Tooling, Instrumentation, Equipment Challenges in Nanocharacterization

The nanotechnology sub-field of nanocharacterization focuses on techniques to analyze materials and processes, including living process, at nanoscale, such as electron microscopy and computer vision.

Introduction

Nanocharacterization, the suite of techniques used to analyze materials and processes at the nanoscale, is fundamental to advancing nanotechnology across diverse fields, from materials science and electronics to medicine and biology. Techniques such as electron microscopy (EM), scanning probe microscopy (SPM), and advanced optical methods provide unprecedented views into the nanoworld.1 However, despite significant progress, numerous instrumentation and tooling barriers persist, limiting resolution, speed, sensitivity, environmental control, and data interpretability. These challenges hinder the ability to fully understand and engineer materials and processes, including those within living systems, at the ultimate scale.3 This report identifies and details approximately 100 significant tooling, instrumentation, and equipment quandaries currently faced in nanocharacterization, drawing upon recent expert opinions and literature, with a focus on electron microscopy, scanning probe microscopy, optical methods, and associated computational analysis tools. The barriers are roughly prioritized based on their perceived impact and the frequency with which they are discussed in contemporary research and strategic documents.

I. Electron Microscopy (EM) Barriers

Electron microscopy, encompassing Transmission Electron Microscopy (TEM), Scanning Transmission Electron Microscopy (STEM), and Scanning Electron Microscopy (SEM), offers unparalleled spatial resolution but faces significant hurdles, particularly concerning beam-sample interactions, environmental control, and data handling.1

A. General Electron Microscopy Challenges

  1. Beam Damage (General): The high-energy electron beam essential for imaging inevitably interacts with the sample, causing structural and chemical alterations. This damage manifests as knock-on displacement (atoms ejected by direct collision) or radiolysis (bond breaking via ionization/excitation), fundamentally limiting observation time and achievable resolution, especially for sensitive materials like polymers, biological specimens, or certain catalysts.6 Persistence is due to the fundamental physics of electron-matter interaction; reducing energy to mitigate knock-on can increase radiolysis, creating unavoidable trade-offs.6 This limits the study of pristine structures and dynamic processes.
  2. Sample Preparation Artifacts (General EM): Preparing samples thin enough for electron transmission (typically <100 nm for TEM/STEM) or with appropriate surface conductivity/topography for SEM often introduces artifacts. Techniques like ion milling (e.g., FIB) can cause ion implantation, amorphization, or redeposition, while chemical fixation or staining for biological samples can distort native structures.8 These artifacts complicate the interpretation of true material structure and properties, persisting due to the inherent need to modify bulk materials for EM analysis.
  3. Vacuum Compatibility Constraints: EM requires high vacuum (<10−4 Pa) to prevent electron scattering by gas molecules and protect the electron source, precluding the study of many materials and processes in their native, ambient, or liquid environments. While environmental TEM (ETEM) and liquid-cell TEM exist, they introduce significant complexity and limitations.7 This barrier persists due to the fundamental requirement of electron beam propagation in vacuum, limiting in situ and operando studies under realistic conditions.
  4. Charging Effects in Insulating Samples: Non-conductive samples accumulate charge under electron beam irradiation, leading to image distortions, drift, and reduced resolution. While conductive coatings can mitigate this, they obscure surface details, add thickness affecting signal quality, and can introduce chemical artifacts, especially in liquid environments.6 The persistence lies in the insulating nature of many important materials (polymers, ceramics, biological tissues) and the limitations of mitigation strategies.
  5. Low Contrast for Low-Z Materials: Materials composed of light elements (e.g., carbon, oxygen, nitrogen in biological samples or polymers) exhibit weak scattering contrast in conventional EM modes (like bright-field TEM or secondary electron SEM). This makes distinguishing features or achieving high resolution difficult without staining or specialized techniques (e.g., Z-contrast STEM, phase contrast methods).9 The persistence is due to the weak interaction cross-section between high-energy electrons and low-atomic-number elements.
  6. Limited 3D Information from 2D Projections (TEM/STEM): Standard TEM/STEM provides 2D projection images, making it difficult to interpret complex 3D nanostructures unambiguously. While electron tomography reconstructs 3D volumes, it requires acquiring numerous tilted images, leading to high cumulative electron doses (exacerbating beam damage) and potential artifacts from the missing wedge of tilt angles.12 The challenge persists due to the projection nature of TEM and the technical difficulties/limitations of tilt-series acquisition.
  7. Depth Resolution Limitations (SEM): While SEM excels at surface topography, determining the depth or thickness of nanoscale features or layers beneath the immediate surface is challenging. The interaction volume of the electron beam extends significantly below the surface, limiting depth resolution compared to lateral resolution.2 This persists due to the physics of electron scattering within the sample volume.
  8. Quantitative Analysis Challenges (EDS/EELS): Accurately quantifying elemental composition using Energy Dispersive X-ray Spectroscopy (EDS) or Electron Energy Loss Spectroscopy (EELS) is complex. Factors like sample thickness, geometry, absorption, fluorescence, and detector efficiency must be carefully accounted for.5 Obtaining reliable quantitative data, especially for light elements (EDS) or complex interfaces, remains a significant challenge requiring sophisticated modeling and calibration, hindering routine quantitative nanoanalysis.
  9. Throughput Limitations: Traditional EM analysis, involving manual sample loading, area selection, focusing, and data acquisition, is inherently slow. Characterizing statistically relevant numbers of features or screening large sample areas is time-consuming.15 This low throughput hinders applications in quality control, high-throughput screening, and the analysis of heterogeneous materials, persisting due to complex instrument operation and manual intervention requirements.
  10. Cost and Accessibility: High-resolution electron microscopes, particularly aberration-corrected (S)TEMs, represent a major capital investment (>$1 million USD) and require specialized facilities (vibration isolation, electromagnetic field cancellation) and highly trained personnel.3 This high cost and complexity limit access for many researchers and institutions, hindering broader adoption and application of advanced EM techniques.

