In nanoscience, techniques based on Atomic Force Microscope (AFM) stand as a cornerstone for exploring local electrical, electrochemical and magnetic properties of microelectronic devices at the nanoscale. As AFM's capabilities evolve, so do the challenges of data analysis. With the aim of developing a prediction model for AFM mappings, based on Machine Learning, this work presents a step towards the analysis and benefit of Big Data recorded in the hyperspectral modes: AFM DataCube. The MultiDAT-AFM solution is an advanced 2000-line Python-based tool designed to tackle the complexities of multi-dimensional measurements and analysis. MultiDAT-AFM offers visualization options, from acquired curves to scanned mappings, animated mappings as movies, and a real 3D-cube representation for the hyperspectral DataCube modes. In addition, MultiDAT-AFM incorporates a Machine Learning algorithm to predict mappings of local properties. After evaluating two supervised Machine Learning algorithms (out of the eight tested) for regression, the Random Forest Regressor model emerged as the best performer. With the refinement step, a root mean square error (RMSE) of 0.18, an R 2 value of 0.90 and an execution time of a few minutes were determined. Developed for all AFM DataCube modes, the strategy and demonstration of MultiDAT-AFM are outlined in this article for a silicon integrated microelectronic device dedicated to RF applications and analyzed by DataCube Scanning Spreading Resistance (DCUBE-SSRM).

Scanning Capacitance Microscopy (SCM) is an essential technique in semiconductor failure analysis. It is widely known in studies of dopant profiles and carrier concentration. Not only that, SCM can be utilized as an electrical fault isolation tool to localize a failing transistor. Compared to Conductive Atomic Force Microscopy CAFM, the main advantage of SCM is that it can be used on both Silicon on Insulator (SOI) and Bulk Silicon wafers. In addition, SCM can scan over a relatively large area in a shorter time than conventional nanoprobing methods. This paper presents case studies illustrating the effectiveness of SCM for die level top-down failure analysis on 45nm node SOI and 14nm FinFET bulk Si technologies.

Milling experiments were undertaken with an AFM-in-SEM system to examine various layers in the SRAMs of a 3 nm finFET node technology chip. The simultaneous Conductive AFM and tomographic milling operation provided feedback as to the exact state of delayering in the sample. Despite the relatively rough appearance of some areas of the chip, the milling by the AFM tip was able to create local areas with high planarity. The AFM measurement provided the exact moment certain structures were polished through. Discussion of various electrical modes in the analysis that might provide clearer indications of breakthrough is also undertaken in this paper.

This paper presents advanced workflows that combine 3D Xray microscopy (XRM), nanoscale tomography, and electron microscopy to generate a detailed visualization of the interior of electronic devices and assemblies to enable the study of internal components for failure analysis (FA). Newly developed techniques such as the integration of deep-learning (DL) based algorithms for 3D image reconstruction are also discussed in this article. In addition, a DL-based tool (called DeepScout) is introduced that uses high-resolution 3D XRM datasets as training data for lower-resolution, larger field-of-view datasets and scales larger-volume data using a neural network model. Ultimately, these workflows can be run independently or complementary to other multiscale correlative microscopy evaluations, e.g., electron microscopy, and will provide valuable insights into the inner workings of electronic packages and integrated circuits at multiple length scales, from macroscopic features on electronic devices (i.e., hundreds of mm) to microscopic details in electronic components (in the tens of nm). Understanding advanced electronic systems through X-ray imaging and electron microscopy, and possibly complemented with some additional correlative microscopy investigations, can speed development time, increase cost efficiency, and simplify FA and quality inspection of printed circuit boards (PCBs) and electronic devices assembled with new emerging technologies.

In recent years, scanning probe microscopy (SPM) has drawn substantial attention for subsurface imaging, since the ultrasharp AFM tip (≈ 10 nm in radius) can deliver and detect, mechanical and electrical signals right above the material’s 3D volume with which it is directly interacting. Electrostatic force microscopy, or EFM, is one of the most common atomic force microscopy (AFM) variants for electrical property characterization. In this work, we demonstrate a method to significantly improve EFM’s subsurface imaging capability. Unlike conventional EFM, where an AC bias is applied to the cantilever, we applied two out of phase AC biases to adjacent subsurface lines and image the resulting cantilever response at the surface. The resulting remote bias induced EFM (RB-EFM) amplitude shows decent contrast of metal lines with a 2.4 μm spacing buried up to 4 μm beneath the surface. This novel method may resolve lines with a horizontal spacing of less than 130 nm at such depth and wider lines to at least 6 μm in depth. In addition, the results are compared with conventional EFM and KPFM that detects subsurface structure with two independent DC biases. A COMSOL simulation model has been developed that reproduces the essential features of the measurement and explains the improvement of subsurface imaging with RB-EFM compared to other electrostatic force imaging techniques. We show, that by biasing independent lines at a small delta in frequency from the cantilever resonance, multiple line traces can be differentiated in the RB-EFM image.

