Figure 1

A schematic representation of the capabilities of various probes in terms of length scales, time scales, and complexity of the property and response functions.Reuse & Permissions

Figure 2

General principles of (a) scanning tunneling microscopy and (b) atomic force microscopy. In STM, a metal probe tip is brought into proximity to a conductive sample, which forms a circuit that is used to control the separation and interactions. In AFM, the tip is typically mounted on a cantilever and forces between the tip and surface are monitored using the mechanical properties of the cantilever. The junction of a tip and a planar surface contains information about electronic electromagnetic, optical, thermal, and chemical properties. These properties can be accessed by stimulating the junction with various perturbations and quantifying the response (c). [(c) is courtesy of Stephen Jesse].Reuse & Permissions

Figure 3

(a) A scanning conductance image of defects in a ${\mathrm{HfO}}_2$ thin film on a Si substrate, demonstrating subnanometer spatial resolution of enhanced current ($\sim 5\mathrm{pA}$) at defects and (b) a mechanism map relating experimental tip-contact conditions to strain-induced conduction variations in Si. From 243.Reuse & Permissions

Figure 4

Nano-impedance spectroscopy. Schematic diagrams of NIM configurations. The tip acts as an electrode in a (a) vertical or (b) lateral configuration. (c) A simple equivalent circuit for the tip-surface junction. Two examples of local impedance spectra, (d) one with multiple interfaces in the current path and (e) one with a single interface. (a)–(c) From 303; (d), (e) from 252.Reuse & Permissions

Figure 5

Complex properties from modulated tip bias techniques. (a) Scanning impedance image of a single nanotube with defects in which contrast is related to the valence band energy of each individual defect. From 89. (b) Comparison of bridge-enhanced nano-impedance microscopy amplitude and phase images (left) with uncorrected NIM (right) demonstrating five orders of magnitude increase in sensitivity. From 266. (c) Topographic image (top) and single-frequency capacitance (bottom) of single and double purple membrane layers. From 93.Reuse & Permissions

Figure 6

Scanning surface potential and Kelvin force microscopy. (a)–(c) Potential measurement of single molecular layers; (d), (e) Potential variation with charge generation. (a) nc-AFM images of the structure of assembled monolayers of a Zn porphyrin on graphite. (b) Comparison of the topographic structure and (c) the surface potential. (d) Comparison of surface potential in the absence and (e) presence of optical illumination of a polymer blend. (a)–(d) From 242. (e), (f) From 52.Reuse & Permissions

Figure 7

(a) Noncontact AFM topography image, area $10\times 8{\mathrm{nm}}^2$, $\Delta f\cong 121\mathrm{Hz}$, $A_{p-p}\cong 28\mathrm{nm}$, $f_0=70\mathrm{kH}z$, $f_{\mathrm{mod}}\cong 452\mathrm{kH}z$, and $U_{\mathrm{ac}}=0.3\mathrm{V}$. (b) Simultaneously recorded potential. (c) Average twin cross sections of topography and potential, as indicated by the boxes in (a) and (b). (d) Schematic snapshot sequences of the AFM tip tracing in (a)–(c): left to right, and (d)–(f): right to left directions. Cutouts of the experimentally recorded cross-section graphs are superimposed onto the models in (d)–(f) and (a)–(c), respectively. Double arrows indicate the key electrostatic tip-surface interactions. From 79.Reuse & Permissions

Figure 8

High-order harmonic detection in $s$-NSOM to determine the dielectric function is illustrated with scattering of $s$-polarized light at four frequencies from a graphite-porphyrin sample. Profiles (top) across a molecular monolayer island are compared. The calculation of image contrast dependence on the dielectric function at the third-order harmonic (middle) can be used to relate scattering variations (right images) to properties. From 242.Reuse & Permissions

Figure 9

A diagram of a scanning nonlinear dielectric microscope illustrating the inclusion of a resonator at the tip-surface junction. From 325.Reuse & Permissions

Figure 10

In nc-AFM, atomic resolution is achieved both in (a) the topography image, which is the $\epsilon (4)$ feedback signal and in (b) the nonlinear dielectric constant image $\epsilon (3)$. The familiar Si(111)-($7\times 7$) surface is used to demonstrate the resolution. There is a direct correlation between the Si adatom locations and the contrast in the $\epsilon (3)$ signal. This implies that the atomic polarization of the Si bonds is probed here. From 325.Reuse & Permissions

