Symposium F.MT02—Multimodal, Functional and Smart Scanning Probe Microscopies for Characterization and Fabrication

Over the years, luminescence spectroscopy has established itself as one of the fundamental methods for analyzing the photophysical properties of a variety of samples, ranging from simple organic molecules, complexes, luminescent proteins to semiconductors. The commonly used steady-state methods (i.e. excitation and emission spectroscopy) provide valuable insights into the photophysics of samples. However, such results give only a partial view of sample behaviors after photoexcitation. A further piece of the puzzle is often revealed by performing time-resolved luminescence spectroscopy.
Studying luminescence lifetime data of materials is a very powerful analytical tool for spectroscopists and microscopists alike, as it provides insights into excited state dynamics of molecules, complexes, nanoparticles, or semiconductors. The fluorescence or phosphorescence lifetime is an intrinsic characteristic of luminescent species. It indicates how long species under consideration will remain in electronically excited states before returning to ground state. Each emitting species has a characteristic luminescence lifetime, that can be influenced by its environment.
A series of spectroscopic and microscopic methods based on luminescence lifetime have been developed and provide further information, which are not accessible by steady-state experiments. For example, with fluorescence lifetime imaging microscopy (FLIM), a very well established imaging method in life science, it is possible to gather lifetime information and spatial information as combination of a sample. This combined data can help to understand biochemical or physical processes by detecting changes in local environment such as pH, temperature, or ion concentration, and to identify molecular interactions or conformation changes via Förster Resonance Energy Transfer (FRET).
Time-resolved photoluminescence measurements are powerful tools in materials science also, e.g. for characterization of key parameters like charge carrier dynamics and mobility in semiconductors. Furthermore, combining spectral, spatial and lifetime information often provide a much better understanding of different dynamic processes occurring in luminescent samples.
Here we will demonstrate the performance of a spectrometer-microscope assembly for characterization and analysis of different materials in terms of lifetime, spectral and spatial resolution, as well as ability to measure long luminescence decays (in µs and ms time scales) using laser drivers with burst capabilities. The abilities of this multi-dimensional approach will be demonstrated with a series of examples reflecting a broad range of applications in materials science research.

There is growing interest in thin-film transparent heaters (TFTH) for optoelectronics, and a plethora of domestic usages such as smart windows, defrosters in automobiles and other applications. In this work, we present the growth and characterization of TFTH made of gallium doped zinc oxide (GZO). GZO thin films are deposited by pulsed laser deposition (PLD) with a varying range of oxygen (0-10 mTorr) on glass. Ga doped (5 atomic weight percentage) ZnO target was made in the laboratory using Ga₂O₃and ZnO powder. GZO is deposited on glass. Atomic force microscopy images reveal that the films are very smooth with excellent surface roughness. The temperature-dependent resistivity was measured using a linear four-probe technique and the sample showed low sheet resistance values. Our TFTH’s exhibited high optical transparency value if (>85%) from transmission data. GZO transparency heaters showed a stable and reproducible Joule heating effect and the temperature reached easily 60°C by the application of low input (~8 V) voltage. This research funding may be beneficial for the potential use of GZO as transparent conducting oxide materials for possible applications in the emerging areas of low-cost power electronics, and flexible and wearable optoelectronic devices.

Energy harvester converts wasted energy into usable electrical energy. Recently, with the rise of Internet of Things (IoT) technology, various types of sensors have been embedded into systems such as mobile and wearable devices, and unmanned equipment. Since it is difficult to supply all the energy required by many sensors with wired power systems, many studies have been conducted to overcome the issue of wired power through energy harvesting technology. Piezoelectric energy harvesters can generate the electrical energy to convert directly the applied strain energy from mechanical energy such as vibration. However, the major challenges of piezoelectric energy harvesters are the relatively low power yield and the degradation of the energy harvesting device due to fatigue caused by tens of millions of vibrations. Here we conducted the accelerated stressed cycle test in the low frequency range of 30 Hz band for a commercially available piezoelectric energy harvester, which is composed of a PZT-5H piezoelectric layer and a flexible substrate. Subsequently, we visualized the piezoelectric fatigue in microscopic region, and evaluated the degradation characteristics using piezoresponse force microscopy (PFM) to correlate the macroscopic fatigue with micron scale ferroelectric domain behavior.

Conjugated polymers have been extensively studied due to their tunable optical and electronic properties relevant to thin-film optoelectronic devices. Several techniques have been used to modify the electrical properties of these polymers, including solvent processing methods to enhance morphology, and molecular doping to enhance conductivity. Addition of a solvent in which the polymer has poor solubility (antisolvent) to the polymer solution prior to spin coating has been shown to improve the morphology of polymer thin films by promoting increased aggregation during solvent evaporation while spin coating. Here, we study the local physical and nanomechanical properties of several common conjugated polymer thin films prepared using an antisolvent dripping method in which the antisolvent is added during, rather than before, the spin coating process. This approach is expected to aid crystallization and produce uniform, stable, films.

Using atomic force microscopy (AFM), we establish a direct relation between the physical, mechanical, and electrical properties of conjugated polymer thin films prepared through antisolvent dripping and electrochemical doping. Phase imaging and fast force mapping (FFM) are used to visualize the nanoscale crystalline and amorphous domains. Additionally, in the case of doped conjugated polymer films, FFM and conductive AFM measurements (CAFM) have been combined for simultaneous measurement of topography, Young’s modulus, and current. The combined FFM and CAFM technique splits the biased sample into a grid of points and records the deflection of the cantilever as a function of tip sample separation at each point, as well as the current measured through the sample by the tip, allowing for a direct comparison of mechanical and electrical properties at the nanoscale. The solvent used for the polymer precursor solution and antisolvent used during spin coating were chosen based on Hansen solubility parameters (HSP). Using HSPs a list of good and poor solvents for the polymers were determined. Choosing a high-boiling-point, good solvent in the precursor solution ensures this solvent remains long enough during spin coating to allow for mixing with the poor solvent added during antisolvent dripping. This provides sufficient time for rearrangement of molecular chains to form a more ordered structure. In addition, for electrochemically doped conjugated polymer films, we find that higher dopant concentrations, and increased conductivity, occur in nanoscale crystalline domains compared to amorphous domains. Raman mapping and energy dispersive spectroscopy support the AFM findings at the micron scale. These results can be used to tune the morphology and resultant electrical properties of conjugated polymer films for use in electronic and optoelectronic devices.

We present a new modality of force-curve-based AFM that eliminates systematic bias in the calculation of elastic modulus from cantilever spring constant calibration errors in force curve analysis. We show that the ratio between the probe deflection and substrate indentation has important implications for the uncertainty of modulus measurement when using generalized adhesive Hertzian indentation models. This error analysis codifies the widely understood rule-of-thumb tradeoff that must be made: if the cantilever deflection is much greater than the material indentation, bias due to error in cantilever calibration will be amplified, whereas if the cantilever deflection is much smaller than material indentation, photodetector noise will dominate. For example, attempts to measure the modulus of a stiff material with a very compliant probe result in a high indentation ratio and a large systematic error due to deflection sensitivity calibration. Our work reveals that, under the common calibration methods which involve the equipartition theorem, a special, “magic” indentation ratio appears between these limits with the ability to annihilate spring constant calibration error. In this presentation, the theory underlying the existence and error-nullifying properties of magic ratio indentation will be shown, followed by demonstrations on real data through custom analysis code and black-box commercial application routines. Caveats and limitations of magic ratio AFM will be discussed, as well as recommendations for users and manufacturers of AFM instruments. Additionally, opportunities for modulus measurement error mitigation with multi-modal and multi-functional methods will be surveyed.

Nanoscale infrared spectroscopy and imaging techniques - AFM-IR and s-SNOM offer non-invasive material characterizations at the sub-diffraction limit. Recent development in AFM-IR and s-SNOM have significantly advanced the spatial resolution and detection sensitivity [1, 2]. However, the spectral range of the techniques are limited to 800-1900 cm^-1 and 2700-3600 cm^-1 which typically involves multiple laser sources. The demand for a versatile laser source that can provide wider tuning range and accommodates both AFM-IR and s-SNOM, therefore, motivates design and implementation of new light source and detection techniques.

In this presentation, we demonstrate a novel method of extending both AFM-IR and sSNOM to the full mid-IR range by coupling with a broadband laser source. Based on OPO/DFG tunable IR generation and seeded by a powerful fs pump laser, the novel broadband laser source delivers an ultra-broadband (6 octave) nano spectroscopy. For AFM-IR, a new modulation scheme is developed to modulate the broadband light at a repetition rate that matches one of the contact resonance frequencies of the AFM cantilever for resonance enhancement. The resulting spectrum is obtained by integration with an FTIR unit achieving 670 – 4000 cm^-1 with a typical 4 cm^-1 spectral resolution in a single laser shot. For s-SNOM measurements, the broadband laser operates in the quasi-cw mode, enabling both s-SNOM spectroscopy and imaging [3]. This versatile system sets a new standard in terms of flexibility and tuning range the for the two complementary nanoscale IR techniques.
In this context, we also highlight the laser performance and demonstrate its application on several representative samples which achieved promising new insights in the understanding of a multitude of different materials.

[1] A Dazzi and C B Prater, Chem. Rev., 117 (2016), p. 5146-5173
[2] E Muller et al., J. Phys. Chem. Lett., 6 (2015), p. 1275-1284
[3] J Yang et al., ACS Nano, 14 (2020), p. 1123−1132

One of the key advantages many SPM techniques have for cell biology studies is their ability to non-destructively image cells in physiological environments. Using time lapse microscopy, it is then possible to observe how the cells change over time either during growth, development or due to drug treatments. For bacterial cells, time lapse atomic force microscopy (AFM) has been very helpful in understanding how cells grow, divide and react to antibiotic stresses. For soft mammalian cells however, long term time lapse using AFM is more difficult due to the complex and time varying force interaction between the tip and the cell surface. For these cells, scanning ion conductance microscopy (SICM) would be better suited for time lapse experiments. Unfortunately, standard SICM instruments require 10s of minutes for good quality images, which is often too slow to follow important cell dynamics. In this talk I will describe a microscope we have developed for time resolved SICM imaging of mammalian cells that enables imaging durations of up to 48h with image acquisition times as low as 3 seconds. We have combined this SICM instrument with super-resolution optical fluctuation imaging (SOFI) to complement the morphological SICM data with biochemical fluorescence information. With this instrument we study cell growth, motility, macropinoytosis and cell differentiation.

Here, we present a multimodal scanning probe microscopy study of a composite anode with a dispersed lithium silicon titanium phosphate (LSTP) lithium ion conductor for all-solid-state batteries. We used electro-chemical strain microscopy (ESM) and lateral force microscopy (LFM) to analyze the electromechanical response and friction force dependence as a function of the measurement parameters such as the tip loading force and AC drive voltage. We found that the sensitivities of friction force and ESM amplitude could be valid markers to identify each component in the composite anode. In addition, we visualized the distribution of active ionic sites of lithium ions in the region of LSTP and binder materials based on Pearson’s correlation analysis between nanoscale ESM and LFM results. With the suggested technique, various components of composite electrodes can be directly visualized and distinguished in ambient conditions, with their properties being measured simultaneously. These methods will provide insights into the optimal conditions of composite electrodes and allow for developing next-generation all-solid-state batteries.

