample)76.
All these modes have been optimized in single-tip SPM systems and, and so far, some studies have proposed initial combinations. For example, a joint characterization of CAFM and KPFM has been performed on the same area of a SiC/Si nanocrystals/SiC sandwich structure72. In this approach, electrons and holes were first injected into the Si nanocrystals with a CAFM tip, and the charging effect and carrier transportation process were analysed by KPFM.
In addition, SPM probe tips could be also used for sample preparation. Apart from the well-known SPM-based lithography based on oxide growth49 and material removal48, recent advancements in tip design have allowed very narrow (<16 nm) metallic lines to be drawn on the surface of samples using a hollow tip filled with metallic particles in a glycol-based dispersion77. This approach provides a better controllability compared with the traditional material deposition method via capillary effects78.
The main difficulty limiting the combination of all these SPM-based techniques is integration. First, finding out (with one technique) the location of the sample previously treated and/or analysed (with another technique) is complex, because in almost all cases, switching between different techniques requires changing the tip and/or tip holder. The problem here is that in this process, the position studied is lost and finding it again is difficult. The use of nanomarks could help to find out (roughly) the locations of interest, but even then, pointing to exactly the same location would be extremely complex. Recently, a special tip holder has been developed that can be used to switch from contact mode to tapping mode without having to change the tip holder79, but even that would be useless when switching between techniques that use different types of probes (for example, CAFM and SThM). Furthermore, the sample will be exposed to atmospheric air while the SPM tips are exchanged, which may have negative and uncontrollable effects in several types of experiments. In addition, combining different SPM techniques simultaneously has never been achieved before. As discussed previously, multiprobe STM, AFM and even CAFM have been developed, but in those systems, all the tips are identical and the maximum that may be achieved is to study topography with one tip and current with another.
Therefore, the concept behind MPC-SPM would be to integrate all these SPM-based nanofabrication and nanocharacterization techniques in one single system with multiple probes, operating in a high-vacuum atmosphere (Fig. 4). To do so, the SPM system should allow different types of tips to be mounted in each tip holder. In some multiprobe systems that are already commercially available, this appears to be impossible due to the reduced size and customized nature of the tip holder65. In this sense, modular multiprobe systems may be more suitable9, though their drawback is that they normally have lower lateral resolution. Moreover, the tips should be able to work on the same area simultaneously, and that would require accurate positioning of the tips without collisions. To solve this issue, the development of a master tip to collect one topographic map of the surface of the sample and locate all tips based on that (as discussed in the previous section) seems to be (again) the best option.
Another issue that would need to be optimized for MPC-SPM is the sample holder and vacuum chamber. On the one hand, using a very small sample holder (diameter <2 cm) facilitates the construction of a vacuum chamber, which is essential to avoid oxygen contamination and related phenomena (such as local anodic oxidation49). In fact, the vacuum chamber should ideally be combined with a heating stage to remove all water molecules from the surface of the sample48. This heating stage can be also very useful in many reliability experiments — most probe stations include a heating chuck, and therefore it should also be included in advanced CAFMs. However, on the other hand, using a small sample holder and vacuum chamber would lead to a closed interface and would strongly obstruct sample manipulation, tip positioning and device visualization. In this regard, CAFMs with large sample holders and open interfaces allow several samples to be loaded simultaneously and external equipment to be connected to the samples or devices for advanced characterization. For example, a CAFM with an open interface has been used to apply an electrical field (using the tip of the CAFM) while the device was externally stressed (using a source meter)80.
CAFM has already played a valuable role in the development of nanoelectronics, but the capabilities of this technique could be significantly enhanced in four key ways. The first and most immediate strategy is to design and optimize novel characterization methods, such as 4D characterization and electromechanical tests for flexible electronics. However, some of these tests may require additional accessories, such as the fabrication of sample holders that can be used to stretch flexible samples. The second strategy is to enhance the electronics of the CAFMs to apply advanced electrical stresses, such as RVS with high current, voltage and temporal resolution (including current limitations) and high-frequency stresses (PVS, microwaves). The third strategy is to develop CAFMs with a simple user interface that allows measurements to be made under different conditions simultaneously, such as controlled environment, different temperatures, intermittent contact (and switch to contact without the need of changing the tip holder) and connection of external systems (for example, source meters). The design and optimization of customized CAFMs that could be used in combination with other large techniques (such as Raman spectroscopy and SEM) is also desirable. Finally, the fourth and most ambitious strategy is to develop multiprobe SPM systems (working in vacuum) that combine different types of probes for versatile in situ nanofabrication and nanocharacterization experiments.
