Keysight Technologies Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization.

Transcription

1 Keysight Technologies Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization Application Note

2 Introduction The scanning microwave microscope (SMM) merges the nanoscale imaging capabilities of an atomic force microscope (AFM) with the high-frequency broadband (from MHz to GHz) impedance measurement accuracy of a vector network analyzer (VNA) (see Figure 1). The typical frequency range of the combined SMM is between 1-20 GHz [1]. It allows characterizing electric and magnetic properties of materials at microwave frequencies with nanometer lateral resolution. Using the microwave signal, impedance nanoscale imaging and doping profiling of the device under test (DUT) can be performed. Typically, the SMM is operated in reflection mode, whereby the ratio of the reflected and incident electromagnetic waves, the so called S 11 scattering parameter, is measured by the VNA at each pixel of the AFM tip-sample contact point. As such a microwave image is generated pixel by pixel, simultaneously to the topographical image of the DUT. Imaging speeds are relatively fast, resulting in a typical acquisition time of 2 minute per image with 256 x 256 pixels. There are two main different imaging modes in SMM. The first is quantitative dopant profiling by means of dc/dv [2], which is a widely used technique for semiconductor failure analysis and detecting leakages in solid state devices with nanometers resolution. The dc/dv mode relies on a low frequency (khz) modulation of the GHz S 11 signal that allows tuning the semiconductor depletion zone and probing the doping concentration through the native oxide interface. The dc/dv signal is not affected by the so called topographic cross-talk [3], therefore it gives immediate information that only depends on the material properties and is not influenced by topographical features of the sample. A dopant calibration sample is required for quantitative imaging. The second SMM mode is complex impedance imaging and it is based directly on the scattering S 11 signal [3]. The S 11 signal is measured by the VNA and depends on the tip-sample relative electrical impedance and on contributions coming from the transmission line cables and RF connectors. Therefore, a de-embedding workflow is required that allows converting the raw S 11 into a calibrated S 11 that depends only on the complex impedance between tip and sample. In this way, using the calibration workflow, the SMM user can obtain calibrated capacitance and resistance images of the DUT. No calibration sample is required because the calibration is done on the DUT itself. dc/dv phase Topography dc/dv amplitude dc/dv phase Figure 1. (a) Sketch of the SMM experimental setup and of the new doping profiling calibration sample with flat topography. The sample consists of 10 different n-implant areas (left region) and 10 different p-implant areas (right region), each area with a width of 2μm, with doping concentrations ranging from to atoms per cm 3. The SMM raw images of this sample are shown including flat topography (b), dc/dv amplitude (c) and dc/dv phase (d).

3 03 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note SMM has been extensively used to gain new insights for semiconductor devices, which is one of the fields where SMM has been developed the most. In the following semicon-smm application note we present the two SMM modes of dopant profiling dc/dv including calibration sample and workflow, as well as the complex impedance S 11 imaging including calibrated capacitance and conductivity. Furthermore we present augmented SMM solutions where the standard SMM mode is extended with other electronic measurement devices from Keysight that opens new and advanced SMM measurement capabilities. The last part shows how 3D EMPro modeling can be used to calculate E-fields and complex impedance values in order to assist SMM data interpretation. SMM Workflows for Semiconductor Applications Quantitative Dopant Profiling Calibration For quantitative dopant profiling with dc/dv a well-established calibration workflow is used that allows extracting the doping concentration of a semiconductor sample using a dopant calibration sample (Figure 2). The method, described in details in [2], converts the dc/dv signal coming from the Dopant Profile Measurement Module (DPMM) to doping concentration values. First, a calibration sample is scanned with SMM, acquiring the amplitude dc/dv image from which the dopant calibration curve is generated (Figure 2, middle panel). This is done for both n-type and p-type polarity. In a second step, the dc/dv amplitude image of the DUT is acquired under the same SMM working conditions used in the first step for the calibration sample (i.e. PNA frequency, % drive voltage, DPMM KHz frequency). Based on the dopant calibration plot (Figure 2, lower panel), the dc/dv amplitude image of the DUT can be transferred in quantitative dopant concentrations. A flat silicon dopant calibration sample (shown in Figure 1), that has both n-type and p-type polarity was recently developed by Infineon Munich [4]. It consists of a p-si substrate (10 15 atoms per cm 3 ) with 10 different p-type and 10 different n-type implant areas, with doping levels in the range atoms per cm 3. The doped areas in the active region are 2μm wide. The n-doped and p-doped regions are next to each other and therefore can be imaged in one SMM scan. This new doping profiling calibration sample can be imaged in a standard top-down configuration, thus eliminating the difficulties associated to having to scan the sample in cross-section [5]. Figure 2. SMM topography (upper panel) and dc/dv images (center panel) of n-type (left; 30 x 30 μm image size) and p-type (right; 10 x10 μm image size) doped semiconductors acquired at 19GHz and modulation frequency of 14kHz. DC tip bias was +1 and -1 V for n- and p-type, respectively. The topo is flat on purpose so no topo-cross talk exists. In accordance with the specification sheets four and five distinguished steps can be obtained for the n- doped and p-doped samples, respectively. The lower panel shows dopant calibration curves showing the dc/dv amplitude with respect to the dopant concentration for the n-doped (left) and p-doped (right) calibration sample.