Interest in graphene has grown considerably in the last decade due to its unique optical and electrical properties. However, some potential applications of graphene, such as transistors, require a bandgap, which is challenging to generate in graphene. One strategy is to tune local electronic properties by chemical modification through oxidation and surface functionalization. Using transition metal complexes, a desirable energy gap can be opened up in graphene.
Understanding and thereby controlling the mechanism of functionalization is yet hampered by the high degree of structural and chemical disorder in graphene and graphene derivatives. Pure graphene contains inhomogeneous morphological defects such as grain boundaries, edges, and wrinkles. Engineering the electronic properties by surface modification can often lead to a larger degree of inhomogeneity in chemical composition, molecular structures, and the degree of sp² conjugation.
Since the number of molecules needed for chemical modification is low, macroscopic methods such as energy-dispersive X-ray spectroscopy or X-ray photoelectron spectroscopy lack the spatial and chemical sensitivity necessary to detect modifications in graphene.
In this work, we employ functional scanning probe microscopy (SPM) techniques that offer spatially resolved, chemically specific and quantitative information on chemically modified graphene derivatives.
Tip-enhanced Raman spectroscopy (TERS) is a powerful surface analysis technique that offers nanometer-resolved images with site-specific chemical fingerprints. We demonstrate nanoscale TERS mapping of structural and functional groups on graphene derivatives. We employ machine learning (ML) tools to extract information on chemical components and their spatial distribution from the hyperspectral TERS maps.
We validate the components resulted from ML-aided analysis by comparing the principal components with density functional theory (DFT) based calculation. The prediction of vibrational modes from DFT offers insights into understanding the site-specific derivative molecules and the bonding between molecules and graphene.
We further extend surface characterization by combining TERS with Kelvin probe force microscopy (KPFM) to locally examine the electrical properties. In conjunction with illumination, we employ KPFM to estimate the local conduction and valence band edge energies. This in situ measurement allows a direct correlation of local topography, chemical composition, and electronic properties.
These methods can be applied to correlated SPM data to obtain complex compositional information that is challenging to gain from any conventional method. By combining nanometer-resolved TERS and KPFM imaging, with computation-aided ML and analysis, we have offered a comprehensive view of chemical and spatial information and demonstrated a generally applicable approach for improving the accuracy of surface characterization. We expect this work to provide a more profound understanding of the molecular binding structures with guidance for optimizing graphene semiconductors and devices.

Near-field scanning optical microscopy (NSOM) has become one of the most crucial and powerful techniques to characterize the chemical, physical, and biochemical properties of materials with nanometer-scale resolution in real-time. A key element for any NSOM systems that combine optical spectroscopy with scanning probe microscopy is the actual probe itself. An ideal NSOM probe provides a strong local electromagnetic field enhancement, efficient far-field to near- field coupling, nanoscale spatial resolution, background-free operation, and broadband photon-plasmon coupling to enable high spatial and temporal resolution. One of the most exciting optical probe architectures based on an optical transformer, “campanile,” was recently developed and has been successfully used for multidimensional spectroscopic imaging of nanostructures with nanoscale resolution, providing insights into novel optoelectronic process¹. However, making such probes requires both the expertise and the facility, and reproducibility is a significant issue. The economic production of reasonable volumes of such probes is not possible. There is a critical need for the development of optical tip technology to yield high-performance and reliable near-field probes and to become available to the broader user community. The realization of such probes would impact nanoscale research at the same level that atomic force microscopes did when paving the way to modern nanoscale science.
We present a wafer-scale realization of Campanile near-field scanning optical probes fabricated on an AFM platform. Figure 1 illustrates a 2D sketch and a scanning electron micrograph (SEM) image of an optical transformer on a cantilever. The figure shows the Boron Nitride flake sample and its topographical image taken by the Campanile AFM probe. Optical measurement results show a promising sign of probes’ functionality with a strong polarization dependence response, which is a tell-a-tale sing of a campanile. This work paves the way for low cost and volume manufacturing of near-field probes suitable for high-resolution hyperspectral imaging.

Accurate monitoring and control of surface chemical reaction by adsorbates on metal oxide semiconductors are of key interest in many electrochemical systems. These nanoscale surface chemistries are often probed by X-ray photon spectroscopy (XPS), scanning tunneling microscopy (STM), and tunneling electron microscopy (TEM) at high vacuum. However, these methods are not suitable for in-operando experiment under various gas environments and the use of energetic particles can potentially disturb the surface chemistry. In contrast, optical second harmonic generation (SHG) is a simple, non-invasive, vacuum-free, and sensitive probing method for surface chemistry. The intrinsic broken inversion symmetry of surface can induce SHG signals, which can be altered by changes in charge distribution at the surface, depending on the presence of chemical adsorbates. Integrated with tip-based scanning near-field optical microscopy (SNOM), optical SHG can interrogate various nanoscale surface phenomena and their dependence on the gas environment without complicated sample preparation.

Here, we demonstrate in-situ tip-based near-field SHG microscopy of chemical reactions at the surface of metal oxide nanowires, including titanium oxide (TiO₂), zinc oxide (ZnO), and vanadium oxide (VO₂). The nanoscale SHG can be amplified by local electric field enhancement by the tip-apex and provide surface chemical state with nanoscale resolution. Based on this method, we in-situ probed the reactions of oxygen and water vapor at the surface of TiO₂ and ZnO, depending on the local vacancy density and the occurrence of catalytic reaction at elevated temperature. In addition, strain effect on the local phase transition in VO₂ nanowires was probed on a mechanical clamped structure at controlled temperature. The correlated TEM inspection of these nanowires showed surface defects and phase state consonant with the near-field SHG imaging. We believe that the presented in-situ near-field SHG microscopy can provide powerful means for in-operando probe of various physicochemical phenomena.

For the last few decades the rapid growth in the field of nanoscience and technology has led to the development of new characterization tools for nanoscale materials. Traditional IR and Raman spectroscopy and imaging offers excellent chemical insights; however, the spatial resolution is limited by the optical diffraction limit (~l/2). Although, recent Super-resolution microscopy techniques [1, 2] offer superior spatial resolution, they are primarily implemented in fluorescence imaging, hence needs external fluorophore tag for detection. Alternatively, high resolution multimodal imaging platforms offer unique combination of SPM and chemical spectroscopy/spectrometry, such as, IR, Raman and SIMS to probe nanoscale chemistry without external fluorophore. One of the above techniques, namely, nanoscale IR spectroscopy/imaging offers a “tag free” spectral detection by exploiting an AFM probe [3] to detect either photothermal expansion force (PTIR/AFM-IR) or near field scattered IR light (s-SNOM) with high spatial resolution beyond optical diffraction limit (2-5 mm).

Recent developments in PTIR/AFM-IR and s-SNOM technology have significantly augmented the speed and spatial resolution for chemical analysis. Exploiting different modalities in AFM, such as, contact, tapping and other heterodyne force detection modes (e.g., ultrasonic and/or acoustic force microscopy), we are able to acquire IR images at a specific absorption band simultaneously with sample topography and viscoelastic properties, providing a complete set of topographical, chemical and mechanical insights with ~10 nm spatial resolution. These high-resolution measurements are currently accompanied by high speed tunable lasers with broad spectral coverage ranging from 800-3600 cm^-1, encompassing full capability of FTIR characterization for dimensions orders of magnitudes smaller than the optical diffraction limit. For example, Figure 1c shows topography and chemical map of PS-b-LDPE block copolymer on a silicon substrate. IR images at 1466/2920 cm^-1 and 1488/3032 cm^-1 highlight the distribution of low-density polyethylene (LDPE) and polystyrene (PS), respectively with feature sizes ~50 nm or smaller.
Along with high spatial resolution, high speed spectral acquisition enables acquiring hyperspectral data cube with large set of data. These high-volume data set opens up new avenues for machine learning algorithm for data processing and analysis, e.g., high resolution spectral map reconstruction using pan-sharpening algorithm [4].

In this presentation, we will highlight the technical background and applications of these emerging technologies in different fields, e.g., nanomaterials, life sciences, polymers, microelectronics etc.
[1] B Huang et al., Annu. Rev. Biochem., 78 (2009), p. 993-1016
[2] T D Harris et al, Appl. Spectrosc., 48 (1994), p. 14A-21A
[3] A Dazzi and C B Prater, Chem. Rev., 117 (2016), p. 5146-5173
[4] N Borodinov et al., npj Comput. Mater., 49 (2019), p. 1-9

Charged ferroelectric domain walls are naturally occurring quasi-2D objects with a range of functional properties and potential technological applications. The hexagonal manganites (RMnO₃, R= Sc, Y, In, Dy to Lu) are an important model system for the study of domain walls, as these improper ferroelectrics naturally develop all orientations of 180° domain walls, including neutral and charged configurations. The charge state and properties are determined by the orientation and geometry of the domain wall, giving rise to rich electronic behavior. At the surface, the "projected" domain wall properties can readily be characterized, for example using scanning probe microscopy (SPM) and photoemission electron microscopy (PEEM), whereas mapping of their characteristics in three dimensions remains challenging.

In this work, I will discuss the use of focused ion beam (FIB) techniques for extending the surface information from SPM studies into 3D. As a first example case, I will discuss the use of FIB to prepare ferroelectric lamella samples with well-defined domain wall configurations determined via correlated microscopy techniques. In order to access the complete 3D information, surface SPM is combined with FIB tomography to image ferroelectric domain walls with nanoscale resolution and correlate the local sub-surface geometry with the measured properties at the surface. Understanding how the 3D geometry of the domain walls influences their local electronic response is an important step toward unlocking their full intrinsic functionality and application in nanoelectronics devices.

Modern materials research in combination with innovative fabrication and processing techniques enable development of increasingly complex materials and advanced components. Analyzing these new materials and understanding their microstructure requires sophisticated characterization tools. Focussed ion beam scanning electron microscopes (FIB-SEMs) are well established for sample analysis and preparation at the nano- to micro-scale in 2D, 3D or even 4D. However, when it came to sample modification and analysis at even larger scales up to millimeters FIB-SEMs were facing a challenge.

With the LaserFIB we present and introduce a new system to solve this challenge and bridge this gap. The LaserFIB features a fs laser that has been integrated with a ZEISS Crossbeam. This combination enables high-resolution SEM imaging and FIB patterning as well as site-specific, large-scale sample structuring and preparation using the laser. All laser processing is done in a separate chamber extending the airlock to avoid damage and contamination of the components in the instrument’s main chamber (Fig. 1). The massive material ablation by the laser is highly efficient in large volume bulk material removal and offers a way to gain access to deeply buried regions of interest (ROIs). Large cross-sections can be prepared in seconds and subsequently get analyzed by SEM leveraging all sorts of analytical techniques (e.g. EDS or EBSD). If needed the FIB allows to fine-polish the cross-section face for undisturbed imaging, however, it has been shown that details of the microstructure are already visible after laser polishing only and even suitable for e.g. EBSD analysis.

In this contribution we will introduce the new LaserFIB and explain the site-specific workflow in detail. Application examples covering different sample materials and applications will be shown. Among these are e.g. correlative sample preparation for X-ray nano-tomography, micro-mechanical testing and large-scale cross-sectioning.

Antiferromagnetic thin films attract significant interest for future low-power spintronic devices. They are largely insensitive to external magnetic fields, exhibit domains with negligible cross-talk and can be switched at THz frequencies [1]. Furthermore, in multiferroics where antiferromagnetism can coexist with ferroelectricity, it is possible to control the magnetic state electrically - a particularly appealing feature for future device applications.
In this talk, we show how the Qnami ProteusQ^TM -- a commercial room temperature scanning nitrogen-vacancy magnetometer (SNVM) -- can be used to quantitatively image antiferromagnetic spin textures with proven DC field sensitivity of ∼ 2 μT/√Hz. We image the spin cycloid of a multiferroic bismuth ferrite (BiFeO₃) thin film and extract a period of about 103 nm, indicative of a type II spin cycloid. [2] In addition, we take advantage of the quantitative nature of SNVM to extract an average magnetic moment density of 0.08 ± 0.02 μ_B /Fe in the BiFeO₃ film. These findings agree well with reported vibrating sample magnetometer results (M_S = 0.06 μ_B /Fe) on similar, strained BiFeO₃ thin films. [3] Our results shed light on future design requirements for reconfigurable nanoscale spin textures in such multiferroic systems.

References:
[1] V. Baltz, et al., Rev. Mod. Phys. 90, 015005 (2018)
[2] A. Haykal, et al., Nat. Commun. 11, 1704 (2020)
[3] W Eerenstein, et al., Science 307, 1203a (2005)

The electrodes of a polymer electrolyte membrane fuel cell (PEMFC) primarily contain a Pt/C catalyst and Nafion binder. Because these components play crucial roles in the redox reaction and proton transport, respectively, their distributions can directly affect the electrochemical reactivity, and thus, the device performance. Even though analyzing the component distribution is important to understand its electrochemical reactivity and improve the device performance, it is challenging to determine it for the PEMFC electrode. In this presentation, I will summarize our recent effort for visualizing the spatial distribution of the electrode components and their heterogeneous electrical properties using multidimensional current–voltage (I–V) spectroscopy combined with data mining. The electrical properties of the electrode components, i.e., the Pt/C catalyst and Nafion binder, were explored by I–V spectroscopy, and their electrical heterogeneity was spatially classified based on the shapes of the measured I–V curves by cluster analysis. These results imply that it is possible to get more insight on the data analysis in AFM even using a simple machine learning technique. Furthermore, the proposed method is expected to be applicable for investigating in detail not only the spatial properties of PEMFC electrodes, but also the properties of various material systems.

