Who needs an electrical back-contact after all?

Abstract

Atomic force microscopy (AFM) is essential for nanoscale material characterization, yet its electrical modes are fundamentally constrained by the requirement of a physical back-contact (BC). The sample under study often requires complex and destructive sample preparation, such as cleaving, metallization, and the use of conductive adhesives, to close the electrical circuit. Due to their destructive nature and applicability only to small coupon samples, electrical AFM modes are generally incompatible with contamination-controlled, automated, in-line semiconductor metrology. To overcome these barriers, we introduce electron-beam excited AFM (EB-AFM). In this configuration, a low-energy electron beam is focused near the AFM probe to replace the physical BC, acting as a remote and reconfigurable electrode. We elucidate the fundamental parameters governing this electron-beam stimulation and demonstrate contact-free electrical mapping on 2D materials, III–V semiconductors, and fully integrated device structures. Our results show that EB-AFM achieves defect contrast and sensitivity comparable to conventional methods without requiring any sample modification. By eliminating the back-contact constraint and enabling wafer-scale compatibility, EB-AFM provides a pathway toward non-destructive, fully automated electrical metrology, thereby broadening the scope of nanoscale device characterization.

---

Umberto Celano

Umberto Celano is an Associate Professor and Research Director at Arizona State University's School of Electrical, Computer and Energy Engineering. He earned his PhD in physics from KU Leuven. Dr Celano's independent group focuses on the intersection of materials science and semiconductor metrology. His lab employs a comprehensive suite of physical probes—spanning light, X-ray, acoustic, electron-beam, and scanning probe techniques—to explore emerging materials and next-generation architectures. The group's specialized interests in nanoelectronics include nanosheet transistors, advanced memory, thermal management, heterogeneous integration metrology, nitrogen-vacancy centers, and 2D materials.

---

Introduction

Electrical scanning probe microscopy is now a staple of material characterization with an impact on the broader community of materials scientists, engineers and nanotechnologists. Starting from the late twentieth century, the combination of atomic force microscopy (AFM) with electrical secondary imaging modes has begun to offer a wide range of analytical techniques that combine nanoscale-resolved morphology from AFM with various readouts.^1–3 By applying either a static (DC) or alternating (AC) voltage between the tip and the sample, various phenomena, such as tunneling current, conductivity, local surface potential, electrochemical reactions, carrier concentration, and piezoelectric response, among others, can be evaluated.^4–6 The list of techniques is long, and multiple publications describe them in great detail.^7–9 Based on the number of publications, some of the most successful techniques are scanning tunneling microscopy (STM), Kelvin probe force microscopy (KPFM), scanning capacitance microscopy (SCM), piezoresponse force microscopy (PFM), conductive atomic force microscopy (C-AFM), and scanning spreading resistance microscopy (SSRM), among others.^10–14 Since their inception, these methods have been critical for addressing complex challenges in material synthesis, characterization, and optimization, spanning from fundamental research to industrial applications such as semiconductor metrology.¹⁵

While AFM methods for probing surface morphology have been integrated into industrial manufacturing processes, including quality control for automotive, aerospace, and semiconductor manufacturing, electrical AFM modes remain limited to small samples and continue to be measurements that are only partially automated.¹⁶ Sophisticated instruments for AFM morphology, such as those deployed in modern semiconductor cleanrooms, are capable of remarkable scanning stability, large fields of view, automation, and high-speed data acquisition, and can potentially perform the majority of electrical AFM modes. However, a critical design constraint prevents their adoption in complete in-line wafer and panel metrology: the requirement for an electrical back-contact (BC).¹⁷ Whether for two-probe DC techniques, such as C-AFM and SSRM, or AC-biased modes, like SCM, PFM, and KPFM, a physical electrical connection to the sample is essential to close the circuit or establish a potential difference. This reliance on back-contact constitutes a significant roadblock, preventing the automated deployment of electrical AFM modes in complex manufacturing environments. Using semiconductor metrology as an example, when measuring wafers during chip production, the need for BC forces the manufacturing flow to stop, breaking the entire wafer and obtaining a small coupon, and subsequently creating a conductive region by either using ion beam deposition or applying conductive epoxy glue or Ag-paste to create a connection between the sample surface and the AFM chuck. Needless to say, none of these steps are compatible with the level of low contamination required to continue the chip manufacturing process; thus, the entire wafer lot is often scrapped as a result. These steps add prohibitive cost and complexity, which have prevented the adoption of electrical AFM modes for in-line metrology, even forty years after these modes were introduced.