Figure 11

(a) A schematic diagram of patterned peptides on a graphite substrate with light incident on the tip-surface junction. (b) The effect of optical illumination on the capacitance of the monolayer porphyrin containing peptides is illustrated as 2D and 3D histograms. The top 3D histogram in the inset shows the properties with illumination, the bottom without illumination. In the 2D histogram the curve on the right is with illumination, and the curve on the left is without exposure. $\lambda =425\mathrm{nm}$. From 172.Reuse & Permissions

Figure 12

Charge dynamics in a polymer blend solar cell. (a) The sample is illuminated from the bottom through a transparent substrate. (b) The EFM signal is monitored after the light is switched off, and (e) the exponential decay constant is plotted for each pixel. (c) This is compared with the quantum efficiency. (d, e) Images of topography and the charging rate, respectively. From 56.Reuse & Permissions

Figure 13

Left: Photograph and schematic of a STM scan head inside a reactor cell. From 39. Right: STM images of Pt(110) surfaces inside the reactor with gas environments of 1.6 atm of ${\mathrm{H}}_2$ and 1 atm of ${\mathrm{O}}_2$ and CO. The crystal surface reconstructs into missing rows, facets, and step bunching, respectively. From 221.Reuse & Permissions

Figure 14

(a) Photograph of a ${\mathrm{SnO}}_2$ single crystal. The (101) surface displayed in (b) is of particular interest because it can be reversibly produced in fully oxidized and reduced form, making this an ideal model system to study processes relevant in chemical sensing. (c) SEM image of ${\mathrm{SnO}}_2$ “nanobelts.” The wide sides of the nanobelts have (101) orientations. (d) STM images of ${\mathrm{SnO}}_2$ nanobelts with different magnifications. From 23.Reuse & Permissions

Figure 15

(a) Schematic of a ${\mathrm{C}}_2{\mathrm{H}}_2$ molecule adsorbed on Pd(111) and associated STM image. Shown below them is a calculated image showing the modulation of the tunneling current by the orbital structure of the molecule. Tunneling electrons can cause the molecule to rotate in its three fold hollow site. (b) Inelastic electron tunneling spectra obtained by 304 on ${\mathrm{C}}_2{\mathrm{H}}_2$ molecules on Cu(100). Tunneling probability changes due to excitation of CH stretch vibrational modes are visible.Reuse & Permissions

Figure 16

The rutile ${\mathrm{TiO}}_2(110)$ surface has become one of the most studied systems in oxide surface science, and STM measurements have been instrumental in understanding this system. Oxygen vacancies (point defects) are easily created through sputtering and annealing. These contribute to a defect state in the band gap, visible in photoemission. In empty-state STM images, oxygen vacancies appear as protrusions between the rows of (protruding) ${\mathrm{Ti}}_{5\mathrm{c}}$ atoms; surface hydroxyl (OH) moieties are somewhat more protruding. Adapted from 66.Reuse & Permissions

Figure 17

Water adsorption on the nonpolar ZnO(1010) surface. In the lowest-energy state, every other molecule is dissociated. Such a configuration results in a saturated overlayer with ($2\times 1$) periodicity. A ($1\times 1$) overlayer, where every water molecule is still intact, is very close in energy. (a) Atomic models for both structures. (b) The STM image shows rows of water molecules that exhibit either ($2\times 1$), ($1\times 1$), or an “intermediate” (IM) configuration; (c) line profiles for each of these cases. (d) The estimated transition path between these two structures. Note that the activation energy is rather small (0.05 eV). At room temperature, water switches back and forth between these two states; if the switching occurs more rapidly than the sampling time of the STM measurement, an apparent IM structure is observed. From 70.Reuse & Permissions

Figure 18

Fourier transform of cuprate LDOS modulations exhibiting signature Bogoliubov excitations from $d$-wave Cooper pairs. (a) These excitations are detected only in $k$ space in the region shaded white in (b). Beyond the $k$-space line connecting $p$, 0 to 0, $p$, the states are not the excitations of $d$-wave Cooper pairs but localized in $r$ space as in (c). From 184.Reuse & Permissions

Figure 19

Scanning tunneling microscopy data on Fe-doped topological ${\mathrm{Bi}}_2{\mathrm{Te}}_3$ (a), (b) $dI/dV(r,E)$ maps showing interference patterns, with the inset showing the corresponding Fourier transform. (c), (d) Same data as insets in (a), (b) but divided by the Fourier transform at $+60\mathrm{meV}$ to remove nondispersing features at small $\mathbf{q}$ vectors. The $\mathbf{q}$ vector along $LK$ was not reported in previous samples without time-reversal symmetry breaking impurities. (e) Schematic diagram of constant energy contours showing the nested regions that result in the primary $q$ vectors observed in STM at low energies along $LK$ and higher energies along $LM$. From 253.Reuse & Permissions