Machine learning (ML) is permeating many areas of research including drug design, modeling, and materials design. In the realm of characterization, ML is only beginning to be integrated into experimentation and processing to improve the workflow. Here, the potential of ML to play a role in analysis of atomic force microscopy images is explored in two distinct applications.
The first area involves a classic application in ML of image recognition. A convolutional neural net was trained on AFM phase images of different polymer blends and achieved 100% accuracy in differentiating test data, compared with 94% accuracy with a less powerful traditional neural network. This study takes advantage of classic data augmentation methods to increase the sample size since the quantity of AFM images is limited by long acquisition times. The second application involves using ML to conduct particle analysis in challenging AFM images where the background is varied and many of the particles are linked in a matrix phase. This study serves as an example of the ultimate customizable image analysis where the training of the model takes place on the specific class of challenging images. The hope is that this study serves as a building block for future, more powerful applications of the powerful ML tool in more challenging problems that will also be discussed including establishing structure-property relationships and autonomous operation of AFM.

Bio-macromolecules, including peptides, proteins, DNA and peptoids, self-assemble into bio-hybrid materials with two dimensional (2D) hierarchical architectures at solid-liquid interfaces, exhibiting relative epitaxial matching. While the structural details have been elucidated experimentally and rationalized through molecular simulations, in-situ characterizations of dynamics defining orientational selection of bio-macromolecular arrays at single molecular level with temporal resolution in second, are still limited.
In this presentation, we will report the most recent study of angular dynamics of individual protein nanorods¹ on (001) plane of muscovite mica by high-speed atomic force microscopy (HS-AFM) and machine learning (ML) based analysis. We quantitatively analyzed their free energy landscapes and elucidated the transition possibilities between various orientational states for a range of nanorod lengths and electrolytes. After comparing ML based analysis with standard physical models for nanorod dynamics in 2D, we found that the transitions between minima go through bimodal mechanisms. Unlink the in-plane Brownian motion between adjacent minima in the energetic barrier heights, the motion between more distant minima is analogous to Levy-flights, indicating the existence of a second high-energy angular motion for the proteins. This work provides the workflow for in-situ understanding biomacromolecular dynamics at solid-liquid interfaces and the insights for further improving the general approach of managing bio-macromolecular hierarchical architectures with high accuracy and efficiency.

1 Pyles, H., Zhang, S., De Yoreo, J. J. & Baker, D. Controlling protein assembly on inorganic crystals through designed protein interfaces. Nature 571, 251-256, doi:10.1038/s41586-019-1361-6 (2019).

The dynamic shapes of biologial tissues emerge from a complex interplay of physics, chemistry and genetics, which determines--at each temporal and spatial scale--the mechanical properties that eventually form the structures of living organisms. Shape and mechanical stability of living organisms rely on precise control in time and space of growth, which is achieved by dynamically tuning the mechanical (viscous and elastic) properties of their hierarchically built structures from the nanometer up.
It is now well-established that cellular behaviour (including stem cell differentiation) crucially depends on the mechanical properties of the cells’ environment. Much attention has been directed towards the importance of the stiffness of the natural (extracellular matrix, ECM) or artificial matrices where cells grow, with the purpose of either understanding mechanotransduction, or controlling the behaviour of cells in tissue engineering. While stiffness (i.e. the capacity of a material to elastically store mechanical energy) has been the focus of most experimental research, neither cells nor matrices are elastic. Biological systems dissipate energy (i.e. they are viscous) and hence they do not respond to mechanical deformations instantaneously (like an ideal Hookean spring), but present different time responses at different spatial scales that characterise their responses to external stimuli.
Measuring viscoelasticity (especially at the nanoscale) has remained experimentally challenging [1,2]. In my talk I will present atomic force microscopy (AFM)-based techniques to measure and map the viscoelasticity of living tissues, cells, membranes, collagen, ECMs, and tissue engineering matrices across the spatial and temporal (from Hz to 100s of kHz) scales, and chirp-based spectroscopic techniques to assess viscoelasticity from Hz to 100s kHz at the nano and micro scale developed in my lab. I will also present tests for assessing which viscoelasic model better fits the experimental AFM results.
Our results have uncovered that extracellular matrices of both living plants and tumours present an almost perfect linear viscoelastic behaviour that is key to understand their growth and shape. I will present our work showing how the growth and shape of the roots, leaves and hypocotyl of living Arabidopsis thaliana living plants are related to the nanoscale viscoelasticity of plant cell walls at the time scales probed by multifrequency AFM.

References:
[1] “Multifrequency AFM reveals lipid membrane mechanical properties and the effect of cholesterol in modulating viscoelasticity” 2019. Z Al-Rekabi, S Contera; Proceedings of the National Academy of Sciences 115 (11), 2658-2663.
[2] “Mapping nanomechanical properties of live cells using multi-harmonic atomic force microscopy”2011 A Raman, S Trigueros, A Cartagena, APZ Stevenson, M Susilo, E Nauman, S Contera. Nature Nanotechnology 6 (12), 809.

The ability to characterize and manipulate materials down to the atomic scale for tuning their mechanical and physical properties is the cornerstone of material science. Two-dimensional (2D) materials in particular are getting large attention because their extremely thin size gives rise to unique and groundbreaking properties.
Here we explore two different research topics. On one side we show how heat delivered by a scanning probe can be used for manipulating the electrical properties of 2D materials, and on the other side we present a recent scanning probe microscopy method to perform Å indentation experiments in supported 2D materials.
Among various novel materials, two-dimensional molybdenum disulfide (MoS₂) is of particular interest due to its large band gap, low dielectric constant, and heavy carrier effective mass. Currently, a key issue in creating high performing field-effect transistors (FETs) based on MoS₂ and other transition metal dichalcogenides (TMDC) films is the poor quality of the metal contacts fabricated on these atomic layers, and the difficulty to pattern dopants. Here, we report a strategy based on thermal scanning lithography (tSPL) to fabricate metal contacts on 2D materials with high reproducibility [1], and we demonstrate the integration of tSPL with a flow-through reactive gas cell to achieve nanoscale control of dopants in monolayer MoS₂ [2]. The tSPL produced defects can present either p- or n-type doping on demand, depending on the used gasses, allowing the realization of field effect transistors, and p-n junctions with precise sub-μm spatial control, and a rectification ratio of over 10⁴. Doping and defects formation are elucidated by means of X-Ray photoelectron spectroscopy, scanning transmission electron microscopy, and density functional theory. We find that p-type doping in HCl/H₂O atmosphere is related to the rearrangement of sulfur atoms, and the formation of protruding covalent S-S bonds on the surface. Alternatively, local heating MoS₂ in N₂ produces n-character.
Moreover, we report on the elastic properties of supported 2D graphene and h-BN films by the use of modulated Å-Indentation, a novel scanning probe microscopy (SPM)-based technique capable of achieving sub-Å indentation depths during force-indentation measurements [3, 4]. By using extremely small amplitude oscillations (<<1 Å) at high frequency, we show how Å-Indentation enables non-destructive local accurate measurements of the contact stiffness and out-of-plane elastic moduli of ultra-thin ultra-stiff films, including CVD diamond films (thus even up to TPa range stiffness), as well as the transverse moduli of supported graphene and h-BN (<1 nm thickness) films. Å-Indentation thus obtains in-situ elasticity measurement combining superior resolution and indentation depths as small as 0.3 Å with AFM nanoscale topographical imaging. Thanks to this technique, we experimentally demonstrate that at room temperature and under localized indentation pressure, a single layer of graphene on top of a carbon interface layer (buffer layer), on SiC, exhibits a transverse stiffness superior to bulk diamond [3]. Extending the capabilities of the Å-indentation technique to the analysis of other 2D materials, these studies explore similar pressure-induced phase transition in hexagonal boron nitride (h-BN). We demonstrate that 2-layer h-BN flakes, on SiO₂, consistently exhibit a transverse elastic modulus almost two times larger than that of the bare substrate. This stiffening effect is observed only for flakes of thickness between 2 to 5 atomic layers, but not in single- nor multilayer (> 6-layer) h-BN, and is related to the conversion of h-BN to cubic BN (c-BN) induced by the pressure applied through the indenter.

1 Zheng et al. Nature Electronics, 2, 17–25 (2019)
2 Zheng et al. Nature Communication, (2020) https://doi.org/10.1038/s41467-020-17241-1
3 Gao et al. Nature Nanotechnology 13.2 (2018): 133
4 Cellini et al. Scientific reports 9.1 (2019): 4075

Atomic Force Microscopy (AFM) is one of the most powerful and versatile tools to image and characterize materials with nanometer scale resolution, leading to the development of nano-/bio- science and technology. While AFM technology has advanced significantly by expanding its functionality beyond a nanoscale imaging tool, the majority of AFM users still find it difficult to employ the advanced techniques and interpret the data correctly due to artifacts. Such difficulties mainly originate from the complex dynamics of the AFM probe system, used as a mechanical transducer to deliver the tip-sample interaction force.

We address this issue by altering the AFM cantilever design. The current state of the art AFM probe is a single-body system, in the shape of a rectangular or triangular beam, which is not ideal for carrying more than one type of information. In contrast, the new AFM probe, so-called paddled cantilever, is reshaped to have a two-field microcantilever design with a considerable dimensional discrepancy in the component part. By doing so, this mechanical transducer provides two discrete transduction channels such that they respond independently to the variations in surface topography and material properties/functionality. Thus, AFM users can effortlessly implement the advanced AFM schemes with minimal ambiguity, complications, and artifacts, by simply employing this new AFM probe.

Since the prototype device of paddled cantilever was first introduced, our team has taken effort to develop a batch fabrication process and, thus, have this advanced technology widely available for AFM users. The detailed dynamics of fabricated paddled cantilevers have been investigated by using the interferometric displacement sensor (IDS) with AFM to directly measure their deflection. The measured deflection along the whole cantilever revealed that the frequency of inner-paddle is not affected by the tip stiffness while its amplitude is proportional to the displacement input delivered through the tip. This is the basis how the paddled cantilever can provide two discrete transduction channels for functional AFM. A theoretical model based on a 2 degree-of-freedom system is also presented to understand the dynamic characteristics of paddled cantilever under various operational conditions. Because the design of paddled cantilever is simple to feasibly integrate into, and compatible with, current AFM probes and systems, this technology will easily extend the AFM functionality to characterize multi-physical properties beyond topography by providing a dual-channel mechanical transducer.

A novel mode of AFM, FT-NanoDMA (Fourier transform dynamical mechanical analyzer), can be used to obtain information on the viscoelastic properties of individual cells. Compared to the recently introduced AFM-NanoDMA, FT-NanoDMA allows obtaining storage and loss modulus of multiple frequencies simultaneously. Also, the FT-NanoDMA design allows measuring the mechanics of biological cells, both fixed and alive, directly in a culture dish. Although it was demonstrated on the example of plant cells, the thick wall of those cells did not allow to resolve any features of the cell structure.
Here we demonstrate the ability of FT-NanoDMA to obtain high-resolution information on the viscoelastic properties of individual eukaryotic cells, which express a highly diverse distribution of its mechanical properties. Zebrafish melanoma cells and human osteoblast cells were studied. The storage, loss modulus and loss angle (the phase difference between the sample oscillation and the response of AFM cantilever) were simultaneously recorded for 9 different frequencies of oscillation ranging between 15 to 300 Hz. The obtained maps can be compared with a fluorescent optical image of the same cells. The results are discussed and different ways of visualizing these multidimensional data presented.