A new scanning capacitance microscope, with an optimized, modern RF circuitry is described. The new design results in improved ease of use and sensitivity. We will discuss the design details and show application examples on semiconductor devices and ferroelectric materials.

The results of analyses on a commercially available 7 nm SRAM, using an in-situ AFM inside a SEM, are presented. In addition to typical results for conductive AFM, a novel method is described that uses the SEM beam to prepare a region for additional material removal, thus bringing out clearer electrical data. This would be of exceptional value for technology nodes using cobalt as a contact material. Finally, techniques making use of the current from the SEM beam as the source of current during the measurement are described. The technique may have value for well resistance measurements using in-situ structures on live product, a survey of junction health, or the localization of point defects.

A novel method of fabrication of all-metal probes for scanning probe microscopy was developed. The motivation for this work was to develop probes and a method of fabricating them, which can be applied to measure friction of pairs of many different materials at the nanoscale. The main process of the presented manufacturing technique is nickel electrodeposition. The other steps are similarly simple and cheap. Moreover, the technique can be easily modified to manufacture probes of different materials and with different tip shapes.

In 1986 the Atomic Force Microscope (AFM) was invented by Gerd Binnig, Christoph Gerber, and Calvin Quate [1]. Since then, numerous analytical techniques have been developed and implemented on the AFM platform, evolving into what is collectively called the Scanning Probe Microscope (SPM). The SPM has since become well established as a mainstream analytical instrument with a continually increasing role in the development of nanoscale semiconductor technologies providing critical data from initial concept to technology development to manufacturing to failure analysis [2]. Scanning Capacitance Microscopy (SCM) has a longstanding, well-established track record for detecting dopant-related mechanisms that adversely affect device performance on planar (Field Effect Transistor) FETs as well as other structures (e.g., diodes, capacitors, resistors). The semiconductor industry’s transition to three dimensional FinFET devices has resulted in many challenges with regard to device analysis. This is especially true when it is necessary to perform detailed dopant analysis on a specific device; the device may be comprised of a single or multiple fins that have been called out specifically through traditional fault localization techniques. Scanning Capacitance Spectroscopy (SCS) is an analytical method, implemented on the SCM platform in which a series of DC bias conditions is applied to the sample and the carrier response is recorded using SCM [3]. SCS has a proven history of highlighting dopant related anomalies in semiconductor devices, which, in some instances, might not otherwise be “visible”. This paper describes successful application of SCM and SCS in showing, in full detail, a dopant-related failure mechanism on an individual, location-specific 14 nm FinFET.

In this paper, the authors report their successful attempt to acquire the scanning nonlinear dielectric microscopy (SNDM) signals around the floating gate and channel structures of the 3D Flash memory device, utilizing the custom-built SNDM tool with a super-sharp diamond tip. The report includes details of the SNDM measurement and process involved in sample preparation. With the super-sharp diamond tips with radius of less than 5 nm to achieve the supreme spatial resolution, the authors successfully obtained the SNDM signals of floating gate in high contrast to the background in the selected areas. They deduced the minimum spatial resolution and seized a clear evidence that the diffusion length differences of the n-type impurity among the channels are less than 21 nm. Thus, they concluded that SNDM is one of the most powerful analytical techniques to evaluate the carrier distribution in the superfine three dimensionally structured memory devices.

Scanning nonlinear dielectric microscopy is continuously developed as an AFM-derived method for 2D dopant profiling of semiconductor devices. In this paper, the authors apply 2D carrier density mapping to Si and SiC and succeed a high resolution observation of the SiC planar power MOSFET. Furthermore, they develop software that combines dC/dV and dC/dz images and expresses both density and polarity in a single distribution image. The discussion provides the details of AFM experiments that were conducted using a Hitachi environmental control AFM5300E system. The results indicated that the carrier density decreases in the boundary region between n plus source and p body. The authors conclude that although the resolutions of dC/dV and dC/dz are estimated to be 20 nm or less and 30 nm or less, respectively, there is a possibility that the resolution can be further improved by using a sharpened probe.

Two-dimensional semiconductors such as atomically-thin MoS2 have recently gained much attention because of their superior material properties fascinating for the future electronic device applications. Here we investigate the nanoscale dominant carrier distribution on atomically-thin natural and Nbdoped MoS2 mechanically exfoliated on SiO2/Si substrates by using scanning nonlinear dielectric microscopy. We show that a few-layer natural MoS2 sample is an n-type semiconductor, as expected, but Nb-doped MoS2, normally considered as a p-type semiconductor, can unexpectedly become an n-type semiconductor due to strong unintentional electron doping.

With decreasing device sizes, nanometer-sized defects on the wafer substrates can already limit the performance of the devices. The detection and precise classification of these defects requires additional characterization methods with a resolution in the nanometer range. It is well known that AFM can measure not only surface morphology but also mechanical and electrical properties. However, the versatility of AFM is not fully utilized in industrial applications due to the various limitations. Various limitations include low throughput and tip life in addition to the laborious efforts for finding the defects in inline automated defect review (ADR). In this paper we introduce the ADR AFM with mechanical and electrical characterization capability of defects in addition to high throughput, high resolution, and non-destructive means for obtaining 3D information for nm-scale defect review and classification.