B. Transmission Electron Microscopy (TEM/STEM) Specific Challenges

  1. Aberration Correction Complexity and Stability: While aberration correctors enable sub-Ångström resolution 5, they are complex, expensive systems requiring meticulous alignment and environmental stability. Maintaining optimal correction over time, especially during in situ experiments with changing conditions (temperature, fields), is challenging.11 This complexity limits routine access to the highest resolutions and can introduce subtle image artifacts if not perfectly tuned.
  2. Resolution Limits in Environmental/Liquid Cells: The membranes (e.g., SiN) required to contain gas or liquid environments within TEM holders scatter electrons, degrading spatial and energy resolution compared to high-vacuum imaging.7 Thicker liquid layers exacerbate this effect. While graphene liquid cells offer thinner confinement 21, their fabrication, handling, and flow control remain challenging, limiting achievable resolution in realistic in situ environments.
  3. Electron Dose Management for Dynamics: Observing dynamic processes in situ requires repeated imaging, leading to high cumulative electron doses that can induce artifacts or halt the process being studied.7 Balancing the need for temporal resolution (requiring frequent images) with minimizing beam damage is a critical challenge, especially for sensitive materials or slow processes.23 This fundamental trade-off limits the types and duration of dynamic phenomena observable.
  4. Spatial Resolution vs. Temporal Resolution Trade-off: Acquiring images with high spatial resolution typically requires longer acquisition times (higher electron dose per image) to achieve sufficient signal-to-noise ratio (SNR). Conversely, capturing fast dynamics requires short acquisition times (low dose per image), often sacrificing spatial resolution or SNR.5 Optimizing this trade-off for specific dynamic processes remains a persistent challenge in in situ TEM.
  5. Drift Correction during In Situ Experiments: Sample drift, induced by thermal changes, mechanical instability, or environmental interactions, is a major issue during in situ TEM, blurring images and hindering high-resolution observation over time.25 While software-based correction exists, real-time, robust hardware drift correction integrated into microscopes, especially for high-temperature or complex experiments, is still lacking universal availability and perfect performance.25
  6. Accurate Temperature Measurement/Control at Nanoscale: Precisely measuring and controlling the local temperature of the specimen area under observation within the TEM is difficult, especially during heating/cooling experiments or in ETEM.25 Thermal gradients across the sample/grid, uncertainties in thermocouple readings, and beam heating effects complicate quantitative thermal studies.25 Lack of accurate local temperature knowledge hinders the extraction of reliable thermodynamic and kinetic data.
  7. Quantitative Strain Mapping Limitations: Techniques like geometric phase analysis (GPA) or peak-pair analysis from HRTEM or diffraction patterns allow strain mapping, but accuracy is limited by image quality, noise, aberrations, sample thickness variations, and the chosen analysis algorithms.18 Achieving reliable, quantitative strain mapping with nanometer resolution, especially in 3D or under dynamic conditions, remains challenging, hindering understanding of mechanical properties at the nanoscale.
  8. Electron Holography Phase Reconstruction Artifacts: Off-axis electron holography maps electromagnetic fields but relies on complex phase reconstruction algorithms sensitive to experimental parameters, noise, and mean inner potential effects.14 Artifacts in the reconstruction can lead to misinterpretation of field strengths and distributions, particularly for weak fields or complex structures. Robust, automated reconstruction remains an area of active development.
  9. Spectroscopy Signal Delocalization (EELS): The inelastic scattering signal in EELS, particularly low-loss plasmons, can be delocalized over several nanometers due to long-range Coulomb interactions. This limits the spatial resolution of chemical or electronic information obtainable from these signals compared to the probe size.12 This fundamental physical effect restricts the analysis of sharp interfaces or isolated nanoparticles using low-loss EELS.
  10. EELS Energy Resolution Limits (Non-Monochromated): Standard TEMs have an energy resolution of ~1 eV, limiting the ability to resolve fine details in EELS spectra, such as vibrational modes or subtle chemical shifts.5 While monochromators significantly improve resolution (down to meV), they reduce beam current (affecting SNR and acquisition speed) and add complexity and cost.5 This limits routine access to high-energy-resolution EELS.
  11. EDS Quantification for Light Elements: EDS struggles to detect and accurately quantify light elements (Z < 11) due to low X-ray fluorescence yields, absorption within the sample and detector window, and peak overlaps.5 While windowless or thin-window detectors improve sensitivity, reliable quantification remains challenging compared to heavier elements, hindering analysis of organic materials, oxides, and nitrides.
  12. Tomography Reconstruction Artifacts (Missing Wedge): Practical limitations on sample tilting (typically ±70°) result in a "missing wedge" of data in reciprocal space for electron tomography. This leads to anisotropic resolution and elongation artifacts in the reconstructed 3D volume, particularly in the direction perpendicular to the tilt axis.12 Overcoming these artifacts requires advanced reconstruction algorithms or alternative acquisition strategies, complicating routine 3D analysis.
  13. Cryo-EM Sample Preparation Challenges (Vitrification): Achieving optimal ice thickness (thin enough for transmission, thick enough to embed particles) and uniform particle distribution during plunge-freezing for single-particle analysis (SPA) or cryo-ET is challenging.26 Issues like particle aggregation, preferred orientation at air-water interfaces, and sample damage during blotting persist, hindering high-resolution structure determination for many biological macromolecules.
  14. Cryo-ET Lamella Preparation (Cryo-FIB): Preparing thin (~100-300 nm) lamellae from vitrified cells or tissues for in situ structural biology using cryo-FIB milling is technically demanding.26 Challenges include precise targeting, minimizing curtaining artifacts, preventing sample devitrification or contamination, and handling fragile lamellae.26 This complex workflow limits the throughput and accessibility of cellular cryo-ET.
  15. Cryo-EM Data Acquisition Throughput: Acquiring the large datasets (thousands to millions of particle images for SPA, hundreds of tilt series for cryo-ET) needed for high-resolution reconstructions is time-consuming, often requiring days of automated microscope operation.27 Optimizing acquisition strategies, improving automation reliability, and managing data flow remain key bottlenecks limiting overall structural biology throughput.27