Moore, G. E. Progress in digital integrated electronics. In Technical Digest. IEEE Int. Electron Devices Meet. 11–13 (IEEE, 1975); https://doi.org/10.1109/N-SSC.2006.4804410
International Roadmap for Devices and Systems (IRDS, 2017); https://irds.ieee.org/roadmap-2017
Allport, P. P. et al. FOXFET biased microstrip detectors. Nucl. Instrum. Methods Phys. Res. A 310, 155–159 (1991).
Xiang, J. et al. Ge/Si nanowire heterostructures as high-performance field-effect transistors. Nature 441, 489–493 (2006).
Lanza, M. et al. Recommended methods to study resistive switching devices. Adv. Electron. Mater. 1, 1800143 (2018).
Murrell, M. P. et al. Spatially resolved electrical measurements of SiO2 gate oxides using atomic force microscopy. Appl. Phys. Lett. 62, 786–788 (1993).
Capella, B. & Dietler, G. Force–distance curves by atomic force microscopy. Surf. Sci. Rep. 34, 1–104 (1999).
Pan, C., Shi., Y., Hui, F., Grustan-Gutierrez, E. & Lanza, M. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 1 (Wiley-VCH, 2017).
Lewis, D., Ignatov, A., Krol, S., Dekhter, R. & Strinkovsky, A. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 13 (Wiley-VCH, 2017).
Lanza, M. et al. Electrical resolution during conductive AFM measurements under different environmental conditions and contact forces. Rev. Sci. Instrum. 81, 106110 (2010).
Iglesias, V., Jing, X. & Lanza, M. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 10 (Wiley-VCH, 2017).
Iglesias, V. et al. Correlation between the nanoscale electrical and morphological properties of crystallized hafnium oxide-based metal oxide semiconductor structures. Appl. Phys. Lett. 97, 262906 (2010).
Binnig, G. & Rohrer, H. Scanning tunneling microscopy. Surf. Sci. 126, 236–244 (1983).
Frammelsberger, W., Benstetter, G., Kiely, J. & Stamp, R. C-AFM-based thickness determination of thin and ultra-thin SiO2 films by use of different conductive-coated probe tips. Appl. Surf. Sci. 253, 3615–3626 (2007).
Pirrotta, O. Leakage current though the poly-crystalline HfO2: trap densities at the grains and grain boundaries. J. Appl. Phys. 114, 134503 (2013).
Lanza, M. et al. Trapped charge and stress induced leakage current (SILC) in tunnel SiO2 layers of de-processed MOS non-volatile memory devices observed at the nanoscale. Microelectron. Reliab. 49, 1188–1191 (2009).
Muensterman, R. et al. Correlation between growth kinetics and nanoscale resistive switching properties of SrTiO3 thin films. J. Appl. Phys. 108, 124504 (2010).
Kajewski, D. et al. Local conductivity of epitaxial Fe-doped SrTiO3 thin films. Phase Transit. 84, 5–6 (2011).
Gomez-Navarro, C. et al. Tuning the conductance of single-walled carbon nanotubes by ion irradiation in the Anderson localization regime. Nat. Mater. 4, 534–539 (2005).
Uppal, H. J. Breakdown and degradation of ultrathin Hf-based (HfO2)x(SiO2)1–x gate oxide films. J. Vac. Sci. Technol. B 27, 443–447 (2009).
Lanza, M. A review on resistive switching in high-k dielectrics: a nanoscale point of view using conductive atomic force microscope. Materials 7, 2155–2182 (2014).