Furthermore, when targeting ultra-scaled devices, forming an electrical back-contact is complicated by the need to navigate specific electrical signal paths. Great care must be taken to avoid excessive contact resistance or interference with adjacent components. Ultimately, the placement and material selection of the back-contact significantly influence measurement quality; this is particularly critical when characterizing high-resistivity layers or integrated solid-state devices. In these scenarios, high series resistance within the tip–sample electrical loop can severely degrade the signal-to-noise ratio and overall sensitivity. This limitation applies to virtually all two-probe AFM techniques, hindering the full potential of these methods. Consequently, despite their widespread use in offline metrology labs and academic environments, electrical AFM modes have not yet been widely adopted for in-line wafer inspection.

To address the limitations imposed by physical back-contacts, we present a novel solution that replaces the mechanical connection with a low-energy electron beam focused on the sample surface near the AFM probe. We refer to this configuration as electron-beam excited AFM (EB-AFM). This approach yields results comparable to those of conventional techniques while unlocking applications previously inaccessible due to geometric or material constraints. In this work, we first describe the experimental setup and the key parameters that govern e-beam stimulation. We then demonstrate the efficacy of this strategy by performing contact-free electrical characterization on various samples, such as 2D materials, metals, and III–V semiconductors, showcasing the competitive defect-mapping capabilities of electron-beam excited conductive AFM (EBC-AFM). Finally, we discuss the potential for extending this e-beam excitation strategy to other modes, specifically scanning spreading resistance microscopy and scanning tunneling microscopy, as well as its applications to fully integrated device structures.

Results and discussion

Fig. 1 shows a general comparison between a conventional setup used for electrical AFMs (Fig. 1a), and the EBC-AFM concept proposed in this work (Fig. 1b). Here, we operate the AFM system with a self-sensing (piezo-resistive) probe, which is inside the chamber of a scanning electron microscope (SEM). In our setup, we utilize an AFM-in-SEM module (Litescope 2.0, NenoVision) in conjunction with various SEM systems, including a Nova 200 NanoLab and Helios 5 UX (Thermo Scientific). The Pt-coated self-sensing probe operates in contact mode, utilizing a Wheatstone bridge to replace the traditional optical beam deflection system for the distance control feedback loop that follows the sample surface. This is a well-established implementation of a compact AFM system, with recent demonstrations of high-speed and high-precision operations when deployed for metrology applications.¹⁸ In this implementation (Fig. 1b), the EBC-AFM setup also offers the advantage of operating in a vacuum, which is consistent with the SEM chamber environment. A primary validation of this setup is reported elsewhere, where the authors report on the use of EBC-AFM for the analysis of 2D materials including isolated exfoliated flakes, and as-grown MoS₂ on sapphire.¹⁷

Fig. 1 (a) Basic schematic of the conventional setup of an electrical AFM. It utilizes a laser-photodiode combination system for probing and requires a physical back-contact to apply the tip–sample voltage for injecting charge carriers. (b) Schematic illustrating the basic setup of an EBC-AFM, which utilizes an e-beam to inject charge carriers without requiring a back-contact. For probing purposes, a Wheatstone bridge-supported self-sensing mechanism is used (as shown in the inset).