C. Scanning Electron Microscopy (SEM) Specific Challenges

  1. Surface Sensitivity vs. Interaction Volume: While SEM primarily images surface topography using secondary electrons (SEs), the primary electron beam penetrates microns into the sample, generating backscattered electrons (BSEs) and X-rays from a larger volume. This limits true surface sensitivity and complicates the analysis of thin films or buried interfaces.2 Balancing surface information (SE) with compositional/structural information (BSE, EDS) from different depths remains a challenge.
  2. Resolution Limits in Conventional SEM: The resolution of conventional SEM (~1-5 nm) is often limited by the electron probe size, interaction volume, and signal detection efficiency, especially at low accelerating voltages needed for surface sensitivity or beam-sensitive samples.1 Achieving sub-nanometer resolution typically requires specialized field-emission guns, immersion lenses, or STEM-in-SEM capabilities, increasing cost and complexity.11
  3. Low Voltage SEM Operation Challenges: Operating SEM at low voltages (<5 kV) enhances surface detail and reduces charging/damage but suffers from lower brightness, larger probe sizes (poorer resolution), and reduced signal (lower SNR) compared to high-voltage operation.18 Optimizing resolution and signal quality at low kV requires advanced electron optics and detectors, posing a persistent instrumentation challenge.18
  4. Environmental SEM (ESEM) Resolution/Pressure Trade-off: ESEM allows imaging in gaseous environments (up to ~kPa) by using specialized detectors and differential pumping, enabling studies of hydrated or outgassing samples. However, gas scattering degrades resolution, and the achievable pressure is still far below ambient conditions.10 Balancing environmental pressure with imaging resolution remains a fundamental limitation.
  5. Quantitative BSE Imaging Complexity: Backscattered electron (BSE) intensity depends on the average atomic number (Z-contrast), but also on topography, crystal orientation (channeling contrast), and detector geometry. Extracting quantitative compositional information solely from BSE images is difficult without standards and careful calibration.12 This limits its use for precise quantitative mapping compared to EDS/WDS.
  6. EDS Spatial Resolution Limits in SEM: Although the SEM probe can be nanometer-sized, the interaction volume from which characteristic X-rays are generated for EDS is typically much larger (microns), especially at higher accelerating voltages. This limits the spatial resolution of EDS mapping in SEM compared to STEM-EDS.5 Achieving nanoscale elemental mapping in SEM requires low voltages or thin samples, often sacrificing signal intensity.
  7. EBSD Sample Preparation Requirements: Electron Backscatter Diffraction (EBSD) requires a highly polished, deformation-free sample surface for optimal pattern quality and indexing accuracy.29 Preparing such surfaces, especially for soft, multiphase, or brittle materials, can be challenging and time-consuming, limiting the routine application of EBSD for crystallographic analysis.
  8. EBSD Indexing Challenges (Nanomaterials, Deformed Materials): Indexing EBSD patterns from nanocrystalline materials (due to weak/diffuse patterns), highly deformed structures (due to pattern distortion), or complex phases can be difficult and prone to errors.3 Robust indexing algorithms and improved pattern quality are needed to reliably analyze these challenging materials.
  9. STEM-in-SEM Integration and Performance: While adding STEM detectors to SEMs offers transmission imaging capabilities at lower cost than dedicated (S)TEMs, performance (resolution, signal quality) is often limited by the SEM's electron optics (designed for surface imaging) and lower accelerating voltages.11 Achieving high-resolution STEM imaging and analysis within an SEM platform remains an ongoing development challenge.
  10. Beam-Induced Deposition/Etching: In SEM, interaction of the electron beam with residual hydrocarbons in the vacuum chamber or precursor gases (in Gas Injection Systems) can lead to unwanted deposition of carbonaceous material or localized etching. This contamination can obscure features or modify the sample surface during analysis, particularly during long acquisitions or focused beam work.28 Maintaining ultra-clean vacuum or carefully controlling beam parameters is crucial but challenging.

II. Scanning Probe Microscopy (SPM) Barriers

SPM techniques, including Atomic Force Microscopy (AFM) and Scanning Tunneling Microscopy (STM), provide exceptional surface topographical and property mapping capabilities but face challenges related to probe tips, scan speed, environmental sensitivity, and data interpretation.1

A. General Scanning Probe Microscopy Challenges

  1. Scan Speed Limitations: SPM relies on mechanically scanning a probe across the surface, pixel by pixel, limiting imaging speed compared to optical or electron microscopy.15 Capturing dynamic processes in real-time or imaging large areas quickly is challenging due to mechanical resonance frequencies of the scanner and cantilever, and feedback loop response times. This low throughput hinders dynamic studies and statistical analysis.31
  2. Probe Tip Durability and Consistency: The sharpness and chemical identity of the SPM probe tip are critical for resolution and data interpretation, but tips wear down or become contaminated during scanning, especially in contact modes or on rough surfaces.32 Fabricating tips with consistent geometry and long-term durability remains a major challenge, affecting reproducibility and quantification.16 Lack of reliable, long-lasting probes hinders routine high-resolution imaging and force measurements.34
  3. Tip-Sample Convolution Artifacts: The finite size and shape of the probe tip inevitably distort the measured topography, especially for features comparable in size to the tip radius. This "tip convolution" effect makes accurate dimensional measurements of sharp or complex nanostructures difficult without deconvolution algorithms, which require knowledge of the tip shape.3 Persistence stems from the physical nature of probe-based imaging.
  4. Image Artifacts (Drift, Noise, Feedback Errors): SPM images are susceptible to artifacts from thermal drift, mechanical vibrations, electronic noise, and imperfect feedback loop response.15 These can manifest as image distortion, streaks, or apparent height variations unrelated to the true surface topography, complicating interpretation, especially for atomic/molecular resolution imaging or subtle feature detection.17 Achieving ultimate stability often requires costly environmental isolation.17
  5. Difficulty Imaging High Aspect Ratio / Complex Structures: Imaging deep trenches, vertical sidewalls, or highly convoluted surfaces is challenging for SPM due to the tip's geometry (limited aspect ratio) and the inability of the feedback loop to track abrupt changes accurately.36 Specialized tips or scanning modes are often required, limiting routine analysis of complex 3D nanostructures.
  6. Environmental Sensitivity (Temperature, Humidity): SPM measurements, particularly those relying on sensitive force or tunneling current detection, are highly susceptible to environmental fluctuations like temperature changes (causing drift) or humidity variations (affecting capillary forces in ambient AFM).17 Maintaining stable environmental conditions is crucial but often difficult outside specialized vacuum or controlled-atmosphere systems.
  7. Data Interpretation Complexity (Multi-parameter): Many advanced SPM modes simultaneously measure multiple parameters (e.g., topography, adhesion, modulus, conductivity, potential). Deconvolving these signals and correlating them accurately with specific material properties can be complex, often requiring sophisticated models and assumptions about tip-sample interactions.15 Lack of straightforward interpretation hinders broader adoption of multi-property mapping.
  8. Calibration Challenges (Force, Distance, Signal): Accurate quantitative measurements (e.g., force spectroscopy, elastic modulus mapping, potential measurements) require careful calibration of the probe (spring constant, tip radius, sensitivity) and the instrument's response (scanner displacement, detector sensitivity).21 Performing reliable and traceable calibrations remains a non-trivial task, limiting the accuracy and comparability of quantitative SPM data.
  9. Limited Subsurface Information: Standard SPM techniques primarily probe the surface or near-surface region. Obtaining information about buried structures, interfaces, or bulk properties is generally not possible without destructive techniques (like AFM tomography) or integration with other methods.33 This surface sensitivity limits SPM's applicability for analyzing layered systems or internal device structures.
  10. Operator Skill Dependency and Reproducibility: Achieving optimal results, especially with advanced SPM modes or challenging samples, often requires significant operator expertise in tip selection, parameter tuning, and data interpretation.30 This reliance on skilled operators contributes to variability and challenges in reproducibility between different users and labs, hindering standardization and broader application.31