A novel mode of AFM, FT-NanoDMA (Fourier transform dynamical mechanical analyzer), can be used to obtain information on the viscoelastic properties of individual cells. Compared to the recently introduced AFM-NanoDMA, FT-NanoDMA allows obtaining storage and loss modulus of multiple frequencies simultaneously. Also, the FT-NanoDMA design allows measuring the mechanics of biological cells, both fixed and alive, directly in a culture dish. Although it was demonstrated on the example of plant cells, the thick wall of those cells didn’t allow to resolve any features of the cell structure.
Here we demonstrate the ability of FT-NanoDMA to obtain high-resolution information on the viscoelastic properties of individual eukaryotic cells, which express a highly diverse distribution of its mechanical properties. Zebrafish melanoma cells and human osteoblast cells were studied. The storage, loss modulus and loss angle (the phase difference between the sample oscillation and the response of AFM cantilever) were simultaneously recorded for 9 different frequencies of oscillation ranging between 15 to 300 Hz. The obtained maps can be compared with a fluorescent optical image of the same cells. Using different load forces, we can probe mechanical properties at different indentation depth. This can bring a possibility to present a sort of tomography of cell mechanics. The excessive amount of data raises the problem of data visualization. The results are discussed and different ways of visualizing these multidimensional data are presented.

Fast, high-resolution mapping of the viscoelastic properties of soft matters, represents a major goal of atomic force microscopy (AFM) [1]. Bimodal AM-FM AFM is a suitable method for this purpose, since it allows the simultaneous acquisition of nanomechanical properties without losing in resolution and acquisition speed. This multifrequency configuration combines the robustness and simplicity of an amplitude modulation (AM) feedback in the first mode, with the sensitivity and a high signal-to-noise ratio of a frequency modulation (FM) feedback in the second mode. Finally, through the use of the appropriate contact mechanics model, it is possible to determine elastic and viscous properties of the analyzed sample [2].
Here we show how bimodal AM-FM is applied to extract the elastic properties of soft samples, such as proteins and polymers. Specifically, we characterize the elastic properties of a single protein in liquid, the 20S proteasome which in living organism plays a proteolytic role [3], and the elastic properties of a poly(styrene-block-methylmethacrylate) (PS-b-PMMA) copolymer sample [4].

[1] C. A. Amo, A. P. Perrino, A. F. Payam, and R. Garcia, ACS Nano 11, 8650 (2017).
[2] E. T. Herruzo, A. P. Perrino, and R. Garcia, Nat. Commun. 5, 3126 (2014).
[3] S. Benaglia, V. G. Gisbert, A.P. Perrino, C. A. Amo R. Garcia, Nat. Protoc. 13, 2890 (2018).
[4] S. Benaglia, C. A. Amo, R. Garcia, Nanoscale 11, 15289 (2019).

Despite the extensive use of biodegradable polyester nanoparticles for drug delivery, there is a lack of systematic studies that report the mechanics of these nanoparticles at physiologically relevant conditions. Given that nanoparticle mechanics have been reported to strongly influence nano-bio interactions, the accurate characterization of polymer nanoparticles is critical to understanding their behavior in vivo. Here, we use indentation experiments on poly(lactic acid) (PLA) and poly(lactide-co-glycolide) (PLGA) nanoparticles using atomic force microscopy (AFM) to identify their mechanical properties at different temperatures in both air and in water. While the particles are rigid at room temperature under dry conditions, their stiffness decreases by as much as 50 fold under physiological conditions (i.e. in water at 37 °C). Differential scanning calorimetry (DSC) confirms that this decrease in stiffness upon heating and immersion in water can be attributed to the phase transition of the nanoparticles from the glassy state to the rubbery state. This combination of AFM and DSC characterization allows us to determine the plasticizing effects of nanoscale size, polymer molecular weight, and immersion in water. Specifically, PLA and PLGA nanoparticles under 50 kDa molecular weight were found to exhibit glass transition temperatures below 37 °C in water, indicating that these biomedically-relevant nanoparticles will be rubbery in biological environments. Lastly, we investigate indentation experiments at elevated temperatures in air and find that insufficient AFM probe heating complicates the determination of nanoparticle temperature, thus preventing accurate recapitulation of the transition behavior for the nanoparticles. The artifacts observed while measuring particle mechanics in air further motivates nanoparticle mechanical characterization in water. Collectively, these experiments provide insight for experimentalists exploring the relationship between polymer nanoparticle mechanics and in vivo behavior, namely that thermoanalytical characterization can distinguish whether the physiologically-relevant mechanical properties can be accurately estimated using bulk material characterization and help identify processes necessary for direct characterization of polymer nanoparticles.

Naturally occurring biopolymers are promising building blocks for a new generation of multifunctional materials that can work at the biotic-abiotic interface and better interact with the environment. Particularly, silk fibroin – a structural protein extracted from Bombyx mori silkworm cocoons, has attracted a lot of attention in the recent decades due to the potential use as technical material in a broad range of applications, encompassing drug delivery, regenerative medicine, food coating, printable sensors and optoelectronics etc. As more manufacturing methodologies with silk fibroin are being developed, a key question yet to be answered is how to endow silk fibroin-based bulk materials with complex nanoscale features. Unlike most of the current nanofabrication strategies that mainly focused on manipulation of nanostructures from the top-down perspective, we choose to approach this question by designing specific folding and assembly pathways of silk fibroin that result in highly ordered β-sheeted nanofibrils. Atomic force microscopy (AFM) in combination with a set of correlative optical, spectroscopy and X-ray scattering techniques are employed to probe the detailed mechanism of nanofibrils formation, from the driving forces for disorder-to-order transition to sequence-structure-property relationships across scales. In particular, high resolution nanomechanical data of silk fibroin at single fibril level in both dried and hydrated states are acquired by using AFM-based bimodal force microscopy and fast force mapping. Factors including silk fibroin molecular weight and crystallinity are found to directly correlate with the nanomechanical properties.

The mechanical response of heterogeneous samples like polymer composites and thin films is controlled by the properties of the microstructure of the material. Since confinement effects and interphase formation can alter properties of the microphases, measurements on these samples must be localized at the nanometer scale to provide the needed property distribution. Mechanical properties of polymers are generally time dependent, so a full understanding requires measurements over a range of frequencies and temperatures as is provided (for bulk samples) by techniques such as dynamic mechanical analysis (DMA).
Atomic Force Microscopy (AFM) has the resolution and mechanical sensitivity to provide the needed local mechanical property information. However, previous AFM measurements of viscoelastic properties have suffered from some significant drawbacks. Contact resonance [1], for example, provides mechanical property maps at known frequencies, but the cantilever resonances are many orders of magnitude higher than DMA, making comparisons indirect at best. Intermittent contact methods like TappingMode [2], force volume, and PeakForce Tapping [3] face challenges in calculating intrinsic mechanical properties like storage and loss modulus (or tan delta) at well-defined frequencies due to the non-linear process of making and breaking contact.
A new mode called AFM-nDMA provides viscoelastic results which can be directly compared with bulk DMA. It provides spectra of storage and loss modulus across frequency and temperature allowing construction of master curves through Time Temperature Superposition [4]. These measurements are localized by the probe-sample contact, allowing observations to be co-located with high-resolution AFM imaging modes. This presentation will apply AFM-nDMA to investigate the mechanical properties of heterogeneous polymeric samples near interfaces between components and within microscopic domains. Time (and temperature) dependence of these properties will also be examined.
[1] U. Rabe, S. Amelio, E. Kester, V. Scherer, S. Hirsekorn, and W. Arnold, Ultrasonics, 2000, 38, 430.
[2] O. Sahin, C. Quate, O. Solgaard, and A. Atalar, Phys. Rev. B, 2004, 69, 1.
[3] B. Pittenger and D. G. Yablon, Bruker App. Note 149, 2017, doi: 10.13140/RG.2.2.15272.67844
[4] M. L. Williams, R. F. Landel, and J. D. Ferry, J. Am. Chem. Soc.,1955, 77, 3701.

Tomographic Atomic Force Microscopy (AFM) uniquely enables volumetric mapping of nanoscale materials properties such as ferroelectricity, conductivity or photoconductivity. Based on alternating imaging and milling with the AFM probe itself, this approach is effective for surface planarization to diminish topographic artifacts. It can also be extended for more complex surface patterning, such as fabricating smooth or stepped gradients in the remaining material thickness. Accordingly, the milling efficiency and conditions are compared for epitaxial monolithic thin films, multilayers, and polycrystalline samples, with the results demonstrating a coupling to bulk mechanical properties and also grain boundary energies. Such insight is beneficial to optimizing future AFM-based shape-engineered surfaces, which are valuable for follow-on investigations of thickness dependencies, subsurface structures, and other volumetric inhomogeneities.

Scanning electrochemical probe microscopy (SEPM) techniques such as scanning electrochemical microscopy (SECM) and in particular the hybrid techniques such as atomic force microscopy combined with scanning electrochemical microscopy (AFM-SECM) enable high resolution topographical imaging along with electrochemical information [1] but also electrochemical force spectroscopic measurements [2]. Recently, we introduced a new type of electrochemical scanning probe tip for combined AFM-SECM measurements providing instead of a conventional AFM tip, a conductive colloidal sphere attached to an otherwise insulated cantilever [3]. Such conductive spherical probes may consist of boron-doped diamond [4], a highly attractive electrode material, or may be electrochemically modified, e.g., functional polymers, suitable for force spectroscopic measurements such as adhesion measurements under potential control.
In this contribution, we present AFM-SECM measurements and electrochemical force spectroscopy using conductive colloidal AFM probes. Due to the spherical geometry and the low spring constant of the cantilever, these probes are highly suitable for mapping nanomechanical properties at soft samples, i.e. bacterial cells. Colloidal probes with BDD as spherical electrode can be used for various SPM experiments including conductive AFM given the physical and electrochemical inertness of BDD, for force spectroscopy under potential control providing information on electrified interfaces, and for AFM-SECM measurements. Notably, all these experiments can be performed with the same BDD-AFM-SECM probe. Colloidal probes can be modified with polymers i.e. polydopamine, which properties may be altered by applying potential to the spherical electrode. Single cell force spectroscopy on bacterial adhesion along with studies at electrified interfaces will be presented.

[1] A. Patel, C. Kranz, Annu. Rev. Anal. Chem. 2018, 11, 329.
[2] S. Daboss, J. Lin, M. Goddjehans, C. Kranz, 2020, in press
[3] P. Knittel, H. Zhang, C. Kranz, G. G. Wallace, M. J. Higgins, Nanoscale 2016, 8, 4475.
[4] S. Daboss, P. Knittel, C. E. Nebel, C. Kranz, Small 2019, 15, 1902099.

The manipulation of single molecules has become possible with the continued development of scanning probe microscopy techniques. Thereby, the manipulation path can be chosen at will, and energy barriers between potential minima on that pathway can be quantified, as can the energy landscape around the molecule before and after manipulation. To explore the practicality of this novel pathway to catalysis research, we selected iodobenzene molecules on a Cu (100) surface as a model system. To this end, we first break the iodine atom from the benzene ring by applying a controlled bias voltage. A specific manipulation path is chosen and then the tip is moved at constant but continuously reduced heights while recording the oscillation amplitude A and phase phi with the microscope operated in our recently developed tuned-oscillator (TO) detection scheme [1]. To preserve the accuracy of the recovered tip-sample interaction potentials and forces, we use oscillation amplitudes significantly larger than the decay length of the tip-sample interaction potential [2,3]. We then manipulated the benzene radical away from the iodine atom and measured the required energy barrier. With increasing distance from the iodine atom, the energy barrier required to move benzene radical decreases. Our experiments are disclosing the effect of local surface chemistry on controlled manipulation of molecules.For further insight, we also compared with computational results obtained using a Bayesian Optimization Structure Search (BOSS) protocol [4], a machine learning technique developed at Aalto University that delivers the minimum energy pathways and entire energy landscapes with minimal computational effort.