Fig. 2 illustrates the core concept of e-beam-excited C-AFM. Here, the e-beam impinges on the sample surface in proximity to the area measured by the probe (i.e., a few µm to a few hundred µm). Previous work discussed the impact of e-beam position and distance on the EBC-AFM signal-to-noise ratio and sensitivity.¹⁷ The inset of Fig. 2a shows the equivalent circuit that is established when using the e-beam to inject charge and the probe as a path to electrical ground. In summary, we observe that the beam current of the SEM can be used to regulate the charge carriers injected at the tip–sample junction, and the measurements can be performed with a distance between the e-beam and the probe. Depending on the specific nature of the thin film under study, i.e., its thickness, conductivity, uniformity, and substrate, the e-beam can act (a) as a local source of electrical potential establishing a difference between the probe and the point where the e-beam lands, which creates a force that pushes charges to ground and (b) as a current source, thus pumping electrons into the sample and forcing their collection at the probe. These two effects are visible in more detail in Fig. S1 of the SI, where we compare measurements collected for bulk conductors and thin layers grown on a thick dielectric. A combination of the two effects is also possible in situations where the sheet resistances of the sample and the substrate create a competing mechanism. Examples of different material systems are shown in Fig. 2b–d, where we report the current channel of EBC-AFM for the case of graphene grown on Cu (Fig. 2b), monolayer MoS₂ grown on SiO₂/Si (Fig. 2c), and 41.5 µm thick GaN grown on sapphire (Fig. 2d). Notably, for each of the three samples, we can observe spatially resolved features in the current channel with lateral resolution ranging from hundreds of nm down to <20 nm, as in the case of step terraces in GaN, as shown in Fig. 2d. Here, we see the appearance of dislocations in correspondence with the step terraces, as indicated by the inset of Fig. 2d. EBC-AFM was performed as depicted in Fig. 2a, where the beam shines 100 µm away from the scanning probe in spot mode at 5 kV acceleration voltage (V_acc) and 0.4 nA beam current (I_beam). No voltage was applied between the probe and chuck. Similar results can be obtained by statically positioning the e-beam at a distance of hundreds of µm from the area under test. While these are used here as representative model systems, their diversity in thickness and surface conductivity suggests a wide range of applications for EBC-AFM where the presence of a physical back-contact is not feasible or inconvenient.

Fig. 2 (a) Schematic showing the tip–sample resistance during EBC-AFM and the equivalent circuit of EBC-AFM. (b) Current map of monolayer graphene grown on Cu. (c) Current map of a monolayer of MoS₂ grown on SiO₂/Si. (d) Current map of GaN showing step terraces and a circular dislocation spot in a zoomed-in view. Additionally, a line profile of the current value is presented from the section of the sample surface indicated by the black arrow.

Conventional C-AFM relies on contact-mode operation, where a bias is applied between the sample and an ultra-sharp conductive probe. This technique functions under relatively low mechanical loads, typically with contact forces in the range of 1 to 500 nN. In contrast, other established electrical modes necessitate distinct operational regimes: SSRM utilizes high-pressure contact to ensure an ohmic connection, while scanning tunneling microscopy operates across a vacuum gap without any physical contact (Fig. 3). SSRM is widely utilized for 2D carrier profiling in semiconductors, operating in a high-force regime (often exceeding several µN). This high pressure is necessary to plastically deform the sample surface, penetrate native oxides, and establish a stable β-Sn pocket in the area of the tip–sample junction (for Si samples).¹⁹ The resulting spreading resistance is measured using a logarithmic amplifier to map carrier concentrations across a wide dynamic range by comparing the resistance profile obtained with a calibration reference sample. Conversely, STM relies on the quantum mechanical tunneling effect rather than physical contact. In this mode, a bias voltage drives a tunneling current across a sub-nanometer gap separating the tip and the conductive surface. Because the tunneling current is exponentially dependent on the tip–sample separation distance, STM provides sub-nanometer (∼0.1 nm) resolution of both surface topography and the local electronic density of states.²⁰ Given their principles of operation, practical implementation for both techniques can be based on e-beam excited electrical charge injection.

Fig. 3 (a and b) Schematics showing the proposed EB-SSRM and EB-STM setup without any BC. (c) SSRM mapping of the Infineon dopant calibration sample and corresponding SEM image showing the landed probe on the sample surface during scanning. (d) STM map on a polycrystalline Cu surface obtained using EB-STM.