Giannazzo, F., Sonde, S., Rimini, E. & Raineri, B. Lateral homogeneity of the electronic properties in pristine and ion-irradiated graphene probed by scanning capacitance spectroscopy. Nanoscale Res. Lett. 6, 109 (2011).
Wang, Z. L. & Song, J. H. Piezoelectric nanogenerators based on zinc oxide nanowire arrays. Science 312, 242–246 (2006).
Hung, S. C., Su, Y. K., Chang, S. J. & Chen, Y. H. Vertically aligned GaN nanotubes — fabrication and current image analysis. Microelectron. Eng. 83, 2441–2445 (2006).
Lv, Y., Cui, J., Jiang, Z. M. & Yang, X. J. Composition and conductance distributions of single GeSi quantum rings studied by conductive atomic force microscopy combined with selective chemical etching. Nanotechnology 24, 065702 (2013).
Shi, Y. et al. Electronic synapses made of layered two-dimensional materials. Nat. Electron. 1, 458–465 (2018).
Wen, Y. et al. Multilayer graphene-coated atomic force microscopy tips for molecular junctions. Adv. Mater. 24, 3482–3485 (2012).
Frederix, P. L. T. M. et al. Assessment of insulated conductive cantilevers for biology and electrochemistry. Nanotechnology 16, 997–1005 (2005).
Xiao, J. X. et al. Room temperature ferroelectricity of hybrid organic-inorganic perovskites with mixed iodine and bromine. J. Mater. Chem. A 6, 9665–9676 (2018).
Han, T. et al. Photo-electrochemical water splitting in silicon based photocathodes enhanced by plasmonic/catalytic nanostructures. Mater. Sci. Eng. B 225, 128–133 (2017).
Bhaskar, U. K. et al. A flexoelectric microelectromechanical system on silicon. Nat. Nanotechnol. 11, 263–266 (2016).
Pan, C. et al. Suppression of nanowire clustering in hybrid energy harvesters. J. Mater. Chem. C. 4, 3646–3653 (2014).
Song, X. et al. Enhanced piezoelectric effect at the edges of stepped molybdenum dissulfide nanosheets. Nanoscale 9, 6237–6245 (2017).
Ranjan, A. et al. Analysis of quantum conductance, read disturb and switching statistics in HfO2 RRAM using conductive AFM. Microelectron. Reliab. 64, 172–178 (2016).
Benstetter, G., Hofer, A., Liu, D., Frammelsberger, W. & Lanza, M. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 3 (Wiley-VCH, 2017).
Krause, O. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 2 (Wiley-VCH, 2017).
Khun, N. W. Scratch-induced wear behavior of aluminum alloy under dry and wet conditions. J. Mechatron. 3, 301–306 (2016).
Simultaneous Electrical and Mechanical Property Mapping at the Nanoscale with PeakForce TUNA Application Note 132 (Bruker, 2011); https://www.bruker.com/products/surface-and-dimensional-analysis/atomic-force-microscopes/modes/modes/nanoelectrical-modes/pf-tuna.html
Pacheco, L. & Martinez, N. F. in Conductive Atomic Force Microscopy: Application in Nanomaterials (ed. Lanza, M.) Ch. 12 (Wiley-VCH, 2017).
Shi, Y. et al. In situ demonstration of the link between mechanical strength and resistive switching in resistive random-access memories. Adv. Electron. Mater. 1, 1400058 (2015).
Kim, K. S. et al. Large-scale pattern growth of graphene films for stretchable transparent electrodes. Nature 457, 706–710 (2009).
Yao, J. et al. Highly transparent nonvolatile resistive memory devices from silicon oxide and graphene. Nat. Commun. 3, 1101 (2012).
Pan, C. et al. Coexistence of grain-boundaries-assisted bipolar and threshold resistive switching in multilayer hexagonal boron nitride. Adv. Funct. Mater. 27, 1604711 (2017).
Hui, F. et al. Graphene and related materials for resistive random access memories. Adv. Electron. Mater. 3, 1600195 (2017).
Miao, F. et al. Anatomy of a nanoscale conduction channel reveals the mechanism of a high-performance memristor. Adv. Mater. 23, 5633–5640 (2001).