References:
[1] O. E. Dagdeviren et al., Nanotechnology 27, 065703 (2016).
[2] O. E. Dagdeviren et al., Physical Review Applied 9, 044040 (2018).
[3] O. E. Dagdeviren et al., Review of Scientific Instruments 90, 033707 (2019).
[4] M. Todorović et al., NPJ Computational Materials 5, 35 (2019). <div id="UMS_TOOLTIP" style="position: absolute; cursor: pointer; z-index: 2147483647; background-color: transparent; top: -100000px; left: -100000px; background-position: initial initial; background-repeat: initial initial;"> </div>

Ferroelectric domain walls are naturally occurring interfaces, separating regions with different orientation of the spontaneous polarization. The domain walls can have fundamentally different electronic properties than the surrounding domains. It is now established that the walls exhibit unusual electronic conduction under direct voltages (d.c.) and distinct vibrational modes under alternating voltages (a.c.) in the microwave regime. In contrast, very little is known about their behavior at low a.c. frequencies, where the walls respond adiabatically.
Here, we map the response of charged ferroelectric domain walls in the adiabatic regime under a.c. voltages in the kilo- and megahertz range. In our model material, ErMnO₃, a.c.-cAFM conduction contrasts drop at significantly different frequencies for different types of domain walls. The characteristic frequencies at which the domain wall contrast vanishes is determined by the local polarization configuration of the adjacent domains. Extending the research on electronic transport mechanisms into the a.c. regime is an important step towards a better understand of the unusual conduction at ferroelectric domain walls and the future development of domain wall based nanoelectronic devices, such as filters, phase shifters, or tunable capacitors.

Tomographic AFM (TAFM) extends atomic force microscopy beyond the measurement of surface properties and into the realm of three-dimensional analysis. For certain materials, the AFM tip can be used as a both a probe and a milling tool, measuring properties such as current or piezoresponse while simultaneously removing material from the surface. Instead of thousands of 2-D pixels, this yields datasets with millions of 3-D voxels, but all still with nanoscale resolution. The resulting tomographic reconstructions enable unprecedented visualization and analysis of heterogeneous materials, particularly those with isolated or networked sub-surface features, surface coatings, or vertical property gradients. This is especially applicable for 3-D mapping of thickness-dependent properties in functional materials including ferroelectric domain structures and interconnected conduction pathways. Examples are highlighted for both BTO/STO superlattices with multilayer chip capacitor applications, and also ferroelectric domain wall-based nanoelectronic devices such as single crystal ErMnO₃.

Ferroelectric thin films epitaxially grown on silicon have attracted great interest for their potential application in energy-efficient electronics for their controllable ferroelectric polarization switching. As ferroelectric materials are made thinner, however, the ferroelectricity is usually suppressed, especially in the archetypal ferroelectric perovskite systems have so far proved unsuitable for thickness scaling and integration with modern semiconductor processes. Recently, Zr-doped HfO₂ grown by atomic layer deposition on silicon has been to be an exciting thin film ferroelectric material that is compatible with conventional Si processing[1, 2] However, such switching is not be easily demonstrated by conventional bulk-based characterization techniques due to the current leakage involved with such ultrathin films. Even though the conventional scanning probe microscopy-based PFM has unique advantages in terms of nanoscale spatial resolution and ease of use, but is subject to crosstalk associated with long-range cantilever-sample electrostatic interactions which can be mistaken for localized piezoelectricity. We have developed and implemented an interferometric displacement sensor (IDS) that mitigates this problem.[3, 4] In addition, we have developed a new, high-sensitivity scanning capacitance microscopy (SCM) module. Recently, we have used these new approaches to study ferroelectricity in ultra-thin HZO films (1 nm).[2] We will discuss SCM and IDS butterfly loops that provide unambiguous evidence of localized ferroelectric behavior. Most importantly, both techniques eliminate the commonly observed artifacts thus provide conclusive evidence of polarization switching in this technologically relevant ultra-thin ferroelectric film grown on silicon.
[1] Chernikova, A. et al, “Ultrathin Hf0.5Zr0.5O2 Ferroelectric Films on Si”, ACS Appl. Mater. Interfaces 8, 7232−7237 (2016).
[2] S Cheema, et.al, “Enhanced ferroelectricity in ultrathin films grown directly on silicon.” Nature 580, 478–482 (2020).
[3] Labuda, A and R. Proksch “Quantitative measurements of electromechanical response with a combined optical beam and interferometric atomic force microscope”, Applied Physics Letters 106, 253103 (2015).
[4] Collins, L. et.al, “Quantitative electromechanical atomic force microscopy.” ACS Nano 13, 8055–8066 (2019).

In the last years new open-loop (OL) Kelvin probe force microscopy (KPFM) modes [1] were introduced to facilitate a faster and more accurate surface potential characterization at the nanoscale. Based on high-speed data capture, the OL KPFM modes observe the tip-sample electrostatic interaction in the time domain, which potentially can increase the temporal resolution by few orders of magnitude in comparison to common closed-loop (CL) KPFM implementations that rely on frequency domain detection [1, 2]. In this work, the high-speed digitization for open-loop sideband frequency-modulation (FM) KPFM is presented in terms of amplitude and frequency responses of the cantilever. In both cases, the tip-sample contact potential difference (CPD) is extracted from the time series analysis of the signals. This interpretation is demonstrated to be superior to the more common used analysis of the parabolic bias dependence of the OL KPFM signals. The OL sideband FM-KPFM amplitude and frequency responses were compared side-by-side to their CL AM-KPFM and CL FM-KPFM counterparts on materials of different surface potentials and subjected to various bias voltages. Notably, the OL sideband FM-KPFM frequency retains the tip-confinement sensitivity that is characteristic to the FM-KPFM based modes while the OL sideband FM-KPFM amplitude suffers from cantilever stray capacitive couplings much in the same way as the common CL AM-KPFM mode [3].

[1] L. Collins, A. Belianinov, S. Somnath, N. Balke, S. V. Kalinin, and S. Jesse, Sci. Reports 6, 30557 (2016).
[2] L. Collins, M. Ahmadi, T. Wu, B. Hu, S. V. Kalinin, and S. Jesse, ACS Nano 11, 8717 (2017).
[3] G. Stan, Nanotechnology in press (2020).

Interfaces characterization is crucial for several types of devices like light emitting diodes, solar cells, photodetectors or transistors. In the particular case of photodetectors it is important to characterize such interfaces to see the impact of passivation, where defects present at the interfaces impact the dark current of the device. Moreover, such parameters like surface photovoltage (SPV) or carrier lifetime can reflect the influence of defects, even when embedded [L Borowik et al, Nanotech. (2010)].
For this particular purpose, we use heterodyne Kelvin Probe Force Microscopy (h-KPFM) [Yasuhiro Sugawara et al, Appl. Phys. Lett. (2012)] under illumination to characterize the silicon interfaces embedded under different types of oxides, specifically Al₂O₃, Al₂O₃/Ta₂O₅, HfO₂, SiN, SiO₂. In addition, we adopt a pump probe approach [Murawski, J. et al, J. Appl. Phys. (2015)] in order to time-resolve the photovoltage (tr-SPV). Besides of the capability to measure SPV, tr-SPV and charge carrier dynamics, the particularity of this setup is the analysis of these parameters as function of different illumination wavelengths and the possibility to perform in-depth dependent measurements.
In this work, we will present the developed experimental method using h-KPFM under frequency modulated and wavelength dependent illumination in order to probe interfacial physical processes (i.e. the charge carrier dynamics and the influence of interface on surface photovoltage). Furthermore, we will correlate the h-KPFM results with additional interfacial characterization procedures (Corona Oxide Characterization Of Semiconductor method) [Marshall Wilson et al, AIP Conference Proceedings (2001)]. Finally, we will demonstrate the interest of a time-resolved approach (i.e. pump probe h-KPFM) for the characterization of embedded interfaces.

Ge₂ Sb₂ Te₅ (GST-225) is one of the most promising materials for use in phase change memory devices, which presents good scalability and high endurance. The operation of the PCM devices consists in switching between a high conductive (On) state and a low conductive (Off) state, which correspond respectively to the crystalline and amorphous phase. The switching from the OFF to the ON state (Set) is obtained by applying voltage pulses that heat the GST-225 material by Joule effet, leading to its crystallization. The switching from the ON state to the OFF state (Reset) is obtained by applying shorter pulses with higher voltages in order to quench the GST-225 and obtain the amorphous phase. For resistive memory cells the current necessary to obtain the phase change is in the order of serveral hundred of microamperes.
Conductive AFM (C-AFM) is widely used for the characterization of the electrical properties of materials at the nanometer scale. It allows in principle to obtain electrical information at nanometric scale while avoiding expensive lithographic processes. At present, the cyclability of writing and erasing process with C-AFM is limited by various pratical factors, namely the high oxidation of the GST-225 in ambient conditions and the current limitation due to the tip degradation. For those reasons, successive Set and Reset steps were not observable, until now, on unpatterned samples by C-AFM. Those limitations can be overcome by using ultra-high vaccum in order to protect the GST-225 against oxidation and utilization of full diamond tips that resist to high current conditions.
In this work, we studied the electrical characteristics of phase change materials at nanometric scale by C-AFM under Ultra High Vaccum in pulse mode. We show that the used experimental configuration allows to cycle unpatterned GST-225 layers between the set and the reset state. We measured the resistive switching characteristics of GST-225 and reached 50 cycles. The experimental results will be compared to temperature and resulting crystalline fraction obtained by finite element simulation.

The molecular organisation and dynamics of liquids near the surface of solids underpins processes such as molecular exchanges, electrochemistry, nanofluidics, biomolecular function and lubrication. Depending on the local topographical and chemical properties of the solid, the behaviour of the interfacing liquid can vary significantly at the nanoscale, with consequences for interfacial processes. Scanning probe microscopies can derive local insights into the equilibrium organisation of interfacial liquids, often with sub-nanometre resolution. However, deriving quantitative information about the local liquid dynamics remains a considerable challenge.
Here we review some of our recent results obtained by combining high-resolution atomic force microscopy with highly localised rheological measurements to deduce quantitative information about the motion of liquids and dissolved molecules along immersed solids. We show that the dynamics is highly system-dependant and can be dominated by either the local configurational entropy or the interface or enthalpic interactions. We also describe a novel approach to quantify the preferred flow direction naturally adopted by liquids near walls, locally and with nanometre precision. The approach, based on AFM, combines a vertically oscillating tip with high frequency lateral oscillations to derive directional information about the interfacial liquid at each point of the system. To illustrate the method’s capabilities, we investigate the dynamics of aqueous solutions containing different metal ions along the surface of a graphene oxide flake. We identify ion-specific flow dynamics that can be related to the well-established sieving properties of graphene oxide membranes.

Investigating nucleation from solutions is challenging, because it is a consequence of unstable density fluctuations, making the structures and events that must be probed both transient in nature and small in spatial extent. Moreover, when nucleation is heterogeneous, it is inherently linked to the structure and dynamics of the interfacial region between the bulk solution and the underlying substrate. Thus high-speed, molecularly resolved AFM combined with fast force mapping (FFM) provides a powerful and unrivalled combination to investigate nucleation from solution onto surfaces. Here I illustrate the deep level of fundamental insight into mechanisms of nucleation and the structure of the interfacial region made possible with these methods using results from investigations into nucleation of gibbsite (Al(OH)₃) on muscovite mica (001) and the structure of electrolyte solutions above mica and boemite (AlOOH) surfaces. In the case of gibbsite nucleation, we combine AFM observations of individual molecular adsorbates, transient clusters, and stable islands with Monte Carlo simulations to put together a coherent picture of surface speciation and nucleation. The results reveal a surface population of ions that is dominated by hydrolyzed species (Al(OH)₂⁺ and Al(OH)²⁺) even though Al³⁺ vastly dominates the bulk solution. These adsorbed ions evolve into subcritical clusters with increasing saturation state and temperature, constituting a population that decreases exponentially with size and exhibits dynamic fluctuations consistent with classical predictions. However, severe discrepancies with classical theory emerge when the values of key thermodynamic parameters are extracted from the AFM data. These discrepancies are resolved when the impact adsorbate charge on the capacitance of the mica-solution system is taken into account in calculating the work of cluster formation, but reveal a film formation process that occurs within a single phase region. Thus, the results imply that the gibbsite films constitutes a nanostructured 2D mesophase. To define the solution structure above boehmite and mica surfaces at lattice resolution, we developed a self-consistent scheme to decouple long-range tip-sample interactions from short-range solvation forces. In the case of boehmite, the results are benchmarked against molecular dynamics simulations that explicitly include the effects of the tip with different levels of approximation and systematically account for tip size, chemistry, and confinement effects. We find four laterally structured water layers within one nanometer of the surface, with the highest water densities at sites adjacent to hydroxyl groups. The findings reveal a complex relationship between site-specific chemistry, water density, and long-range particle interactions; and represent a major step forward towards quantitative data interpretation and widespread implementation of 3D FFM.