Fig. 3a shows the schematic operation of EB-SSRM. Here, the self-sensing probe was equipped with a diamond coating (Fig. S2 of the SI), and the measurement was performed with high pressure between the tip and the surface, in order to minimize the tip–sample contact resistance and make the spreading resistance established at the point of maximum pressure under the tip apex dominant. Multiple reports provide a more detailed description of conventional SSRM operations.^21–23 For our measurement, we use a dopant calibration sample supplied by Infineon Technologies AG, comprising multiple n-type and p-type implanted stripes with systematically varying doping concentrations. Controlled carrier densities were achieved through ion implantation, and secondary ion mass spectrometry (SIMS) data provided by Infineon Technologies AG confirmed a uniform dopant distribution within the upper 200 nm of the surface region. The implanted stripes display widths of 0.5 µm, 1.0 µm, and 2.0 µm, arranged with a periodic pitch of 2.5 µm. Here, we eliminate the need for a BC and operate the measurement with the e-beam positioned 300 µm away from the area of interest in spot mode at V_acc = 5 kV and I_beam = 0.4 nA. The results are shown in Fig. 3c, where n-type regions appear on the left side of the staircase-like profile. These observations demonstrate the capability of the EB-SSRM implementation to resolve both the spatial distribution and the doping of semiconductor structures with sub-micrometer precision. Despite these results, current self-sensing probe technology remains a bottleneck for high-pressure applications. The conventional monolithic design, where the piezoresistive element is embedded directly in the cantilever, restricts the tuning of critical parameters, such as the spring constant and thermal budget.^24,25 To address these limitations, new designs such as tri-layer AFM probe technology offer a robust solution.²⁶ By structurally decoupling the sensing element from the mechanical lever, the tri-layer architecture enables high stiffness and optimized coatings required for SSRM without compromising the sensor's performance or thermal stability.

In addition to SSRM, we report on the attempt to perform STM using an e-beam excited electrical charge injection mechanism (Fig. 3b). To this end, we selected the surface of a bulk metal as a sample, cleaned the surface by mild annealing (100 °C for 10 min), and then stored the sample in a high-vacuum environment. In this case, the e-beam was positioned 400 µm away from the area of interest. Fig. 3d shows the surface of polycrystalline copper (Cu) obtained from the experiment, where the current between the tip and the sample was used for the feedback of the z-actuator, i.e., STM constant current feedback mode (Fig. 3b). It is worth noting that the scan range of the AFM system used here is not compatible with that required for high-resolution STM, particularly in maintaining the probe's position with high accuracy at the sub-nanometer scale. Additionally, the base vacuum level was 1 × 10⁻⁵ mbar, which is considered suboptimal for STM, typically operated under ultrahigh vacuum (UHV) conditions.^27,28 However, we report that the basic operations can be performed, and the tip–sample current (i.e., 1 nA) generates a stable contact, allowing for the imaging of the Cu surface, where multiple islands were detected with grain sizes ranging from 200 to 500 nm. While future work is clearly required to support the possibility of extending the family of EB-induced AFM methods to SSRM and STM, the results presented here are encouraging.