B. Atomic Force Microscopy (AFM) Specific Challenges

  1. Contact Mode Sample/Tip Damage: In contact mode AFM, the tip is dragged across the surface, potentially damaging soft samples (like biological molecules or polymers) or wearing down the tip itself due to friction and adhesion forces.4 While tapping mode reduces lateral forces, some tip-sample interaction is unavoidable. This limits the applicability of contact mode for delicate samples and contributes to tip wear issues.
  2. Tapping Mode Feedback Complexity: Maintaining optimal cantilever oscillation amplitude and phase in tapping mode requires sophisticated feedback control, especially on surfaces with varying topography or adhesion.16 Improper tuning can lead to imaging artifacts or loss of resolution. Developing robust feedback systems for high-speed, high-resolution tapping mode remains an engineering challenge.16
  3. Capillary Forces in Ambient Conditions: In ambient air, a thin layer of adsorbed water creates meniscus forces between the tip and sample, significantly affecting force measurements and imaging contrast, particularly in contact or tapping modes.21 These uncontrolled capillary forces complicate quantitative measurements and necessitate operation in controlled humidity, liquid, or vacuum environments for reliable results.
  4. Quantitative Nanomechanical Mapping Challenges (Modulus, Adhesion): Extracting accurate quantitative mechanical properties like Young's modulus or adhesion force from AFM force curves or specialized modes (e.g., PeakForce Tapping, AM-FM) is complex.37 Results depend heavily on the chosen contact mechanics model (e.g., Hertz, DMT, JKR), accurate tip shape/radius determination, cantilever calibration, and assumptions about material behavior, limiting reliability and comparability.31
  5. High-Speed AFM Limitations: While high-speed AFM (HS-AFM) systems exist, achieving video-rate imaging typically involves trade-offs, such as smaller scan areas, limitations on sample height variation, specialized small cantilevers, and potential compromises in force control or resolution.4 Balancing speed with image quality, scan size, and applicability to diverse samples remains a challenge for routine dynamic biological or materials studies.
  6. AFM in Liquid Environments (Signal Damping, Contamination): Imaging in liquid is crucial for biological samples but introduces challenges like viscous damping of the cantilever (reducing sensitivity/Q-factor), potential sample/tip contamination from the liquid, and difficulties in controlling tip-sample forces accurately.4 Optimizing AFM performance for stable, high-resolution imaging in physiological or reactive liquid environments remains difficult.
  7. Conductive AFM (CAFM) Tip Reliability: Standard metal-coated Si tips used for CAFM degrade quickly due to high current densities and mechanical friction, leading to unreliable and non-reproducible electrical measurements.34 While more durable solid metal or diamond tips exist, they are more expensive or stiffer (potentially damaging samples), posing a trade-off between durability, cost, and sample integrity.33 Consistent, durable, and affordable CAFM probes are needed.
  8. CAFM Data Interpretation (Contact Resistance, Spreading Effects): Interpreting CAFM current maps quantitatively is complicated by unknown tip-sample contact resistance, current spreading effects within the sample, and potential modification of the sample by the high electric fields or currents at the tip apex.38 Relating measured current to intrinsic material conductivity requires careful modeling and understanding of the nanoscale contact physics.
  9. Magnetic Force Microscopy (MFM) Resolution and Artifacts: MFM resolution is often limited by the size of the magnetic tip coating and the tip-lift height needed to separate magnetic from topographic signals.3 Furthermore, the tip's magnetic field can potentially alter the magnetic state of the sample, leading to measurement artifacts. Achieving high-resolution, non-perturbative magnetic imaging remains challenging, especially for weakly magnetic or complex domain structures.
  10. Kelvin Probe Force Microscopy (KPFM) Artifacts and Resolution: KPFM measures surface potential but is susceptible to artifacts from stray capacitance, variations in tip work function, and topographic crosstalk.15 Achieving high spatial resolution (~10s of nm) and accurate potential measurements simultaneously requires careful experimental setup, tip selection, and data processing to minimize these artifacts.

C. Scanning Tunneling Microscopy (STM) Specific Challenges

  1. Requirement for Conductive Samples: STM relies on quantum mechanical tunneling current between the tip and sample, fundamentally restricting its application to electrically conductive or semiconductive materials.17 Insulating surfaces cannot be imaged directly, limiting its use for many polymers, ceramics, and biological samples without conductive coatings (which obscure atomic detail). This is the primary limitation of STM.
  2. Tip Preparation and Stability (Atomic Sharpness): Achieving and maintaining an atomically sharp and stable tip apex is crucial for atomic resolution STM but is often challenging.2 Tip crashes, contamination, or changes in the apex structure during scanning can lead to loss of resolution or image artifacts. Reliable in situ tip preparation and characterization methods are needed but add complexity.
  3. Distinguishing Topographic vs. Electronic Effects: The STM tunneling current depends on both the tip-sample separation (topography) and the local density of electronic states (LDOS) of the sample and tip.17 Disentangling these contributions, especially on electronically inhomogeneous surfaces or when performing scanning tunneling spectroscopy (STS), can be difficult, complicating the interpretation of atomic-scale images and spectra.17
  4. Ultra-High Vacuum (UHV) and Cryogenic Requirements: Achieving stable atomic resolution and performing STS often requires UHV conditions to prevent surface contamination and cryogenic temperatures to reduce thermal drift and improve energy resolution.14 These requirements necessitate complex and expensive instrumentation, limiting the accessibility and applicability of high-performance STM.17
  5. Low Tunneling Current Sensitivity: The tunneling current is typically very small (pA to nA range), making STM measurements highly sensitive to electronic noise and external vibrations.17 Robust vibration isolation and low-noise electronics are essential but add significantly to system cost and complexity.17

III. Optical Microscopy & Spectroscopy Barriers

Optical methods, including super-resolution microscopy (SRM) and Raman spectroscopy, offer non-invasive characterization, often in physiological conditions, but face limitations in resolution (though improved by SRM), sensitivity, penetration depth, and labeling.1