Collagen is the most abundant structural protein of the extracellular matrix. The assembly of collagen fibrils play relevant roles in a variety biological processes. The formation fibrils during the self-assembly process of collagen I have been studied by AFM [1,2]. Those studies lacked the time and mechanical properties resolution to clarify the mechanism of the earlier stages of collagen assembly and fibril structure formation. We have developed a high-speed bimodal AFM that combines the ms time resolution of high-speed AFM [3] with the nanomechanical force sensitivity of bimodal AFM [4,5]. High-speed bimodal AFM characterizes the earliest stages of the self-assembly of the collagen fibrils by proving time-resolved and high-spatial resolution maps of the evolution of the elasticity of the fibrils during the growth.

[1] F. Jiang, H. Hörber, J. Howard, D.J. Müller., J. Struct. Biol. 148, 268–278 (2004).
[2] D.R. Stamov, E. Stock, C.M. Franz, T. Jähnke, H. Haschke, Ultramicroscopy 149, 86–94 (2015).
[3] T. Ando. Nanotechnology 23, 062001 ( 2012)
[4] S. Benaglia, V.G. Gisbert, A.P. Perrino, C.A. Amo, R. Garcia, Nat. Protoc. 13, 2890 (2018)
[5] E.T. Herruzo, A.P. Perrino, R. Garcia, Nat. Commun. 5, 3126 (2014).

Solid-liquid interfaces are highly relevant for a number of phenomena in nature and technology. However, when it comes to hydrophobic surfaces, the structure of water in close proximity to the surface and its role in mediating interactions are not well understood. Two dimensional materials provide a variety of large and atomically flat surfaces that are mildly hydrophobic. We exploited the angstrom resolution capabilities of three-dimensional AFM [1,2] to image the interfacial water organization on MoS2, MoSe2, WS2, WSe2 and graphene.
We demonstrate that these interfaces are characterized by layers of oscillating density within 2 nm above the solid surface (Fig. 1a). The distances between adjacent layers for MoS2, MoSe2, WS2 and WSe2 and graphene are ≈ 0.50 nm (Fig. 1b). This value is larger than the one predicted and measured for water density oscillations (≈ 0.30 nm) [2,3]. The experiments indicate that on 2D material surfaces water molecules are expelled from the vicinity of the surface and replaced by several molecular-size hydrophobic solvation layers.

References:

[1] T. Fukuma, Y. Ueda, S. Yoshioka, and H. Asakawa, Phys. Rev. Lett. 104 (2010) 016101
[2] T. Fukuma, R. Garcia, ACS Nano 12 (2018) 11785
[3] D. Martin-Jimenez, R. Garcia, Nat. Comm. 7 (2016) 12164
[4] M. R. Uhlig, D. Martin-Jimenez, R. Garcia, Nat. Comm. 10 (2019) 2606

High-throughput screening technologies such as microtiter plates enable chemists and biologists to simultaneously perform thousands of experiments in an automated fashion. While transformatively smaller than the vials used by the prior generation, the typical working volume for microtiter plates is in the range of microliters and there are still far too many compounds to study even with this degree of miniaturization. Scanning probe lithography (SPL) could provide a path to continuing this miniaturization in fluid handling as it encompasses approaches to pattern fluids with volumes at least 10⁹ times smaller than what is achievable in microtiter plates. While the patterning of such sub-fL volumes has been demonstrated with various different materials over the years, to achieve controlled and reliable experimentation with SPL, it is crucial to have precise control over the size of features and amount of fluid that is transferred in real time. Here, we enable closed-loop control over the fluid written using SPL by combining a novel process for quantifying the fluid on a cantilever, the utilization of a tipless cantilever architecture, and a control strategy for adjusting writing parameters in situ. Specifically, we employ a two harmonic mode inertial sensing algorithm to reliably quantify the amount and location of fluid on the cantilever. Critically, sensing the mass of liquid on the cantilever before and after a fluid transfer event allows us to monitor the deposited amount with precision approaching the fL scale. This capability allows us to investigate the fundamental mechanisms and characteristics of fluid transport at this scale. Taking inspiration from experiments of fluid transfer between macroscopic parallel plates, we find that the fraction of the liquid that is transferred from the cantilever to the substrate is only a function of speed of the cantilever and surface energies of the substrates. Based on this discovery, we show that the amount of fluid that will be transferred can be predicted based on the amount of ink on the cantilever, hence it is possible to control the transferred amount by selecting writing parameters. With this insight in hand, we develop a closed-loop process for writing patterns with designed feature sizes. Given the versatility of scanning probes to characterize fluids once they are written, this work lays the foundation for multifunctional research systems in which chemical experiments are set up, performed, and evaluated at the nanometer scale.

AFM is a valuable tool for biologists and biophysicists to investigate structure, mechanics, and dynamics of biological samples from single proteins or DNA molecules through lipid layers, protein matrices and hydrogels to whole cells or cellular assemblies. To preserve the physiological, hydrated state of these entities, most of these experiments are performed in an aqueous environment. Our newest instrument, the P6, is optimized to facilitate both high-resolution and large area measurements in any liquid environment. The fully motorized scan head includes a three-axis flexure-based tip scanner. Its proprietary design features an ultra-low-noise detection system with non-moving light sources. Moreover, the scan head is equipped with photothermal excitation (PTE). This is superior to the common piezo-acoustic excitation as it solely drives the cantilever without moving any other parts of the system, resulting in unparalleled excitation stability, especially in liquid operation. Textbook-like cantilever tuning results in high ease-of-use imaging and enables quantitative understanding of tip-sample interaction forces. The linear frequency response and high bandwidth of PTE enables multifrequency and high-speed applications. The instrument is compatible with any inverted optical microscope allowing for combined investigation of biological samples with AFM and various optical microscopy techniques, such as fluorescence imaging or super resolution imaging techniques. Apart from imaging, PTE also opens up new horizons for measurement modes and analysis techniques. PTE is, for instance, used to measure the mass of living cells at unprecedented time and mass resolution. [2] More advances in PTE-based techniques are to be expected soon.
Here, we show how we could overcome the engineering challenges posed by the different features of the scan head (full motorization, tip scanner and PTE) by employing a new opto-mechanic design. [1] This allowed building a stiff, large-range scanner that, in combination with sophisticated lowest noise-electronics and PTE, allows to routinely achieve high-resolution imaging in aqueous buffer solution, such as revealing the secondary structure of dsDNA.

[1] EP3450994A1
[2] D. Martinez-Martin et al. (2017) Nature 550: 500-5

Liquid-solid interfaces occur in many systems, such as batteries, catalysts, or in naturally occurring geochemical systems. Surface reactions may be strongly influenced by the presence of a liquid. In heterogenous catalysis, a solvation environment can be crucial in enabling surface certain reactions, such as oxidation reactions or electro catalysis reactions. The presence of the liquid-solid interface can speed up the reactions, increase the selectivity of products and open up new favorable reaction pathways. Therefore, it is important to understand the role of solvent in the liquid-solid interface surface reactions. For electrochemical systems, such as in batteries or electrocatalysts, the electrode potential can also have a significant impact on the chemisorption of molecules at metal surfaces as well as surface reactions. For instance, the electric field can affect the adsorption strength, the adsorption site and the intermolecular vibrational frequencies. Solvent and electrical field effects may both strongly influence surface reactivity. However, modeling efforts often ignore these two phenomena, or may not treat such phenomena appropriately. Methods to better describe electrochemical environments are still being developed and are an ongoing area of research. Herein, we modeled using density functional theory to simulate the ethanol oxidation reactions (EOR). We assessed how both the solvent and electrical fields influence important C-C and C-H bond scission reactions. Different solvation models exist for treating solvation. Implicit solvation models treat the solvent as a continuum and are generally “cheap” in terms of computational time. Explicit solvation models involve modeling the solvent molecules using density functional theory. Such calculations may be more computationally expensive, but in principle more accurate. Hybrid solvation models combine both the implicit and explicit approaches. We evaluated all three solvation models for describing surface chemistry on the Rh(111) surface, which is a typical catalyst for the EOR. Our results show that the reactions involving C-C and C-H scission are favorable in the presence of explicit water molecules due to intermolecular H bonding. However, the influence of hydrogen bond is negligible in the implicit solvation models. Thus, implicit models fail to properly describe solvent effects on ethanol oxidation. Our results do indicate that C-C and C-H bond scission becomes more favored when using a hybrid solvation model. The hybrid solvation model allows an efficient combination of both implicit and explicit solvation approaches. Field effects are also important in that they can have cause the adsorption energies of intermediates to fluctuate depending on the applied potential. The potential also can affect the energy barrier for bond breaking reactions. Our results assess the influence of the applied potential, and how modulating the potential may influence desired reactions. Overall, our work provides valuable insights on the effects of solvation and electric fields on surface chemistry in systems like batteries or electrocatalysts. We highlight a compelling approach to accurately and efficiently modeling electrochemical interfaces.

Scanning electrochemical cell microscopy (SECCM) is a nanoelectrochemical scanning probe technique that creates a confined electrochemical cell (droplet) via meniscus contacting the surface by employing an electrolyte filled micropipette containing quasi-reference counter electrode(QRCE).[1] It is particularly useful to investigate the heterogeneity, or local electrochemical properties of the electrode surfaces by direct mapping the electroactivity with nanoscale resolution.
In this study, the electrochemically reversible [Ru (NH₃)₆]³⁺/²⁺ electron transfer process at a highly ordered pyrolytic graphite (HOPG) surface was recorded using Park NX12 AFM system. A single barrel glass nanopipette with Ag/AgCl inserted as QRCE was utilized. Using previous successful experience in commercializing pipette-based electrochemical microscopy,[2] Park Systems’ hardware and software enables localized nanoscopic cyclic voltammetry measurements conducted at each pixel each time the meniscus contacts the surface. Thus, a spatially resolved surface electroactivity mapping of HOPG at various scan rates were created with high-throughput at the micro-/nanoscale. The difference in electrochemical activity measured at the edge plane relative to the basal plane surface was shown.
This work demonstrates that Park Systems’ commercial SECCM option is promising for quantitative electroanalysis at the nanoscale, with potential applications in energy storage (battery) studies and corrosion research.
[1] Gao, R., Edwards, M. A., Qiu, Y., Barman, K., & White, H. S. (2020). Visualization of Hydrogen Evolution at Individual Platinum Nanoparticles at a Buried Interface. Journal of the American Chemical Society.
[2] Shi, W., Goo, D., Jung, G., Pascual, G., Kim, B., & Lee, K. Simultaneous Topographical and Electrochemical Mapping using Scanning Ion Conductance Microscopy–Scanning Electrochemical Microscopy (SICM-SECM).

Imaging ion transport in nanostructure materials has applications in systems from neuromorphic computing, to electrochemical energy storage, and solar cells. We utilize multimodal imaging to probe transport phenomena in these systems. For instance, by combining nanoscale vibrational spectroscopy with structural imaging we can correlated local anion diffusion with structural phase transitions in conjugated polymer films being explored for neuromorphic computing and bioelectronics applications. We also combine similar IR imaging with time-resolved electrostatic force microscopy (trEFM) to uncover the dynamics of ion transport in various 3D and 2D halide perovskites of interest for applications in optoelectronics, especially solar cells, where ion transport can be a primary contributor to device aging and failure. This work bridges advances in scanning probe microscopy with tools from data science to better understand materials design rules for new applications.