Finally, we focus on the practical application of e-beam excitation for site-specific characterization in fully fabricated chips. This is an important area of application for electrical AFM modes with ample use of these techniques for failure analysis (FA), process development, and reliability assessment. Emerging integrated logic devices such as gate-all-around field-effect transistors (GAAFETs) and 3D memory concepts such as current 3D NAND, and future designs for 3D dynamic random-access memory (DRAM), all depend on intricate three-dimensional structures to sustain performance scaling and follow the industry's ongoing miniaturization goals.^29,30 While enabling improved performance and reduced power consumption, these integration schemes introduce substantial challenges for conventional characterization techniques, which often lack the spatial resolution or sensitivity required to probe three-dimensional electrical and material properties at the nanoscale. In particular, chiplets and multi-die 3D integration pose a high level of complexity in accessing the area of interest in modern chips.³¹ This is generally addressed by a combination of material removal methods, including grinding, polishing, and the use of plasma-assisted focused ion beam (PFIB), to access the area of interest (Fig. 4a). However, access to the area of interest does not imply the ability to perform a localized electrical measurement, i.e., C-AFM or others. In many cases, the device under test can be further electrically isolated by the presence of a junction, shallow trench isolation, or other electrical obstacles that limit the application of DC and AC signals between the tip and the sample. A simple example is reported in Fig. 4b for the case of a modern 3D NAND vertical memory. Here, modern 3D NAND devices can have 200+ worldline layers, which translates into a vertical channel with a length exceeding 10 µm, resulting in substantial series resistance when measuring conductivity in the channel using an AFM probe. Similarly, a scaled logic device, i.e., a transistor, could be isolated from the substrate because it is built using silicon-on-insulator technology, or because it is placed on a series of doped wells.³² For all these applications, our results indicate that e-beam excited probing with AFM can offer a potential solution. Fig. 4b shows an example of EBC-AFM obtained in a de-processed 3D NAND memory array, after removing the top contacts and exposing the area of the vertical poly-Si channel. The bottom insets in Fig. 4b show the region under test, as imaged by scanning electron microscopy (left) and EBC-AFM (right). Here, the contrast in the current map of EBC-AFM is generated without any physical back-contact by leveraging the presence of the SEM e-beam, thus obtaining high resolution probing of the conductive poly-Si channel at a fraction of the complexity for the sample preparation. Finally, another potential case study is represented by the analysis of materials that are grown selectively and thus are locally isolated from any electrical grounding. Selective growth methods enable the direct integration of 2D materials onto target substrates or silicon complementary metal–oxide–semiconductor (CMOS) platforms, eliminating the need for a separate, often damaging, transfer process.³³Fig. 4c shows the case of selective area deposition of synthetic MoS₂ grown inside patterned Al₂O₃ trenches. Clearly, the absence of a continuous closed film (SEM image in Fig. 4c) would make it impossible to apply classic C-AFM. However, EBC-AFM can be directly applied to the area of interest without the need for any extra steps for the formation of a back-contact, as visible in the bottom inset of Fig. 4c, showing morphology and current for one of the isolated MoS₂ islands.

Fig. 4 (a) Schematic of the external view of 3D NAND and multi-die chip stack. (b) Schematic of 3D NAND showing the top view of the poly-Si channel and the corresponding cross-section. At the bottom, SEM and EBC-AFM images of the circular poly-Si channel from the top side. (c) Top: SEM image. Bottom: AFM morphology and current map of the selective area deposition of MoS₂ grown (CVD) inside patterned Al₂O₃ trenches. Scale bar is 1 µm.

Finally, we investigate the underlying contrast mechanism of EBC-AFM, which deviates substantially from classical two-contact measurements. At first glance, the use of an electron beam as a current source to drive a signal through the tip–sample junction might suggest that the system operates similarly to constant-current C-AFM. Historically, such configurations have been employed to mitigate the limitations of voltage-induced, back-contacted C-AFM and specifically to prevent current-induced sample damage or to provide contrast in regions with highly variable sheet resistance.³⁴

In those traditional systems, a constant current is maintained via an external ammeter coupled with a tunable voltage source. In EBC-AFM, however, the interaction between the electron beam and the sample surface is far more complex. These interactions trigger a richer dynamic of phenomena, generating various electromotive forces (EMFs) that drive current through the tip–sample junction, distinguishing it from a simple regulated power supply. Fig. 5 shows a comparison where the type of material and the geometrical confinement of the layers involved in electronic transport can determine a substantially different behavior compared to a constant current C-AFM experiment.

Fig. 5 (a) Schematic of EBC-AFM performed on a bulk conductive sample. (b) Schematic of EBC-AFM performed on an ultra-thin semiconductor on top of an insulating substrate. (c and d) Comparing the EBC-AFM current level in Au and MoS₂ samples in terms of e-beam energy (at I_beam = 0.8 nA) and beam current (at V_acc = 1 keV). The Au sample is 50 nm thick while the MoS₂ sample is a monolayer.