SuperFlat AFM. Kleindiek Nanotechnik https://www.nanotechnik.com/sf-afm.html (2019).
Wen, C. et al. In situ observation of current generation in ZnO nanowire based nanogenerators using a CAFM integrated into an SEM. ACS Appl. Mater. Interfaces 11, 15183–15188 (2019).
Celano, U. Metrology and Physical Mechanisms in New Generation Ionic Devices (Springer Theses, Springer, 2016)
Garcia, R., Knoll, A. W. & Riedo, E. Advanced scanning probe lithography. Nat. Nanotechnol. 9, 577–587 (2014).
Liu, D., Benstetter, G. & Frammelsberger, W. The effect of the surface layer of tetrahedral amorphous carbon films on their tribological and electron emission properties investigated by atomic force microscopy. Appl. Phys. Lett. 82, 3898 (2003).
Celano, U. et al. Conductive-AFM tomography for 3D filament observation in resistive switching devices. In 2013 IEEE Int. Electron Devices Meet. 21.6.1–21.6.4 (IEEE, 2013); https://doi.org/10.1109/IEDM.2013.6724679
Chen, S. et al. On the limits of scalpel AFM for the three dimensional electrical characterization of nanomaterials. Adv. Funct. Mater. 28, 1802266 (2018).
Hammock, S. M. 3D c-AFM Imaging of Conductive Filaments in HfO 2 Resistive Switching Devices (Texas A&M University Libraries, 2017).
Buckwell, M., Montesi, L., Hudziak, S., Mehonic, A. & Kenyon, A. J. Conductance tomography of conductive filaments in intrinsic silicon-rich silica RRAM. Nanoscale 7, 18030–18035 (2015).
Luria, J. et al. Charge transport in CdTe solar cells revealed by conductive tomographic atomic force microscopy. Nat. Energy 1, 16150 (2016).
Porti, M., Aguilera, L., Blasco, X., Nafria, M. & Aymerich, X. Reliability of SiO2 and high-k gate insulators: a nanoscale study with conductive atomic force microscopy. Microelectron. Eng. 84, 501–505 (2007).
Ranjan, A. et al. CAFM based spectroscopy of stress-induced defects in HfO2 with experimental evidence of the clustering model and metastable vacancy defect state. In 2016 IEEE Int. Reliability Phys. Symp. 7A-4-1–7A-4-7 (IEEE, 2016); https://doi.org/10.1109/IRPS.2016.7574576
Jiang, L. et al. Dielectric breakdown in chemical vapor deposited hexagonal boron nitride. ACS Appl. Mater. Interfaces 9, 39758–39770 (2017).
Wu, Y. L., Lin, J. J., Chang, S. H. & Huang, C. Y. The degradation of thin silicon dioxide films subjected to pulse voltage stresses at nanoscale. ECS Trans. 28, 339–343 (2010).
Foissac, R., Blonkowski, S. & Kogelschatz, M. Nanoscale characterization of high-K/IL gate stack TDDB distributions after high-field prestress pulses. IEEE Trans. Device Mater. Reliab. 15, 298–307 (2015).
Lau, C. N., Stewart, D. R., Williams, S. & Bockrath, M. Direct observation of nanoscale switching centers in metal/molecule/ metal structures. Nano Lett. 4, 569–572 (2004).
Miao, F. et al. Anatomy of a nanoscale conduction channel reveals the mechanism of a high-performance memristor. Adv. Mater. 23, 5633–5640 (2011).
Park NX-Hivac: high vacuum atomic force microscope. Park Systems https://www.parksystems.com/index.php/products/small-sample-afm/park-nx-hivac (2019).
Products. PrimeNano https://primenanoinc.com/products.html (2018).
LT Nanoprobe. Scienta Omicron https://www.scientaomicron.com/en/products/lt-nanoprobe/1399 (2019).
Higuchi, S., Kubo, O., Kuramochi, H., Aono, M. & Nakayama, T. A quadruple-scanning-probe force microscope for electrical property measurements of microscopic materials. Nanotechnology 22, 285205 (2011).