The functional properties of halide perovskite semiconductors at the micro and nanoscales are heterogeneous,¹ and there is a pressing need to understand how these properties (electrical, chemical, etc.) influence one another and affect overall photovoltaic (PV) performance. In this presentation, we begin by showing a variety of correlative microscopy methods that have been applied to perovskite solar cells to investigate the local electrical, chemical, and optical properties of single-junction devices.^2–4 For this, Kelvin probe force microscopy, nano-IR imaging, time-of-flight secondary ion mass spectroscopy depth profiling among other characterization techniques were implemented. These results reveal that the perovskite’s (i) nanoscale voltage response depends on its chemical composition (ii) charge carrier dynamics depend on the incident photon energy (i.e. wavelength), and (iii) altered chemical properties due to the inclusion of potassium additives impact the electrical response in three dimensions.

Beyond single-junction perovskite PV, and a promising real-world application for this material class, is the multi-junction perovskite/c-Si solar cell. Not only are perovskites complex on their own, but when embedded within a textured heterojunction, the interactions between device architecture and material performance introduces new challenges. Commercial c-Si solar cells are etched to form micro-scale, randomly distributed pyramids on their front side, complicating the growth of the perovskite top cell. It is not well understood how the local light-matter interactions of the perovskite material behave when incorporated into this type of multijunction architecture. Therefore, we present optoelectronic microscopy measurements correlated with optical simulations of perovskite on textured c-Si solar cells. Through wide-field, hyperspectral photoluminescence (PL) imaging, we measure the photon out-coupling of the perovskite material and find both a spectral and spatial dependence on the geometrical patterning which dominates any grain-to-grain variation that is typically found in perovskite PL measurements.⁵ Such heterogeneity reveals the importance of the underlying c-Si texture design, suggesting that tuning the surface morphology could lead to a homogenization of the perovskite’s PL emission, which may boost the multi-junction solar cell efficiency.

(1) Tennyson, E. M.; Doherty, T. A. S.; Stranks, S. D. Heterogeneity at Multiple Length Scales in Halide Perovskite Semiconductors. Nat Rev Mater 2019, 4 (9), 573–587 - invited review, front cover
(2) Tennyson, E. M.; Roose, B.; Garrett, J. L.; Gong, C.; Munday, J. N.; Abate, A.; Leite, M. S. Cesium-Incorporated Triple Cation Perovskites Deliver Fully Reversible and Stable Nanoscale Voltage Response. ACS Nano 2019, 13 (2), 1538–1546 - cover
(3) Tennyson, E. M.; Howard, J. M.; Roose, B.; Garrett, J. L.; Munday, J. N.; Abate, A.; Leite, M. S. The Effects of Incident Photon Energy on the Time-Dependent Voltage Response of Lead Halide Perovskites. Chem. Mater. 2019, 31 (21), 8969–8976 - front cover
(4) Tennyson, E. M.; Abdi-Jalebi, M.; Ji, K.; Garrett, J. L.; Gong, C.; Pawlicki, A., A.; Ovchinnikova, O., S.; Munday, J. N.; Stranks, S. D.; Leite, M. S. Correlated Electrical and Chemical Nanoscale Properties in Potassium-Passivated, Triple-Cation Perovskite Solar Cells. Adv. Mater. Inter. 2020 - in press
(5) Tennyson, E.M.; Frohna K.; Drake, W.K.; Sahli, F.; Yang, T. C.-Y.; Fu, F.; Werner, J.; Jeangros, Q.; Ballif, C.; Stranks, S.D. Multimodal Microscale Imaging of Textured Perovskite/c-Si Tandem Solar Cells - in preparation

The design, synthesis, and characterization of new semiconductor materials are important for the growth of electronics and energy solutions such as solar cells and fuel cells. Perovskite materials in particular are being investigated due to their compositional flexibility resulting from a wide range of stoichiometry in their design in contrast to conventional materials like silicon, thereby leading to a wide variety of optoelectronic properties. Some perovskites such as organic-inorganic lead halide perovskites and perovskite oxides are well-known for their power conversion efficiency and for their potential catalytic applications respectively. However, their respective toxicity / instability issues and the large energy gaps (>3 eV) pose serious challenges for their practical optoelectronic applications. Recently, transitional metal perovskite chalcogenides have emerged as a viable semiconductor that can address these issues (stability, abundance, toxicity, cost, and performance) as high performance photovoltaic materials, but a detailed understanding of their electronic and optical properties remains a key open question.
Here, we report the Scanning Probe Microscopy studies on the surfaces of single crystals of transition metal perovskite chalcogenides such as barium zirconium sulfide (BaZrS₃, Ba₃Zr₂S₇) using Kelvin Probe Force Microscopy (KPFM), and Raman and photoluminescence spectroscopy. Besides nominally undoped crystals, we also studied transitional metal doped (Cr doping on Zr sites). While the nominally undoped BaZrS₃ has a bandgap of 1.8 eV, it shows strong photoluminescence at ~1.25 eV when a dopant such as chromium is introduced. KPFM was used to map the surface potential as well as electro-mechanical contrast proportional to capacitance. In addition, a precise optical coupling of the Raman microscope with the Scanning Probe Microscope allows surface potential changes to be mapped under dark and illuminated conditions. The surface potential of Ba₃Zr₂S₇ and Cr-doped BaZrS₃ changes dramatically in a reversible manner with laser illumination, highlighting the strong light-matter interaction and luminescent characteristics. It was found that this behavior is strongly dependent on the wavelength and laser power. The measured contact potential difference values provide indirect information on the evolution of work function during light illumination providing qualitative information on the quasi-Fermi level splitting induced by the photo-excitation. In summary, a unique combination of optical microscopy and scanning probe microscopy instrumentation were used to study optical, electronic, and photovoltaic properties of perovskite chalcogenides to evaluate their suitability for advanced photonic and energy applications.

Metal halide perovskites (MHPs) have shown promising applications in optoelectronic devices. Further improvement of MHPs devices toward commercialization necessitates a clear understanding of the factors impacting the functionality of MHPs. To date, ionic migration, ferroic behavior, structural distortion are mostly debated mechanisms credited with the functionality of MHPs. Given the interconnected nature of ionic states, ferroic properties, and structural distortion due to the screening effect and the compensation effect, the interactions among these properties can no longer be ignored. Here, in the combination of scanning probe microscopy (SPM) with other chemical and functional imaging techniques, as well as the machine learning approach, we have systematically investigated the ion migration, ferroic behavior, lattice strain, crystallographic orientation, and charge transport in MHPs and discovered a number of interesting phenomena. We have discovered the hysteretic ion migration—including intrinsic compositional ions (e.g. CH₃NH₃⁺, I^-) and decomposition chemical species (e.g. CH₃NH₃⁺ decomposition products)—in MHPs based devices by directly detecting ion migration by a newly developed time-resolved time-of-flight secondary ion mass spectrometry (tr-ToF-SIMS). Using time-resolved Kelvin probe force microscopy (tr-KPFM), we revealed how this hysteretic ion migration affects the local charge distribution in MHPs. We for the first time specified the contribution of CH₃NH₃⁺ and I^- migration to the current-voltage hysteresis in an MHP device. In addition, we found that the CH₃NH₃⁺ migration is strongly coupled with a local lattice strain change and hence impacts the ferroic behavior of CH₃NH₃PbI₃. Confocal photoluminescence microscopy study combining with piezoresponse force microscopy, electron backscatter diffraction, etc. directly unveiled the impact of chemical, strain, and ferroic properties on the charge recombination and transfer in MHPs. Altogether, our work offers a picture describing the interconnection of ionic, structural, ferroic, and optoelectronic properties of MHPs, which is expected to help to enhance the performance and the stability of MHPs-based optoelectronic devices.

The electrodes of a polymer electrolyte membrane fuel cell (PEMFC) primarily contain a Pt/C catalyst and Nafion binder. Because these components play crucial roles in the redox reaction and proton transport, respectively, their distributions can directly affect the electrochemical reactivity, and thus, the device performance. Even though analyzing the component distribution is important to understand its electrochemical reactivity and improve the device performance, it is challenging to determine it for the PEMFC electrode. In this presentation, I will summarize our recent effort for visualizing the spatial distribution of the electrode components and their heterogeneous electrical properties using multidimensional current–voltage (I–V) spectroscopy combined with data mining. The electrical properties of the electrode components, i.e., the Pt/C catalyst and Nafion binder, were explored by I–V spectroscopy, and their electrical heterogeneity was spatially classified based on the shapes of the measured I–V curves by cluster analysis. These results imply that it is possible to get more insight on the data analysis in AFM even using a simple machine learning technique. Furthermore, the proposed method is expected to be applicable for investigating in detail not only the spatial properties of PEMFC electrodes, but also the properties of various material systems.

One of the key advantages many SPM techniques have for cell biology studies is their ability to non-destructively image cells in physiological environments. Using time lapse microscopy, it is then possible to observe how the cells change over time either during growth, development or due to drug treatments. For bacterial cells, time lapse atomic force microscopy (AFM) has been very helpful in understanding how cells grow, divide and react to antibiotic stresses. For soft mammalian cells however, long term time lapse using AFM is more difficult due to the complex and time varying force interaction between the tip and the cell surface. For these cells, scanning ion conductance microscopy (SICM) would be better suited for time lapse experiments. Unfortunately, standard SICM instruments require 10s of minutes for good quality images, which is often too slow to follow important cell dynamics. In this talk I will describe a microscope we have developed for time resolved SICM imaging of mammalian cells that enables imaging durations of up to 48h with image acquisition times as low as 3 seconds. We have combined this SICM instrument with super-resolution optical fluctuation imaging (SOFI) to complement the morphological SICM data with biochemical fluorescence information. With this instrument we study cell growth, motility, macropinoytosis and cell differentiation.

Here, we present a multimodal scanning probe microscopy study of a composite anode with a dispersed lithium silicon titanium phosphate (LSTP) lithium ion conductor for all-solid-state batteries. We used electro-chemical strain microscopy (ESM) and lateral force microscopy (LFM) to analyze the electromechanical response and friction force dependence as a function of the measurement parameters such as the tip loading force and AC drive voltage. We found that the sensitivities of friction force and ESM amplitude could be valid markers to identify each component in the composite anode. In addition, we visualized the distribution of active ionic sites of lithium ions in the region of LSTP and binder materials based on Pearson’s correlation analysis between nanoscale ESM and LFM results. With the suggested technique, various components of composite electrodes can be directly visualized and distinguished in ambient conditions, with their properties being measured simultaneously. These methods will provide insights into the optimal conditions of composite electrodes and allow for developing next-generation all-solid-state batteries.

The dynamic shapes of biologial tissues emerge from a complex interplay of physics, chemistry and genetics, which determines--at each temporal and spatial scale--the mechanical properties that eventually form the structures of living organisms. Shape and mechanical stability of living organisms rely on precise control in time and space of growth, which is achieved by dynamically tuning the mechanical (viscous and elastic) properties of their hierarchically built structures from the nanometer up.
It is now well-established that cellular behaviour (including stem cell differentiation) crucially depends on the mechanical properties of the cells’ environment. Much attention has been directed towards the importance of the stiffness of the natural (extracellular matrix, ECM) or artificial matrices where cells grow, with the purpose of either understanding mechanotransduction, or controlling the behaviour of cells in tissue engineering. While stiffness (i.e. the capacity of a material to elastically store mechanical energy) has been the focus of most experimental research, neither cells nor matrices are elastic. Biological systems dissipate energy (i.e. they are viscous) and hence they do not respond to mechanical deformations instantaneously (like an ideal Hookean spring), but present different time responses at different spatial scales that characterise their responses to external stimuli.
Measuring viscoelasticity (especially at the nanoscale) has remained experimentally challenging [1,2]. In my talk I will present atomic force microscopy (AFM)-based techniques to measure and map the viscoelasticity of living tissues, cells, membranes, collagen, ECMs, and tissue engineering matrices across the spatial and temporal (from Hz to 100s of kHz) scales, and chirp-based spectroscopic techniques to assess viscoelasticity from Hz to 100s kHz at the nano and micro scale developed in my lab. I will also present tests for assessing which viscoelasic model better fits the experimental AFM results.
Our results have uncovered that extracellular matrices of both living plants and tumours present an almost perfect linear viscoelastic behaviour that is key to understand their growth and shape. I will present our work showing how the growth and shape of the roots, leaves and hypocotyl of living Arabidopsis thaliana living plants are related to the nanoscale viscoelasticity of plant cell walls at the time scales probed by multifrequency AFM.