Fig. 5a and b show a comparison of two different cases of sample geometry and conductivity, which are used here to study two types of electromotive forces that can be observed in EBC-AFM. In the presence of a conductive surface, where the sample's geometry allows for the total penetration depth of the e-beam to reside in the layer under study (i.e., thick and conductive sample), the impinging e-beam and a large amount of secondary-generated charge carriers are all transferred to ground (and sensed) via the self-sensing probe. This is visible in the case of a thick Cu foil, as shown in the SI, where we observe a higher fraction of the e-beam current converted into EBC-AFM current (Fig. S1 of the SI). This behavior is analogous to constant-current C-AFM operation, in which the EBC-AFM current is governed by the e-beam current, with an additional contribution proportional to the material-specific secondary electron emission yield.

Interestingly for the case in Fig. 5, we select two films with a difference in thickness of ca. 10 times, and in both cases, we observe a strong dependence of the EBC-AFM current on the beam acceleration voltage. Although not identical in the two samples, we observed a dependence between the beam acceleration voltage and the EBC-AFM current, indicating that in both cases the e-beam will penetrate from the surface into the insulator beneath. Here, this is responsible for the combined effect of (a) a fraction of the impinging electrons that conduct through the topmost film into the self-sensing probe, and (b) a fraction of the e-beam current that is lost in the substrate with no direct contribution to the EBC-AFM current. In addition, an interesting correlation between beam current and EBC-AFM current (Fig. 5d) is observed, and the sample geometry plays a significant role in determining the proportionality factor between the two. This is reflected in a relatively constant response between the EBC-AFM current and the beam acceleration voltage (Fig. 5c), suggesting that a good fraction of the e-beam current is lost in the substrate with no direct contribution to the EBC-AFM current. However, we expect that the insulating substrate exposed to the e-beam will undergo substantial charging, as often reported using Kelvin probe force microscopy (KPFM) in the area exposed to the e-beam.^35,36 Importantly, in samples with a strong tendency to electrostatic doping, as in the case of MoS₂, the net effect of the fraction of e-beam in the substrate can lead to an additional surface potential contribution, acting as a local source of potential difference between the tip and the sample. It is noteworthy that in our experience, when aiming at the acquisition of 2D conductive profiles, this effect does not impact the local conduction in a detrimental way. This effect, which is clearly detrimental to SEM imaging, creates an additional electrostatic potential on the surface of the thin semiconductor, thereby contributing to the strength of the resulting electromotive forces that assist current collection in the EBC-AFM experiment. This behavior shows similarities to constant current C-AFM operation, where the EBC-AFM current is imposed by the e-beam current plus a component proportional to the material-specific surface charging, or to the nanoscopic implementation of the electron-beam-induced current (EBIC) method.³⁷

Conclusions

In summary, we have presented a transformative method for performing electrical AFM modes that eliminates the need for a physical back-contact. By replacing conventional ohmic connections with a remote, reconfigurable e-beam, we remove the primary obstacle to non-destructive nanoscale characterization. We have demonstrated that this approach is versatile across a wide range of materials, including III–V semiconductors, 2D materials, and fully fabricated device structures, and supports multiple modalities such as EB-SSRM and EB-STM. A key highlight of this work is the application to fully integrated 3D NAND structures, where EB-AFM successfully probed vertical poly-Si channels that were otherwise electrically isolated. This capability reduces sample preparation complexity and enables a “correlative metrology” workflow that combines structural and electrical analysis for multiple types of FA applications. Given the ubiquity of electron-beam systems in semiconductor manufacturing, we believe EB-AFM to be a ready-to-deploy solution that can unlock electrical modes on a full-wafer scale, enabling in-line metrology applications without interrupting the chip manufacturing process flow.

Conflicts of interest

There are no conflicts to declare.

Disclaimer

Certain commercial equipment, instruments, software, or materials are identified in this document to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, or to imply that the materials or equipment identified are necessarily the best available for the purpose.