Klein, A. E., Janunts, N., Tunnermann, A. & Pertscch, T. Investigation of mechanical interactions between the tips of two scanning near-field optical microscopes. Appl. Phys. B 108, 737–741 (2012).
Nakayama, T. et al. Development and application of multiple-probe scanning probe microscopes. Adv. Mater. 24, 1675–1692 (2012).
Takahashi, M., Ko, H., Ushiki, T. & Iwata, F. Interactive nano manipulator based on an atomic force microscope for scanning electron microscopy. In Proc. 2011 International Symposium on Micro-NanoMechatronics and Human Science 495–500 (IEEE, 2011); https://doi.org/10.1109/MHS.2011.6102241
Erlbacher, T., Yanev, V., Rommel, M., Bauer, A. J. & Frey, L. Gate oxide reliability at the nanoscale evaluated by combining conductive atomic force microscopy and constant voltage stress. J. Vac. Sci. Technol. B 29, 01AB08 (2011).
Matey, J. R. & Blanc, J. Scanning capacitance microscopy. J. Appl. Phys. 57, 1437 (1985).
Xu, J., Xu, J., Zhang, P., Li, W. & Chen, K. Nanoscale quantification of charge injection and transportation process in Si-nanocrystal based sandwiched structure. Nanoscale 5, 9971–9977 (2013).
Shi, L. & Majumdar, A. Scanning thermal microscopy of carbon nanotubes using batch-fabricated probes. Appl. Phys. Lett. 77, 4295–4297 (2000).
Gu, Y. et al. Near-field scanning photocurrent microscopy of a nanowire photodetector. Appl. Phys. Lett. 87, 04311 (2005).
Lai, K., Kundhikanjana, W., Kelly, M. & Shen, Z. X. Modeling and characterization of a cantilever-based near-field scanning microwave impedance microscope. Rev. Sci. Instrum. 79, 063703 (2008).
Aoki, N., Cunha, C. R., Akis, R., Ferry, D. K. & Ochiai, Y. Imaging of integer quantum Hall edge state in a quantum point contact via scanning gate microscopy. Phys. Rev. B 72, 155327 (2005).
Yeshua, T. et al. Micrometer to 15 nm printing of metallic inks with fountain pen nanolithography. Small 14, 1702324 (2017).
Piner, R. D., Zhu, J., Xu, F., Hong, S. & Mirkin, C. A. “Dip-pen” nanolithography. Science 283, 661–663 (1999).
Ribeiro-Palau, R. et al. Twistable electronics with dynamically rotatable heterostructures. Science 361, 690–693 (2018).
Sellier, H. et al. On the imaging of electron transport in semiconductor quantum structures by scanning-gate microscopy: successes and limitations. Semicond. Sci. Technol. 26, 064008 (2011).
Snow, E. S. & Campbell, P. M. Fabrication of Si nanostructures with an atomic force microscope. Appl. Phys. Lett. 64, 1932–1934 (1994).
Kado, H. & Tohda, T. Nanometer-scale recording on chalcogenide films with an atomic force microscope. Appl. Phys. Lett. 66, 2961–2962 (1995).
Houze, F., Meyer, R., Schneegans, O. & Boyer, L. Imaging the local electrical properties of metal surfaces by atomic force microscopy with conducting probes. Appl. Phys. Lett. 69, 1975–1977 (1996).
Sugimura, H. & Nakagiri, N. AFM lithography in constant current mode. Nanotechnology 8, A15–A18 (1997).
Lantz, M. A., O’Shea, S. J. & Welland, M. E. Characterization of tips for conducting atomic force microscopy in ultrahigh vacuum. Rev. Sci. Instrum. 69, 1757–1764 (1998).
Durkan, C., Welland, M. E., Chu, D. P. & Migliorato, P. Probing domains at the nanometer scale in piezoelectric thin films. Phys. Rev. B 60, 16198–16204 (1999).
Landau, S. A. et al. Scanning probe microscopy — a tool for the investigation of high-k materials. Appl. Surf. Sci. 157, 387–392 (2000).