The carrier distribution in solar cell is important evaluation target. Scanning nonlinear dielectric microscopy is applied to the cross section of phosphorus implanted emitter in monocrystalline silicon solar cell and visualizes the carrier distribution quantitatively. The effective diffusivities of phosphorus are estimated from the experimental results. Then, the three-dimensional carrier distribution is simulated. The experimental and simulation results show good correlation.

High resolution observation of density of interface states (Dit) at SiO2/4H-SiC interfaces was performed by time-resolved scanning nonlinear dielectric microscopy (tr-SNDM). The sizes of the non-uniform contrasts observed in the map of Dit were in the order of several tens of nanometers, which are smaller than the value reported in the previous study (>100 nm). The simulation of the tr-SNDM measurement suggested that the spatial resolution of tr-SNDM is down to the tip radius of the cantilever used for the measurement and can be smaller than the lateral spread of the depletion layer width.

High-resolution and high-sensitivity detection of free carriers in semiconductors is critical due to the trend of device miniaturization and diversification. To address this need, the AFM-based techniques of scanning spreading resistance microscopy, scanning capacitance microscopy, scanning nonlinear dielectric microscopy (SNDM), scanning microwave impedance microscopy, and scanning microwave microscopy are used. This paper demonstrates enhanced SNDM with stepwise dC/dV and dC/dz imaging, qualitative analysis, quantitative analysis, and artifact-free carrier-density profiling of semiconductor devices. The trace mode in enhanced SNDM is switched between contact (dC/dV measurement) state and non-contact (dC/dz measurement) state for every line scan whereby the sampling intelligent scan mode is switched these states every pixel. Using IMEC Si standards and Si power MOSFET as examples demonstrates that this SNDM methodology can provide qualitative, quantitative, and artifact-free carrier density profiling of semiconductor devices.

It is widely acknowledged that Atomic force microscopy (AFM) methods such as conductive probe AFM (CAFM) and Scanning Capacitance Microscopy (SCM) are valuable tools for semiconductor failure analysis. One of the main advantages of these techniques is the ability to provide localized, die-level fault isolation over an area of several microns much faster than conventional nanoprobing methods. SCM, has advantages over CAFM in that it is not limited to bulk technologies and can be utilized for fault isolation on SOI-based technologies. Herein, we present a case-study of SCM die-level fault isolation on SOI-based FinFET technology at the 14nm node.

The transistor structure of memory devices and other cutting-edge semiconductor devices has become extremely minute and complicated owing primarily to advances in process technology and employment of three-dimensional structures. Among the various approaches to improve the device performance and functionality, optimizing the carrier distribution is considered to be quite effective. This study focuses on scanning nonlinear dielectric microscopy (SNDM), a capacitance-based scanning probe microscopy technique. First, to evaluate SNDM's potential for high-resolution measurement, the most commonly used metal coated tip with a tip radius of 25 nm was used to measure a quite low-density impurity distribution. Then, after confirming that the SNDM's S/N ratio was sufficiently high for the smaller probe tip, an ultra-fine diamond probe tip with a nominal tip radius of lesser than 5nm as an SNDM probe tip to measure sub-20 nm node flash memory cell transistors was employed. Successful results were obtained and are reported.

Methods are available to measure conductivity, charge, surface potential, carrier density, piezo-electric and other electrical properties with nanometer scale resolution. One of these methods, scanning microwave impedance microscopy (sMIM), has gained interest due to its capability to measure the full impedance (capacitance and resistive part) with high sensitivity and high spatial resolution. This paper introduces a novel data-cube approach that combines sMIM imaging and sMIM point spectroscopy, producing an integrated and complete 3D data set. This approach replaces the subjective approach of guessing locations of interest (for single point spectroscopy) with a big data approach resulting in higher dimensional data that can be sliced along any axis or plane and is conducive to principal component analysis or other machine learning approaches to data reduction. The data-cube approach is also applicable to other AFM-based electrical characterization modes.

Voltage contrast (VC) mode inline E-beam inspection (EBI) at post contact layer provides electrical readout of critical yield signals at an early stage, which could be months before a wafer reaches functional test. Similar to the passive voltage contrast (PVC) technique that is widely used in failure analysis labs, inline VC scanning is based on scanning electron microscopy, where a low keV electron beam scans across the wafer. Conductive atomic force microscopy (CAFM) was successfully implemented as a characterization method for inline VC defects. In this paper, three challenging VC defect analysis case studies are considered: bright voltage contrast (BVC) gate to active short, BVC Junction leakage, and Dark Voltage Contrast gate contact open. Defects exhibiting a hard electrical short, junctional leakage, and open gate contact are used to illustrate how CAFM provides a powerful and comprehensive solution for in-depth characterization of the inline VC defects.