A. Super-Resolution Microscopy (SRM) Challenges

  1. Resolution Limits (Technique Dependent): While SRM techniques (STED, SIM, SMLM like PALM/STORM) break the diffraction limit (~200-250 nm), they still have finite resolution limits, typically ranging from ~20-120 nm laterally depending on the method.39 Achieving resolutions approaching electron microscopy (<10 nm) remains challenging for most biological applications and requires significant optimization. The specific resolution depends heavily on the technique and experimental conditions.39
  2. Phototoxicity and Photobleaching: Many SRM techniques, particularly STED and SMLM, require high laser intensities or long acquisition times, leading to phototoxicity (damage to living cells) and photobleaching (irreversible destruction of fluorophores).39 This limits live-cell imaging duration and the ability to observe dynamic processes over extended periods without perturbing the system.42
  3. Labeling Density and Specificity Challenges: Achieving high labeling density (crucial for resolving dense structures in SMLM/STED) and ensuring label specificity remains difficult.39 Antibody labeling can suffer from linkage errors (distance between fluorophore and target) and steric hindrance, while fluorescent proteins may perturb function or have suboptimal photophysics.39 Developing smaller, brighter, more photostable probes (e.g., nanobodies, improved organic dyes) is critical but ongoing.39
  4. Artifacts in Image Reconstruction (SIM, SMLM): SIM relies on precise knowledge of illumination patterns and complex algorithms, making it susceptible to reconstruction artifacts if parameters are incorrect.43 SMLM reconstruction can suffer from artifacts due to fluorophore blinking (leading to over/undercounting), localization errors, and drift during acquisition, requiring careful calibration and sophisticated analysis software.39
  5. Limited Temporal Resolution (Especially SMLM): SMLM techniques require acquiring thousands of frames to reconstruct a single super-resolved image, resulting in poor temporal resolution (seconds to minutes per frame), making them unsuitable for many fast live-cell dynamics.39 While STED and SIM can be faster, there's often a trade-off between resolution, field of view, and speed.39
  6. Complexity and Cost of SRM Systems: SRM systems are often complex optical setups requiring careful alignment and calibration (especially STED) and represent a significant cost increase compared to conventional fluorescence or confocal microscopes.40 This complexity and cost limit their widespread accessibility and routine use, particularly outside specialized imaging centers.40
  7. Limited Penetration Depth in Tissue: Like conventional optical microscopy, SRM techniques suffer from limited penetration depth (typically tens of micrometers) in scattering tissues due to light scattering and absorption.42 While techniques like light-sheet illumination or tissue clearing can help, imaging deep within intact biological tissues or non-transparent materials at super-resolution remains a major challenge.42
  8. Quantitative SRM Challenges: Accurately counting molecules or quantifying structural parameters using SRM is challenging due to uncertainties in labeling efficiency, fluorophore photophysics (blinking, maturation), localization precision, and potential artifacts.39 Establishing robust methods for quantitative analysis and validating results remains an active area of research.39
  9. Multi-Color SRM Complexity: Performing simultaneous multi-color SRM requires careful selection of spectrally distinct fluorophores with appropriate photophysics for the chosen technique, minimizing crosstalk, and complex optical setups/detection schemes.42 Achieving high-quality, artifact-free multi-color super-resolution imaging, especially with more than two colors, remains technically demanding.
  10. Need for Specialized Fluorophores (SMLM, STED): SMLM relies on photoswitchable or photoactivatable fluorophores, while STED requires dyes resistant to depletion laser bleaching.39 The availability and performance (brightness, stability, switching kinetics) of suitable fluorophores for specific targets and wavelengths can be a limiting factor, driving continuous development of new fluorescent probes.40

B. Other Optical/Spectroscopic Techniques

  1. Raman Spectroscopy Signal Weakness: Raman scattering is an inherently weak phenomenon, resulting in low signal intensity, especially from nanoscale volumes or low concentrations. This necessitates long acquisition times or high laser powers (risking sample damage) to achieve adequate SNR.2 Enhancing the Raman signal (e.g., via SERS) is often required but introduces its own complexities.
  2. Surface-Enhanced Raman Scattering (SERS) Substrate Reproducibility: SERS relies on plasmonic nanostructures (substrates) to dramatically enhance the Raman signal, but fabricating SERS substrates with uniform and highly reproducible enhancement factors across large areas remains challenging.32 This variability complicates quantitative SERS measurements and hinders its reliability for routine analysis.
  3. Near-Field Scanning Optical Microscopy (NSOM/SNOM) Tip Challenges: NSOM/SNOM achieves sub-diffraction-limit resolution using a sharp probe (aperture or scattering tip) as a near-field light source/detector.1 However, fabricating robust, high-transmission aperture probes or sharp, efficient scattering tips is difficult.14 Tip wear, low signal throughput, and artifacts from tip-sample distance control limit resolution and reliability.21
  4. NSOM/SNOM Low Signal Intensity and Speed: The optical signal collected in NSOM/SNOM is often very weak, requiring sensitive detectors and limiting scan speed.14 Balancing resolution, signal strength, and acquisition speed remains a key challenge, particularly for techniques like Tip-Enhanced Raman Spectroscopy (TERS) which combines NSOM/SNOM with Raman.32
  5. Dynamic Light Scattering (DLS) Resolution Limitations: DLS measures particle size distribution based on Brownian motion but has poor resolution for polydisperse samples or mixtures, as scattering intensity scales strongly with size (r6), meaning larger particles dominate the signal.3 It also assumes spherical particles and requires dilute suspensions, limiting its applicability for complex or concentrated systems.3
  6. Photon Correlation Spectroscopy (PCS) Limitations: PCS, often used synonymously with DLS, faces similar limitations in resolving complex size distributions and is sensitive to the presence of small numbers of large particles or aggregates.2 Accurate interpretation requires knowledge of solvent viscosity and temperature, and assumes non-interacting particles.

IV. In Situ / Operando Characterization Barriers

Observing nanoscale processes under realistic operating or environmental conditions (operando or in situ) provides crucial insights but faces major challenges in replicating complex environments within analytical instruments, controlling stimuli accurately, and avoiding instrument-induced artifacts.5