References:
[1] “Multifrequency AFM reveals lipid membrane mechanical properties and the effect of cholesterol in modulating viscoelasticity” 2019. Z Al-Rekabi, S Contera; Proceedings of the National Academy of Sciences 115 (11), 2658-2663.
[2] “Mapping nanomechanical properties of live cells using multi-harmonic atomic force microscopy”2011 A Raman, S Trigueros, A Cartagena, APZ Stevenson, M Susilo, E Nauman, S Contera. Nature Nanotechnology 6 (12), 809.

The ability to characterize and manipulate materials down to the atomic scale for tuning their mechanical and physical properties is the cornerstone of material science. Two-dimensional (2D) materials in particular are getting large attention because their extremely thin size gives rise to unique and groundbreaking properties.
Here we explore two different research topics. On one side we show how heat delivered by a scanning probe can be used for manipulating the electrical properties of 2D materials, and on the other side we present a recent scanning probe microscopy method to perform Å indentation experiments in supported 2D materials.
Among various novel materials, two-dimensional molybdenum disulfide (MoS₂) is of particular interest due to its large band gap, low dielectric constant, and heavy carrier effective mass. Currently, a key issue in creating high performing field-effect transistors (FETs) based on MoS₂ and other transition metal dichalcogenides (TMDC) films is the poor quality of the metal contacts fabricated on these atomic layers, and the difficulty to pattern dopants. Here, we report a strategy based on thermal scanning lithography (tSPL) to fabricate metal contacts on 2D materials with high reproducibility [1], and we demonstrate the integration of tSPL with a flow-through reactive gas cell to achieve nanoscale control of dopants in monolayer MoS₂ [2]. The tSPL produced defects can present either p- or n-type doping on demand, depending on the used gasses, allowing the realization of field effect transistors, and p-n junctions with precise sub-μm spatial control, and a rectification ratio of over 10⁴. Doping and defects formation are elucidated by means of X-Ray photoelectron spectroscopy, scanning transmission electron microscopy, and density functional theory. We find that p-type doping in HCl/H₂O atmosphere is related to the rearrangement of sulfur atoms, and the formation of protruding covalent S-S bonds on the surface. Alternatively, local heating MoS₂ in N₂ produces n-character.
Moreover, we report on the elastic properties of supported 2D graphene and h-BN films by the use of modulated Å-Indentation, a novel scanning probe microscopy (SPM)-based technique capable of achieving sub-Å indentation depths during force-indentation measurements [3, 4]. By using extremely small amplitude oscillations (<<1 Å) at high frequency, we show how Å-Indentation enables non-destructive local accurate measurements of the contact stiffness and out-of-plane elastic moduli of ultra-thin ultra-stiff films, including CVD diamond films (thus even up to TPa range stiffness), as well as the transverse moduli of supported graphene and h-BN (<1 nm thickness) films. Å-Indentation thus obtains in-situ elasticity measurement combining superior resolution and indentation depths as small as 0.3 Å with AFM nanoscale topographical imaging. Thanks to this technique, we experimentally demonstrate that at room temperature and under localized indentation pressure, a single layer of graphene on top of a carbon interface layer (buffer layer), on SiC, exhibits a transverse stiffness superior to bulk diamond [3]. Extending the capabilities of the Å-indentation technique to the analysis of other 2D materials, these studies explore similar pressure-induced phase transition in hexagonal boron nitride (h-BN). We demonstrate that 2-layer h-BN flakes, on SiO₂, consistently exhibit a transverse elastic modulus almost two times larger than that of the bare substrate. This stiffening effect is observed only for flakes of thickness between 2 to 5 atomic layers, but not in single- nor multilayer (> 6-layer) h-BN, and is related to the conversion of h-BN to cubic BN (c-BN) induced by the pressure applied through the indenter.

1 Zheng et al. Nature Electronics, 2, 17–25 (2019)
2 Zheng et al. Nature Communication, (2020) https://doi.org/10.1038/s41467-020-17241-1
3 Gao et al. Nature Nanotechnology 13.2 (2018): 133
4 Cellini et al. Scientific reports 9.1 (2019): 4075

Atomic Force Microscopy (AFM) is one of the most powerful and versatile tools to image and characterize materials with nanometer scale resolution, leading to the development of nano-/bio- science and technology. While AFM technology has advanced significantly by expanding its functionality beyond a nanoscale imaging tool, the majority of AFM users still find it difficult to employ the advanced techniques and interpret the data correctly due to artifacts. Such difficulties mainly originate from the complex dynamics of the AFM probe system, used as a mechanical transducer to deliver the tip-sample interaction force.

We address this issue by altering the AFM cantilever design. The current state of the art AFM probe is a single-body system, in the shape of a rectangular or triangular beam, which is not ideal for carrying more than one type of information. In contrast, the new AFM probe, so-called paddled cantilever, is reshaped to have a two-field microcantilever design with a considerable dimensional discrepancy in the component part. By doing so, this mechanical transducer provides two discrete transduction channels such that they respond independently to the variations in surface topography and material properties/functionality. Thus, AFM users can effortlessly implement the advanced AFM schemes with minimal ambiguity, complications, and artifacts, by simply employing this new AFM probe.

Since the prototype device of paddled cantilever was first introduced, our team has taken effort to develop a batch fabrication process and, thus, have this advanced technology widely available for AFM users. The detailed dynamics of fabricated paddled cantilevers have been investigated by using the interferometric displacement sensor (IDS) with AFM to directly measure their deflection. The measured deflection along the whole cantilever revealed that the frequency of inner-paddle is not affected by the tip stiffness while its amplitude is proportional to the displacement input delivered through the tip. This is the basis how the paddled cantilever can provide two discrete transduction channels for functional AFM. A theoretical model based on a 2 degree-of-freedom system is also presented to understand the dynamic characteristics of paddled cantilever under various operational conditions. Because the design of paddled cantilever is simple to feasibly integrate into, and compatible with, current AFM probes and systems, this technology will easily extend the AFM functionality to characterize multi-physical properties beyond topography by providing a dual-channel mechanical transducer.

Scanning electrochemical probe microscopy (SEPM) techniques such as scanning electrochemical microscopy (SECM) and in particular the hybrid techniques such as atomic force microscopy combined with scanning electrochemical microscopy (AFM-SECM) enable high resolution topographical imaging along with electrochemical information [1] but also electrochemical force spectroscopic measurements [2]. Recently, we introduced a new type of electrochemical scanning probe tip for combined AFM-SECM measurements providing instead of a conventional AFM tip, a conductive colloidal sphere attached to an otherwise insulated cantilever [3]. Such conductive spherical probes may consist of boron-doped diamond [4], a highly attractive electrode material, or may be electrochemically modified, e.g., functional polymers, suitable for force spectroscopic measurements such as adhesion measurements under potential control.
In this contribution, we present AFM-SECM measurements and electrochemical force spectroscopy using conductive colloidal AFM probes. Due to the spherical geometry and the low spring constant of the cantilever, these probes are highly suitable for mapping nanomechanical properties at soft samples, i.e. bacterial cells. Colloidal probes with BDD as spherical electrode can be used for various SPM experiments including conductive AFM given the physical and electrochemical inertness of BDD, for force spectroscopy under potential control providing information on electrified interfaces, and for AFM-SECM measurements. Notably, all these experiments can be performed with the same BDD-AFM-SECM probe. Colloidal probes can be modified with polymers i.e. polydopamine, which properties may be altered by applying potential to the spherical electrode. Single cell force spectroscopy on bacterial adhesion along with studies at electrified interfaces will be presented.

[1] A. Patel, C. Kranz, Annu. Rev. Anal. Chem. 2018, 11, 329.
[2] S. Daboss, J. Lin, M. Goddjehans, C. Kranz, 2020, in press
[3] P. Knittel, H. Zhang, C. Kranz, G. G. Wallace, M. J. Higgins, Nanoscale 2016, 8, 4475.
[4] S. Daboss, P. Knittel, C. E. Nebel, C. Kranz, Small 2019, 15, 1902099.

Imaging ion transport in nanostructure materials has applications in systems from neuromorphic computing, to electrochemical energy storage, and solar cells. We utilize multimodal imaging to probe transport phenomena in these systems. For instance, by combining nanoscale vibrational spectroscopy with structural imaging we can correlated local anion diffusion with structural phase transitions in conjugated polymer films being explored for neuromorphic computing and bioelectronics applications. We also combine similar IR imaging with time-resolved electrostatic force microscopy (trEFM) to uncover the dynamics of ion transport in various 3D and 2D halide perovskites of interest for applications in optoelectronics, especially solar cells, where ion transport can be a primary contributor to device aging and failure. This work bridges advances in scanning probe microscopy with tools from data science to better understand materials design rules for new applications.

The molecular organisation and dynamics of liquids near the surface of solids underpins processes such as molecular exchanges, electrochemistry, nanofluidics, biomolecular function and lubrication. Depending on the local topographical and chemical properties of the solid, the behaviour of the interfacing liquid can vary significantly at the nanoscale, with consequences for interfacial processes. Scanning probe microscopies can derive local insights into the equilibrium organisation of interfacial liquids, often with sub-nanometre resolution. However, deriving quantitative information about the local liquid dynamics remains a considerable challenge.
Here we review some of our recent results obtained by combining high-resolution atomic force microscopy with highly localised rheological measurements to deduce quantitative information about the motion of liquids and dissolved molecules along immersed solids. We show that the dynamics is highly system-dependant and can be dominated by either the local configurational entropy or the interface or enthalpic interactions. We also describe a novel approach to quantify the preferred flow direction naturally adopted by liquids near walls, locally and with nanometre precision. The approach, based on AFM, combines a vertically oscillating tip with high frequency lateral oscillations to derive directional information about the interfacial liquid at each point of the system. To illustrate the method’s capabilities, we investigate the dynamics of aqueous solutions containing different metal ions along the surface of a graphene oxide flake. We identify ion-specific flow dynamics that can be related to the well-established sieving properties of graphene oxide membranes.

Investigating nucleation from solutions is challenging, because it is a consequence of unstable density fluctuations, making the structures and events that must be probed both transient in nature and small in spatial extent. Moreover, when nucleation is heterogeneous, it is inherently linked to the structure and dynamics of the interfacial region between the bulk solution and the underlying substrate. Thus high-speed, molecularly resolved AFM combined with fast force mapping (FFM) provides a powerful and unrivalled combination to investigate nucleation from solution onto surfaces. Here I illustrate the deep level of fundamental insight into mechanisms of nucleation and the structure of the interfacial region made possible with these methods using results from investigations into nucleation of gibbsite (Al(OH)₃) on muscovite mica (001) and the structure of electrolyte solutions above mica and boemite (AlOOH) surfaces. In the case of gibbsite nucleation, we combine AFM observations of individual molecular adsorbates, transient clusters, and stable islands with Monte Carlo simulations to put together a coherent picture of surface speciation and nucleation. The results reveal a surface population of ions that is dominated by hydrolyzed species (Al(OH)₂⁺ and Al(OH)²⁺) even though Al³⁺ vastly dominates the bulk solution. These adsorbed ions evolve into subcritical clusters with increasing saturation state and temperature, constituting a population that decreases exponentially with size and exhibits dynamic fluctuations consistent with classical predictions. However, severe discrepancies with classical theory emerge when the values of key thermodynamic parameters are extracted from the AFM data. These discrepancies are resolved when the impact adsorbate charge on the capacitance of the mica-solution system is taken into account in calculating the work of cluster formation, but reveal a film formation process that occurs within a single phase region. Thus, the results imply that the gibbsite films constitutes a nanostructured 2D mesophase. To define the solution structure above boehmite and mica surfaces at lattice resolution, we developed a self-consistent scheme to decouple long-range tip-sample interactions from short-range solvation forces. In the case of boehmite, the results are benchmarked against molecular dynamics simulations that explicitly include the effects of the tip with different levels of approximation and systematically account for tip size, chemistry, and confinement effects. We find four laterally structured water layers within one nanometer of the surface, with the highest water densities at sites adjacent to hydroxyl groups. The findings reveal a complex relationship between site-specific chemistry, water density, and long-range particle interactions; and represent a major step forward towards quantitative data interpretation and widespread implementation of 3D FFM.