Data availability

The supporting data have been provided as part of the supplementary information (SI). Supplementary information: Fig. S1 and S2. See DOI: https://doi.org/10.1039/d5nr05523g.

Acknowledgements

M. A. R. L., M. J. H., S. C., S. S., and U. C. acknowledge primary support from Applied Materials Inc. for materials characterization development. The authors acknowledge the help of Infineon Technologies AG, particularly Thomas Schweinböck, for providing the dopant calibration sample. Moreover, the authors also acknowledge Jiali Yao from Wonmo Kang's lab for providing the graphene sample and the assistance of facilities from the Eyring Materials Center at Arizona State University.

References

1. R. A. Oliver, Advances in AFM for the electrical characterization of semiconductors, Rep. Prog. Phys., 2008, 71, 076501.
2. P. Eaton and P. West, Atomic force microscopy, Oxford University Press, 2010.
3. P. P. Swain, M. Penedo and G. E. Fantner, A Comprehensive Analysis of Combined AFM/SEM Systems for In situ Nanoscale Characterizations and Multiparametric Correlative Microscopy, Microsc. Microanal., 2025, 31, ozaf110.
4. U. Celano, Electrical Atomic Force Microscopy for Nanoelectronics, Springer, 2019.
5. D. S. Jakob, H. Wang and X. G. Xu, Pulsed force Kelvin probe force microscopy, ACS Nano, 2020, 14, 4839–4848.
6. A. Gruverman, M. Alexe and D. Meier, Piezoresponse force microscopy and nanoferroic phenomena, Nat. Commun., 2019, 10, 1661.
7. G. Benstetter, R. Biberger and D. Liu, A review of advanced scanning probe microscope analysis of functional films and semiconductor devices, Thin Solid Films, 2009, 517, 5100–5105.
8. F. Giannazzo, E. Schilirò, G. Greco and F. Roccaforte, Conductive atomic force microscopy of semiconducting transition metal dichalcogenides and heterostructures, Nanomaterials, 2020, 10, 803.
9. F. J. Giessibl, Advances in atomic force microscopy, Rev. Mod. Phys., 2003, 75, 949.
10. G. Binnig, H. Rohrer, C. Gerber and E. Weibel, Surface studies by scanning tunneling microscopy, Phys. Rev. Lett., 1982, 49, 57.
11. M. Nonnenmacher, M. o'Boyle and H. K. Wickramasinghe, Kelvin probe force microscopy, Appl. Phys. Lett., 1991, 58, 2921–2923.
12. J. Matey and J. Blanc, Scanning capacitance microscopy, J. Appl. Phys., 1985, 57, 1437–1444.
13. A. Kholkin, S. Kalinin, A. Roelofs and A. Gruverman, in Scanning Probe Microscopy: Electrical and Electromechanical Phenomena at the Nanoscale, Springer, 2007, pp. 173–214.
14. D. Alvarez, J. Hartwich, M. Fouchier, P. Eyben and W. Vandervorst, Sub-5 nm-spatial resolution in scanning spreading resistance microscopy using full-diamond tips, Appl. Phys. Lett., 2003, 82, 1724–1726.
15. N. G. Orji, et al., Metrology for the next generation of semiconductor devices, Nat. Electron., 2018, 1, 532–547.
16. A. Moussa, A.-L. Charley and P. Leray, in Metrology, Inspection, and Process Control XXXIX, SPIE, 2025, pp. 218–227.
17. M. A. R. Laskar, et al., Electron–Beam Excited Conductive Atomic Force Microscopy for Back Contact Free, Wafer–Scale and In–Line Compatible Electrical Characterization of 2D Materials, Adv. Sci., 2025, e05113.
18. A. Ahmad, T. Ivanov, T. Angelov and I. W. Rangelow, Fast atomic force microscopy with self-transduced, self-sensing cantilever, J. Micro/Nanolithogr., MEMS, MOEMS, 2015, 14, 031209.