  1. Bridging the Pressure Gap (EM): Performing EM under truly realistic pressures (atmospheric or higher) relevant to catalysis or environmental processes is extremely difficult.10 Environmental TEM (ETEM) typically operates at pressures orders of magnitude lower (<20 mbar) than ambient conditions due to vacuum requirements and resolution degradation from gas scattering.45 Closed-cell holders allow higher pressures but suffer from membrane scattering and limited reaction volumes.19
  2. Bridging the Temperature Gap: Achieving and accurately measuring very high (>1500 °C) or very low (< liquid N2) temperatures relevant to some materials processes within restrictive sample holders, while maintaining imaging resolution and stability, is challenging.25 Thermal drift and accurate local temperature measurement remain persistent problems.24
  3. Liquid Flow Control in Microreactors (TEM/SPM): Precisely controlling liquid flow rates, mixing reactants, and preventing leakage or bubble formation within miniaturized liquid cells for TEM or SPM is technically demanding.19 Stagnant flow or poor mixing can lead to non-representative reaction kinetics, while the small volumes limit throughput and statistical relevance.19 Static graphene liquid cells are common but lack flow control.19
  4. Electrochemical Cell Design Limitations (TEM/SPM): Designing robust in situ electrochemical cells for TEM or SPM that incorporate working, counter, and reference electrodes, allow electrolyte flow, minimize unwanted reactions/corrosion, and enable high-resolution imaging is complex.45 Issues like IR drop, reference electrode stability, and beam effects on electrochemistry persist.45
  5. Correlative Operando Measurements: Simultaneously measuring functional properties (e.g., catalytic activity, electrical resistance, gas evolution) while performing high-resolution imaging/spectroscopy operando is challenging.19 Integrating measurement probes, synchronizing data streams, and ensuring the measurement doesn't interfere with imaging (and vice-versa) requires sophisticated instrumentation and experimental design.6
  6. Sample Representativeness (In Situ vs. Bulk): Ensuring that the behavior observed in the highly constrained environment and small sample volume of an in situ experiment (e.g., thin TEM sample, specific AFM scan area) is representative of the bulk material or device behavior is a critical challenge.3 Surface effects, confinement, and altered reaction conditions can lead to deviations from macroscopic reality.5 Validation against bulk measurements is crucial but often difficult.22
  7. Beam Effects in Reactive Environments: The electron beam can interact not only with the sample but also with the surrounding gas or liquid environment (in situ TEM), generating radicals or inducing unwanted reactions that perturb the process under study.6 Disentangling intrinsic material behavior from beam-induced artifacts in reactive environments is a major challenge requiring careful control experiments.45
  8. Multi-Modal In Situ Integration Complexity: Combining multiple in situ characterization techniques (e.g., TEM + Raman, AFM + Optical) to probe different aspects of a process simultaneously provides richer information but presents significant instrumentation challenges.13 Aligning observation volumes, synchronizing stimuli and data acquisition, and avoiding interference between techniques requires highly customized and complex setups.
  9. Time Resolution for Ultrafast Processes: Capturing extremely fast nanoscale dynamics (femtosecond to picosecond timescale), such as charge carrier dynamics or initial stages of phase transitions, requires specialized ultrafast techniques (e.g., ultrafast electron microscopy/diffraction, pump-probe optical methods).24 These techniques often involve compromises in spatial resolution, signal-to-noise, or require complex, expensive laser-integrated instruments.13

V. Computational and Data Analysis Barriers

The increasing volume, velocity, and complexity of data generated by modern nanocharacterization tools create significant bottlenecks in data storage, processing, analysis, and interpretation, often requiring advanced computational approaches like AI/ML.8

  1. Big Data Handling (Volume, Velocity): Modern detectors (especially in EM, e.g., 4D-STEM) generate data at enormous rates (TB/hour), overwhelming local storage and processing capabilities.27 Efficient data transfer, storage infrastructure, and high-throughput processing pipelines are required but often lacking or underdeveloped.27 This "data deluge" creates a major bottleneck between acquisition and analysis.27
  2. Real-Time Data Processing/Feedback: Analyzing large datasets in real-time to provide feedback for optimizing experiments (e.g., adjusting focus, tracking features, stopping acquisition) is computationally demanding and requires tightly integrated hardware/software systems.15 Lack of real-time processing limits experimental efficiency and the ability to perform autonomous or adaptive experiments.30
  3. Computational Cost of Simulations/Modeling: Accurately simulating complex nanocharacterization experiments (e.g., electron scattering, SPM tip-sample interactions, optical near-fields) or modeling structure-property relationships requires significant computational resources (HPC) and sophisticated algorithms.47 The computational cost limits the scale and complexity of systems that can be modeled, hindering direct comparison with experiments.
  4. Lack of Standardized Data Formats and Metadata: Diverse instrument manufacturers and analysis techniques lead to a plethora of proprietary or poorly documented data formats, lacking standardized metadata.27 This hinders data sharing, interoperability between software tools, reproducibility, and the application of large-scale data mining or AI/ML approaches.8 Adherence to FAIR (Findable, Accessible, Interoperable, Reusable) principles is lacking.27
  5. AI/ML Data Scarcity and Quality: Training robust AI/ML models for tasks like image segmentation, feature recognition, or property prediction requires large, high-quality, well-annotated datasets, which are often scarce in specialized nanocharacterization domains.8 Lack of comprehensive, classified, and formatted databases hinders the development and application of AI in nanoscience.8 Data cleaning and version management are also critical challenges.57
  6. AI/ML Model Interpretability ("Black Box" Problem): Many powerful AI/ML models, particularly deep learning networks, act as "black boxes," making it difficult to understand why they make certain predictions or classifications.57 This lack of interpretability hinders trust, debugging, and scientific discovery, especially in fields requiring mechanistic understanding.57 Developing explainable AI (XAI) methods suitable for complex scientific data is crucial.
  7. Automation of Complex Analysis Workflows: Automating multi-step analysis workflows involving diverse data types (images, spectra, diffraction patterns) and complex algorithms remains challenging.27 Integrating different software tools, handling exceptions, and ensuring robustness require significant software engineering effort, limiting fully automated analysis pipelines.49
  8. 4D-STEM Data Analysis Bottlenecks: 4D-STEM generates massive four-dimensional datasets (2D diffraction pattern at each 2D scan position).23 Processing this data (e.g., for orientation mapping, phase contrast imaging, strain analysis) is computationally intensive, requiring specialized algorithms and often significant processing time, creating a bottleneck after acquisition.52 Real-time analysis is particularly challenging.52
  9. Cryo-EM/ET Image Processing Complexity: Reconstructing high-resolution 3D structures from noisy cryo-EM/ET data involves complex image processing pipelines (motion correction, CTF estimation, particle picking/alignment, classification, reconstruction) requiring significant computational power (often GPUs) and user expertise.27 Streamlining and automating these workflows while ensuring accuracy remains an ongoing challenge.27
  10. SPM Data Analysis Automation: Automating the analysis of SPM data, such as identifying features, quantifying properties from force curves, or correcting for artifacts, is needed to handle increasing data volumes and improve reproducibility.15 Applying ML for tasks like tip state recognition or automated feature classification shows promise but requires further development and validation.15

VI. General / Cross-Cutting Barriers

These barriers affect multiple nanocharacterization techniques or represent overarching challenges in the field.

  1. Multi-Modal Data Integration and Correlation: Combining data from different characterization techniques (e.g., AFM + Raman, TEM + APT, Light + Electron Microscopy) provides complementary information but presents significant challenges in spatial registration, data fusion, and correlative analysis.13 Developing streamlined workflows and robust software for multi-modal correlative nanocharacterization is crucial but underdeveloped.
  2. Lack of Standard Reference Materials and Methods: The absence of widely accepted standard reference materials and standardized measurement protocols for many nanoscale properties (e.g., nanoparticle size distribution, film thickness, mechanical properties measured by AFM) hinders comparability of results between labs and techniques.3 Developing traceable standards is critical for reliable nanomanufacturing and regulation.64
  3. Metrology for Nanomanufacturing: Transitioning nanocharacterization techniques from research labs to reliable, high-throughput metrology tools for process control and quality assurance in nanomanufacturing requires significant improvements in automation, robustness, speed, and standardization.63 Bridging the gap between lab capabilities and industrial needs remains a major barrier.56
  4. Training and Workforce Development: Operating advanced nanocharacterization instrumentation and interpreting complex data requires highly specialized skills.31 There is a need for improved training programs and workforce development to ensure sufficient expertise is available to utilize these powerful tools effectively and develop new methodologies.22 Lack of trained personnel can be a significant bottleneck for facilities and research groups.27
  5. Sustainability and Energy Consumption: Advanced nanocharacterization facilities, particularly those involving large instruments like TEMs, UHV systems, and HPC clusters for data analysis, consume significant amounts of energy.8 Developing more energy-efficient instrumentation, data handling strategies (e.g., efficient data compression, optimized algorithms), and sustainable operational practices is becoming an increasingly important consideration for the field.8