Cai, L., Tabata, H. & Kawai, T. Probing electrical properties of oriented DNA by conducting atomic force microscopy. Nanotechnology 12, 211–216 (2001).
Bietsch, A. & Michel, B. Size and grain-boundary effects of a gold nanowire measured by conducting atomic force microscopy. Appl. Phys. Lett. 80, 3346–3348 (2002).
Hayakawa, J. et al. Current-driven switching of exchange biased spin-valve giant magnetoresistive nanopillars using a conducting nanoprobe. J. Appl. Phys. 96, 3440 (2004).
Masuda, H., Takeuchi, M. & Takahashi, T. Local photocurrent detection on InAs wires by conductive AFM. Ultramicroscopy 105, 137–142 (2005).
Szot, K., Speier, W., Bihlmayer, G. & Waser, A. R. Switching the electrical resistance of individual dislocations in single-crystalline SrTiO3. Nat. Mater. 5, 312–320 (2006).
Sire, C., Blonkowski, S., Gordon, M. J. & Baron, T. Statistics of electrical breakdown field in HfO2 and SiO2 films from millimeter to nanometer length scales. Appl. Phys. Lett. 91, 242905 (2007).
Cen, C. et al. Nanoscale control of an interfacial metal–insulator transition at room temperature. Nat. Mater. 7, 298–302 (2008).
Ramesha, G. G. & Sampath, S. Electrochemical reduction of oriented graphene oxide films: an in situ Raman spectroelectrochemical study. J. Phys. Chem. C. 113, 7985–7989 (2009).
Zhu, J., Lu, L. & Zeng, K. Nanoscale mapping of lithium-ion diffusion in a cathode within an all-solid-state lithium-ion battery by advanced scanning probe microscopy techniques. ACS Nano 7, 1666–1675 (2013).
Celano, U. et al. Three-dimensional observation of the conductive filament in nanoscaled resistive memory devices. Nano Lett. 14, 2401–2406 (2014).
Li, J. J. et al. Microscopic investigation of grain boundaries in organolead halide perovskite solar cells. ACS Appl. Mater. Interfaces 7, 28518–28523 (2015).
Drogeler, M. et al. Spin lifetimes exceeding 12 ns in graphene nonlocal spin valve devices. Nano Lett. 16, 3533–3539 (2016).
Liu, X. et al. Scanning probe nanopatterning and layer-by-layer thinning of black phosphorus. Adv. Mater. 29, 1604121 (2017).
Okino, H. et al. In situ resistance measurements of epitaxial cobalt silicide nanowires on Si(110). Appl. Phys. Lett. 86, 233108 (2015).
This work has been supported by the Young 1000 Global Talent Recruitment Program of the Ministry of Education of China, the Ministry of Science and Technology of China (grant no. BRICS2018-211-2DNEURO), the National Natural Science Foundation of China (grants no. 61502326, 41550110223, 11661131002, 61874075), the Jiangsu Government (grant no. BK20150343), the Ministry of Finance of China (grant no. SX21400213) and the Young 973 National Program of the Chinese Ministry of Science and Technology (grant no. 2015CB932700). The Collaborative Innovation Center of Suzhou Nano Science and Technology, the Jiangsu Key Laboratory for Carbon-Based Functional Materials and Devices, the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the 111 Project from the State Administration of Foreign Experts Affairs are also acknowledged. F.H. acknowledges support from the Technion-Guangdong Fellowship. D. Lewis and R. Dechter from Nanonics, T. Yang from Park Systems, L. Pacheco from Concept Scientific Instruments, O. Krause from Nano World and W. Frammelsberger from Deggendorf Institute of Technology are acknowledged for helpful discussions. X. Jing (Soochow University) and E. Sahagún (Scixel) are acknowledged for support with figure preparation.
The authors declare no competing interests.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Hui, F., Lanza, M. Scanning probe microscopy for advanced nanoelectronics. Nat Electron 2, 221–229 (2019). https://doi.org/10.1038/s41928-019-0264-8
Conductive Atomic Force Microscopy of Semiconducting Transition Metal Dichalcogenides and Heterostructures
Advanced Functional Materials (2020)
2D Materials (2020)