19. V. Domnich and Y. Gogotsi, Phase transformations in silicon under contact loading, Rev. Adv. Mater. Sci., 2002, 3, 1–36.
20. G. Binnig, H. Rohrer, C. Gerber and E. Weibel, 7 × 7 reconstruction on Si (111) resolved in real space, Phys. Rev. Lett., 1983, 50, 120.
21. M. A. R. Laskar, et al., The enduring legacy of scanning spreading resistance microscopy: Overview, advancements, and future directions, Appl. Phys. Rev., 2025, 12, 041305.
22. P. Eyben, T. Janssens and W. Vandervorst, Scanning spreading resistance microscopy (SSRM) 2d carrier profiling for ultra-shallow junction characterization in deep-submicron technologies, Mater. Sci. Eng., B, 2005, 124, 45–53.
23. A. Schulze, R. Cao, P. Eyben, T. Hantschel and W. Vandervorst, Outwitting the series resistance in scanning spreading resistance microscopy, Ultramicroscopy, 2016, 161, 59–65.
24. K. Xu, C. Ji, F. Wei and Z. Zhai, Recent Progress in Self-sensing Probe Technology in Atomic Force Microscope, Curr. Nanosci., 2021, 17, 808–818.
25. H. Mahmoodi Nasrabadi, M. Mahdavi, M. Soleymaniha and S. Moheimani, High resolution atomic force microscopy with an active piezoelectric microcantilever, Rev. Sci. Instrum., 2022, 93, 1979082.
26. N. Hosseini, et al., A hybrid polymer/ceramic/semiconductor fabrication platform for high-sensitivity fluid-compatible MEMS devices with sealed integrated electronics, arXiv, 2023, preprint, arXiv:2307.05221,  DOI:10.48550/arXiv:2307.05221.
27. P.-F. Chung, et al., Design and performance of an ultrahigh vacuum spectroscopic-imaging scanning tunneling microscope with a hybrid vibration isolation system, Rev. Sci. Instrum., 2024, 95, 033701.
28. C. Herbschleb, et al., The ReactorSTM: Atomically resolved scanning tunneling microscopy under high-pressure, high-temperature catalytic reaction conditions, Rev. Sci. Instrum., 2014, 85, 083703.
29. A. Goda, 3-D NAND technology achievements and future scaling perspectives, IEEE Trans. Electron Devices, 2020, 67, 1373–1381.
30. Q. Zhang, Y. Zhang, Y. Luo and H. Yin, New structure transistors for advanced technology node CMOS ICs, Natl. Sci. Rev., 2024, 11, nwae008.
31. D. Das Sharma, G. Pasdast, S. Tiagaraj and K. Aygün, High-performance, power-efficient three-dimensional system-in-package designs with universal chiplet interconnect express, Nat. Electron., 2024, 7, 244–254.
32. C. Zhang, J. Li, R. Dao and R. Ring, in International Symposium for Testing and Failure Analysis, ASM International, 2025, pp. 472–481.
33. F. Liu, et al., Site-selective growth of two-dimensional materials: strategies and applications, Nanoscale, 2022, 14, 9946–9962.
34. J. Weber, Improving the Reliability of Conductive Atomic Force Microscopy, 2024.
35. M. Kovařík, D. Citterberg, E. Paiva de Araújo, T. Šikola and M. Kolíbal, Understanding the Effect of Electron Irradiation on WS₂ Nanotube Devices to Improve Prototyping Routines, ACS Appl. Electron. Mater., 2024, 6, 8776–8782.
36. M. A. Stevens-Kalceff, Kelvin probe microscopy of localized electric potentials induced in insulating materials by electron irradiation, Microsc. Microanal., 2004, 10, 797–803.
37. R. Heiderhoff, R. M. Cramer and L. J. Balk, Nanoscopic EBIC technique in a hybrid SEM/SFM system, in Proceedings of International Reliability Physics Symposium RELPHY-96, IEEE, Dallas, TX, USA, 1996, pp. 366. https://ieeexplore.ieee.org/document/492144.

---