Conclusion

The field of nanocharacterization continues to push the frontiers of measurement science, enabling remarkable insights into the nanoscale world. However, as this compilation demonstrates, significant tooling, instrumentation, and methodological barriers persist across all major techniques. Overcoming limitations related to fundamental physics (e.g., beam-sample interactions, diffraction limit), engineering complexity (e.g., in situ cell design, aberration correction, probe fabrication), data handling (e.g., big data management, real-time analysis, AI integration), and practical constraints (e.g., cost, accessibility, standardization) is paramount. Addressing these challenges through continued innovation in instrument design, detector technology, sample preparation methods, computational algorithms, and collaborative development of standards and databases will be crucial for unlocking the full potential of nanotechnology and translating nanoscale discoveries into impactful real-world applications. The integration of automation and artificial intelligence appears particularly vital for tackling issues of throughput, reproducibility, and the overwhelming complexity of modern nanocharacterization data.

Works cited

  1. Optical and Electron Microscopy for Analysis of Nanomaterials - PubMed, accessed April 17, 2025, https://pubmed.ncbi.nlm.nih.gov/33782877/
  2. Nanocharacterization Techniques | Request PDF - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/308277403_Nanocharacterization_Techniques
  3. Characterization techniques for nanoparticles: comparison and ..., accessed April 17, 2025, https://pubs.rsc.org/en/content/articlehtml/2018/nr/c8nr02278j
  4. Time-Resolved Scanning Ion Conductance Microscopy for Three-Dimensional Tracking of Nanoscale Cell Surface Dynamics | ACS Nano - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/abs/10.1021/acsnano.1c05202
  5. Tutorial on In Situ and Operando (Scanning) Transmission Electron Microscopy for Analysis of Nanoscale Structure–Property Rela - NREL, accessed April 17, 2025, https://www.nrel.gov/docs/fy25osti/90453.pdf
  6. Tutorial on In Situ and Operando (Scanning) Transmission Electron ..., accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acsnano.4c09256
  7. Imaging of Hydrated and Living Cells in Transmission Electron Microscope: Summary, Challenges, and Perspectives | ACS Nano, accessed April 17, 2025, https://pubs.acs.org/doi/full/10.1021/acsnano.5c00871
  8. Nano & AI: A Nobel Partnership | ACS Nano - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acsnano.4c14832
  9. Molecular Electron Microscopy: State of the Art and Current Challenges - PMC, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC2660199/
  10. In Situ TEM Studies of Catalysts Using Windowed Gas Cells - MDPI, accessed April 17, 2025, https://www.mdpi.com/2073-4344/10/7/779
  11. Microscopy and Microanalysis (M&M) Meeting Updates, accessed April 17, 2025, https://microscopy.org/files/galleries/MM21_Event-Site-PDF.pdf
  12. Electron Microscopy Studies of Soft Nanomaterials | Chemical Reviews - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acs.chemrev.2c00461
  13. Volume 30 Issue Supplement_1 | Microscopy and Microanalysis - Oxford Academic, accessed April 17, 2025, https://academic.oup.com/mam/issue/30/Supplement_1
  14. (PDF) Introduction: Nanomaterials characterization using microscopy - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/246865964_Introduction_Nanomaterials_characterization_using_microscopy
  15. Scanning probe microscopy in the age of machine learning - AIP Publishing, accessed April 17, 2025, https://pubs.aip.org/aip/aml/article/1/4/041501/2920721/Scanning-probe-microscopy-in-the-age-of-machine
  16. Constraints and Challenges in Tip-Based Nanofabrication | Request PDF - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/302503284_Constraints_and_Challenges_in_Tip-Based_Nanofabrication
  17. Scanning Probe Microscopy – Including Scanning Tunneling Microscopy and Atomic Force Microscopy – Principles and Applications | Technology Networks, accessed April 17, 2025, https://www.technologynetworks.com/analysis/articles/scanning-probe-microscopy-including-scanning-tunneling-microscopy-and-atomic-force-microscopy-356991
  18. Microscopy & Microanalysis 2019 Meeting, accessed April 17, 2025, https://microscopy.org/files/galleries/MM19_Event-Site-PDF.pdf
  19. Full article: In situ and operando transmission electron microscopy for the synthesis, fundamental understanding and catalytic applications of nanoalloys - Taylor & Francis Online, accessed April 17, 2025, https://www.tandfonline.com/doi/full/10.1080/23746149.2025.2481277?af=R
  20. Liquid cell transmission electron microscopy and its applications | Royal Society Open Science - Journals, accessed April 17, 2025, https://royalsocietypublishing.org/doi/10.1098/rsos.191204
  21. Atomic Force Microscopy Application in Biological Research: A Review Study - PMC, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC3700051/
  22. Tutorial on In Situ and Operando (Scanning) Transmission Electron Microscopy for Analysis of Nanoscale Structure–Property Relationships (Journal Article) - OSTI, accessed April 17, 2025, https://www.osti.gov/pages/biblio/2482574
  23. 4D-STEM of Beam-Sensitive Materials | Accounts of Chemical Research - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acs.accounts.1c00073
  24. Perspective and prospects of in situ transmission/scanning transmission electron microscopy - Oxford Academic, accessed April 17, 2025, https://academic.oup.com/jmicro/article/73/2/79/7450959
  25. Current status and future directions for in situ transmission electron microscopy - PMC, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC5100813/
  26. Microscopy Today: Volume 30 - Issue 1 | Cambridge Core, accessed April 17, 2025, https://www.cambridge.org/core/journals/microscopy-today/issue/4ED15EE8743961FD9E376D7AD2752356
  27. Big data in contemporary electron microscopy: challenges and ..., accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC10492738/
  28. Design and Applications of Environmental Cell Transmission Electron Microscope for In Situ Observations of Gas–Solid Reactions - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/10890199_Design_and_Applications_of_Environmental_Cell_Transmission_Electron_Microscope_for_In_Situ_Observations_of_Gas-Solid_Reactions
  29. Marketing | EDAX Blog, accessed April 17, 2025, https://edaxblog.com/category/marketing/
  30. (PDF) Scanning probe microscopy in the age of machine learning, accessed April 17, 2025, https://www.researchgate.net/publication/375548541_Scanning_probe_microscopy_in_the_age_of_machine_learning