Figure 20

(a) Schematic representation of spin-polarized STM. Spin-polarized tunneling results in a tip-sample distance that depends on the relative magnetization direction of tip and sample. (b) Topography and (c) magnetic $dI/dU$ map of a Cr(001) surface measured with an Fe-coated probe tip at 300 K. Screw (arrows) and edge dislocations (one example marked by an arrow) can be recognized in (a). The tunneling parameters are $I=3\mathrm{nA}$ and $U=-0.7\mathrm{V}$. As for other monatomic step edges the edge dislocations lead to a change of the magnetic contrast due to the layered antiferromagnetic structure of Cr(001). Typically, a domain wall is found between two screw dislocations, e.g., between the two arrows in (b). From 38.Reuse & Permissions

Figure 21

(a) Topography of Mn islands on W(110) measured with an Fe-coated tip. Lines along the [001] direction indicate row-wise antiferromagnetic order. (b) A higher-magnification image reveals that the corrugation is not constant but modulated. (c) Averaged line section (dots) taken along the box in (b). (d) Schematic representation of the antiferromagnetic cycloidal spin spiral. From 38.Reuse & Permissions

Figure 22

Measurement of spin relaxation time. (a) STM topographic image of a Fe-Cu heterodimer and a Cu adatom. (b) Voltage pump-probe measurements of the dimer and two control structures: the dimer probed with a non-spin-polarized tip; a nonmagnetic atom probed with a spin-polarized tip. In region I, the probe precedes the pump; in region II, the pump and probe overlap; in region III, the probe follows the pump and any postexcitation is detected. The Fe-Cu dimer exhibits a relaxation time of $87\pm 1\mathrm{ns}$. From 206.Reuse & Permissions

Figure 23

(a) Phonon-mediated tunneling in graphene. From 359. (b) Magnetic-field dependence of spin excitations in a single magnetic Mn atom and a cluster of three atoms. (c) The density of states showing the gap state in the inset. (d) Temperature dependence of conductance spectra for a single Mn atom and a Mn trimer. From 133.Reuse & Permissions

Figure 24

Distribution of mode energies detected by $d^{2}I/d{V}^2$ imaging of Bi-2212 crystals containing only ${}^{16}\mathrm{O}$ and ${}^{18}\mathrm{O}$. (a) 2D histograms of mode energies plotted vs the associated energy gap values $D$ for Bi-2212 crystals containing ${}^{16}\mathrm{O}$ and ${}^{18}\mathrm{O}$. (b) The mode detected here as part of the superconducting electronic structure involves vibrations of the oxygen atoms. Distribution of mode energies detected by $d^{2}I/d{V}^2$ imaging. From 199.Reuse & Permissions

Figure 25

Single-atom chemical identification by nc-AFM. The schematic illustrates the operation mode, with the tip oscillating at its resonance frequency, modified by interaction forces with the substrate atoms. The image at the bottom right is the atom-resolved topography of a surface alloy composed of Si, Sn, and Pb atoms blended in equal proportions on a Si(111) substrate. From 313.Reuse & Permissions

Figure 26

$1750\times 810{\mathrm{pm}}^2$ 3D force microscopy on graphite, illustrating some of the various information channels that can be recorded simultaneously (3, 4; 24). (a) $x\mathrm{\text{−}}y$ cut through the 3D force data at constant height. The average attractive force in this image is $-2.306\mathrm{nN}$ with a total force corrugation of about 70 pN. (b) $x\mathrm{\text{−}}y$ potential energy data from the same height over the surface as in (a). The average energy is $-5.47\mathrm{eV}$ with a total corrugation of $\sim 38\mathrm{meV}$. (c) 2D cross section along the line plotted in (a), showing how the sample’s surface force field extends into the vacuum (height 220 pm). The average force at each height has been subtracted to enhance contrast. Contour lines mark force differences of 8 pN at every height. (d) Map of the energy dissipated during oscillation. The color scale spreads from 250 meV (dark) to 295 meV (bright). (e) Map of the absolute values of the lateral forces acting on the tip during closest approach. Lateral forces are between 0 and 120 pN. (f) Zoom into the area highlighted in (e), where the direction and strength of the lateral forces are indicated by vectors. Note that in (b) and (e), the positions of the carbon atoms have been marked with an exemplary hexagonal ring. From 4.Reuse & Permissions

Figure 27

Trade-off between chemical specificity and spatial resolution. The combination of chemically specific techniques (NMR, IR, Raman) with high-resolution techniques (STM, AFM) holds promise for measurements providing high-information-resolution products. Adapted from 57.Reuse & Permissions

Figure 28

Comparison of diffraction-limited optical microscopy and near-field optical microscopy. (a) Schematic of the diffraction limit showing the minimum detectable separation of two scatterers. (b) Schematic of aperture scanning near-field microscopy. To a first approximation, resolution is defined by the aperture size and not by the wavelength of the exciting radiation.Reuse & Permissions

Figure 29

Near-field Raman imaging of single-walled carbon nanotubes. (a) Topography showing individual catalyst particles and a hidden carbon nanotube. (b) Near-field Raman scattering spectrum of a carbon nanotube. Near-field Raman images of the (c) $G$-band intensity, (d) ring breathing mode (RBM) intensity, and (e) $D$-band intensity. All images are acquired simultaneously. The RBM frequency of $259{\mathrm{cm}}^{-1}$ identifies the tube as an (8,5) nanotube with a diameter of 0.9 nm. The $D$-band intensity is related to defects in the carbon nanotube structure. (e) The nanotube is locally perturbed by a larger catalyst particle.Reuse & Permissions

Figure 30

Schematics (not to scale) of the scattering scanning near-field infrared nanoscope. The basic building block of the proposed apparatus is an AFM. The tip of the AFM is illuminated with an infrared laser. The scattered signal is registered using an interferometric scheme enabling direct measurements of both amplitude and phase of the scattered light. Interferometric detection is imperative to produce images of local values of optical constants free of topographic artifacts.Reuse & Permissions

Figure 31

Infrared snapshot images taken during the critical temperature range of the Mott transition of ${\mathrm{VO}}_2$. The transition from insulator to metal proceeds by temperature-induced growth and coalescence of initially separate metal puddles. From 270.Reuse & Permissions

Figure 32

(a) Schematics of piezoresponse force microscopy (PFM) measurements. Electric bias applied to an SPM tip in contact with the surface creates an electric field within material (b) that generates strain due to piezoelectric or more complex electromechanical coupling. The nonuniform strain results in surface deformation detected as the SPM tip deflection. (c) Electromechanical coupling in functional materials. The solid lines correspond to estimated limits of nanoindentation and atomic force microscopy, while the dashed lines correspond to the ultimate limits that can be achieved through the instrumentation development including resonant-enhanced modes, low-noise beam deflection position systems, and high-stability platforms. From 169.Reuse & Permissions

Figure 33

(a) The confinement of an electric field by an AFM probe enables probing a bias-induced phase transition within a defect-free volume or at a given separation from defects. From 156. The color map (b) illustrates the disorder potential in an epitaxial lead zirconate titanate film. (c), (d) A single defect in multiferroic ${\mathrm{BiFeO}}_3$ determined from the nucleation bias and fine structure features on a (e), (f) hysteresis loop. From 170.Reuse & Permissions

Figure 34

(a) Domain wall conductance in ${\mathrm{BiFeO}}_3$ thin films. (b) Tip-induced metal-insulator transition in Ca-doped ${\mathrm{BiFeO}}_3$. (c) Simultaneous measurements of bias-induced strain and tip-surface current, illustrating the perfect correlation between polarization switching and transport. The double peak is due to the presence of defect and is observed in $\sim 10\%$ of cases. (d) The combination of high-voltage writing and low-voltage readout enables nonvolatile data storage. From 353 and 216.Reuse & Permissions

Figure 35

Data acquisition and processing. (a) Ferroelectric switching is induced by a pulse train of increasing dc bias while the changes in the piezoresponse, contact stiffness, and dissipation are measured by exciting the cantilever over a narrow frequency band around its contact resonance (b),(c). The ferroelectric hysteresis is measured across a grid of points (d) resulting in a spatially resolved 4D data set, where each point represents the cantilever’s resonant response along the local hysteresis loop (e),(f). The resonant response (amplitude and phase) is then by a simple harmonic oscillator model to yield the resonance amplitude, resonance frequency, and $Q$ factor (g) as a function of the dc bias. From 153.Reuse & Permissions