  31. Towards a Fully Automated Scanning Probe Microscope for ..., accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8123492/
  32. Comparing Commercial Metal-Coated AFM Tips and Home-Made Bulk Gold Tips for Tip-Enhanced Raman Spectroscopy of Polymer Functionalized Multiwalled Carbon Nanotubes, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8840094/
  33. Improving the Consistency of Nanoscale Etching for Atomic Force Microscopy Tomography Applications - Frontiers, accessed April 17, 2025, https://www.frontiersin.org/journals/materials/articles/10.3389/fmats.2019.00203/full
  34. Solid Platinum Nanoprobes for Highly Reliable Conductive Atomic Force Microscopy | ACS Applied Materials & Interfaces - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acsami.3c01102
  35. Scanning Probe Lithography: State-of-the-Art and Future Perspectives - PMC, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC8878409/
  36. Studies of probe tip materials by atomic force microscopy: a review, accessed April 17, 2025, https://www.beilstein-journals.org/bjnano/articles/13/104
  37. Highlights from Microscopy and Microanalysis - Oxford Academic, accessed April 17, 2025, https://academic.oup.com/mt/article/31/2/40/7116011
  38. The most comprehensive source for all elds of microscopy, histology and general laboratory research... - gets an upgrade. - Wiley Analytical Science, accessed April 17, 2025, https://analyticalscience.wiley.com/cms/asset/e91f7274-14d8-4093-bb90-fcec5ed1bffb/emea63digital.pdf
  39. Challenges facing quantitative large-scale optical super-resolution ..., accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC7898072/
  40. Challenges and Opportunities in Superresolution | Features - Photonics Spectra, accessed April 17, 2025, https://www.photonics.com/Articles/Challenges_and_Opportunities_in_Superresolution/a61868
  41. Beyond the Limit: The World of Super-Resolution Microscopy | Labcompare.com, accessed April 17, 2025, https://www.labcompare.com/10-Featured-Articles/342799-Beyond-the-Limit-The-World-of-Super-Resolution-Microscopy/
  42. Super-resolution Microscopy for Nanomedicine Research | ACS Nano, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acsnano.9b05289
  43. Super-resolution Microscopy for Nanomedicine Research - PMC, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC6764015/
  44. Review of Recent Development of In Situ/Operando ... - PubMed, accessed April 17, 2025, https://pubmed.ncbi.nlm.nih.gov/31099081/
  45. Importance and Challenges of Electrochemical in Situ Liquid Cell Electron Microscopy for Energy Conversion Research - ACS Publications, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acs.accounts.6b00330
  46. Mechanistic Insights and Technical Challenges in Sulfur-Based Batteries: A Comprehensive In Situ/Operando Monitoring Toolbox, accessed April 17, 2025, https://pmc.ncbi.nlm.nih.gov/articles/PMC11650778/
  47. Mechanistic Insights and Technical Challenges in Sulfur-Based Batteries: A Comprehensive In Situ/Operando Monitoring Toolbox | ACS Energy Letters, accessed April 17, 2025, https://pubs.acs.org/doi/10.1021/acsenergylett.4c02703
  48. Frontiers of in-situ materials characterization - from new ..., accessed April 17, 2025, https://www.european-mrs.com/frontiers-situ-materials-characterization-new-instrumentation-and-methods-imaging-aided-materials
  49. Machine learning in electron microscopy for advanced ..., accessed April 17, 2025, https://pubs.rsc.org/en/content/articlelanding/2022/nh/d2nh00377e
  50. Lab on a beam—Big data and artificial intelligence in scanning transmission electron microscopy | Request PDF - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/334440776_Lab_on_a_beam-Big_data_and_artificial_intelligence_in_scanning_transmission_electron_microscopy
  51. Streaming Large-Scale Microscopy Data to a Supercomputing Facility - arXiv, accessed April 17, 2025, https://arxiv.org/html/2407.03215v1
  52. Real-Time Interactive 4D-STEM Phase-Contrast Imaging From Electron Event Representation Data: Less computation with the right representation | Request PDF - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/357501415_Real-Time_Interactive_4D-STEM_Phase-Contrast_Imaging_From_Electron_Event_Representation_Data_Less_computation_with_the_right_representation
  53. Accelerating Time-to-Science by Streaming Detector Data Directly into Perlmutter Compute Nodes - arXiv, accessed April 17, 2025, https://arxiv.org/html/2403.14352v1
  54. Real-time 4D signal processing and visualization using graphics processing unit on a regular nonlinear-k Fourier-domain OCT system - Optica Publishing Group, accessed April 17, 2025, https://opg.optica.org/oe/abstract.cfm?uri=oe-18-11-11772
  55. accessed December 31, 1969, https://pubs.aip.org/aip/aml/article/1/4/041501/2920721
  56. nsf-gov-resources.nsf.gov, accessed April 17, 2025, https://nsf-gov-resources.nsf.gov/files/JSR-24-02%20NSF-TIP%20Advanced%20Materials%20Opportunities_11142024_FINAL_0.pdf
  57. Bridging Nanomanufacturing and Artificial Intelligence—A ... - MDPI, accessed April 17, 2025, https://www.mdpi.com/1996-1944/17/7/1621
  58. [2201.00966] AI visualization in Nanoscale Microscopy - arXiv, accessed April 17, 2025, https://arxiv.org/abs/2201.00966
  59. (PDF) Machine learning in electron microscopy for advanced nanocharacterization: current developments, available tools and future outlook - ResearchGate, accessed April 17, 2025, https://www.researchgate.net/publication/364367337_Machine_learning_in_electron_microscopy_for_advanced_nanocharacterization_current_developments_available_tools_and_future_outlook
  60. Automated crystal system identification from electron diffraction patterns using multiview opinion fusion machine learning | PNAS, accessed April 17, 2025, https://www.pnas.org/doi/10.1073/pnas.2309240120
  61. Analysis of Interpretable Data Representations for 4D-STEM Using Unsupervised Learning, accessed April 17, 2025, https://www.osti.gov/pages/biblio/1914356
  62. 4D-Explorer: A visual software for 4D-STEM data processing and image reconstruction, accessed April 17, 2025, https://www.researchgate.net/publication/371605967_4D-Explorer_A_visual_software_for_4D-STEM_data_processing_and_image_reconstruction
  63. Instrumentation and Metrology for Nanotechnology, accessed April 17, 2025, https://www.nano.gov/sites/default/files/pub_resource/nni_instrumentation_metrology_rpt.pdf
  64. www.nist.gov, accessed April 17, 2025, https://www.nist.gov/system/files/documents/mml/nano_small_web-4.pdf