4 04 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note Application to Semiconductor Failure Analysis Defects or leakages on real devices can be detected both with dc/dv and S 11 complex impedance imaging. Figure 3 shows a conventional bipolar p-n SRAM sample investigated with SMM, including topographical image and dc/dv images. Several p-n defect structures within the n-doped channels can be identified in the dopant profiling image (cf. arrow in Figure 3b). The sketch depicts p- and n-doped areas in Figure 3d, showing a clear agreement with the various doping types and concentrations in the dc/dv image. The various regions of interest can be clearly distinguished and the doping polarity and density analyzed. No cross talk is obtained between topography and dopant image. Bipolar p/n junction interfaces are observed with a width of 100nm, with the exact width depending on the tip-bias voltage. Figure 3. Imaging defect structures and p-n junctions (a) Topographical image with 10 x10μm scan size with sketch of the doping polarities and concentrations in (d). (b) Corresponding dc/dv phase image with defect structures in the n-doped channels (cf., arrow). (c) dc/dv image with a scan area of 2 x 2μm provides a closer look at the n-doped channel and p-n junction interfaces (cf., arrows). Sub-surface and Backwafer Imaging SMM can use the penetration capabilities of microwaves to perform non-destructive in-depth imaging of structures located underneath the sample surface [6]. This capability, named as sub-surface imaging, is an important benefit when the area of interest is buried under the sample surface. Figure 4 shows SMM images of Si test samples with varying dopant density ( atoms cm 3 ) covered with dielectric layers of SiO 2 ( nm thickness). SMM, allows sensing the electrical properties of doped silicon structures buried under oxides with more than 400nm thickness. SMM can be therefore used to accurately perform quantitative nanoscale probing of dielectric materials and to perform calibrated capacitance characterization of buried structures. This uncovers applications in the field of semiconductor failure analysis studies including process Figure 4. (a) Experimental setup for sub-surface imaging of a doped Si substrate partly covered with SiO 2. (b), (d) SMM images of a doped Si substrate partly covered with 120nm and 200nm. Panels (b),(d) show the topography images with a sketch of the sample (inset) and panels (c),(e) show the reflection S 11 phase images with the subsurface dopant features.

5 05 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note optimizations for integrated chip fabrication. The concept of sub-surface imaging can be pushed further to realize the so called backside wafer imaging, as recently shown on a Static Random Access Memory (SRAM) device [6] (Figure 5). In this application, the SMM tip is scanned over the back-side of the wafer, with the wafer polished down mechanically to a thickness of nm. The active device region is thus beneath nm of doped Si wafer. The SRAM structure, which is on the opposite side of the tip sample contact, could be imaged through the wafer with a thickness of up to 500 nm. a) b) Topography dc/dv b Si wafer d c 50 nm 50 nm 50 nm SRAM SRAM Figure 5. Back-wafer imaging applied to semiconductor wafer devices. (a) Semiconductor device structures (SRAM) imaged at the backside of the silicon wafer through a thicknesses of 50nm. (b) The left panel shows the topographical image and the right panel shows the differential capacitance images (dc/ dv amplitude). Complex Impedance: Calibrated Capacitance and Resistivity Complex impedance calibration An SMM S 11 complex impedance calibration workflow has been recently developed where no calibration sample is required [3]. From the measured S 11, calibrated capacitance and resistance images of a metallic, insulating or semiconducting material are obtained. The calibration procedure is based on the simultaneous acquisition of Electrostatic Force Microscopy (EFM) and S 11 approach curves in order to calculate the three error coefficients in a standard S 11 black box calibration. A low frequency EFM approach curve is used to measure the capacitance change with the distance when the cantilever approaches the sample surface. The method is available as a script in Pico Café [7]. Figure 6 shows the calibration script graphical user interface, which allows selecting the most appropriate equivalent electrical network model. SMM calibration script Figure 6. Graphical User Interface (GUI) of the Picoscript for the automated complex impedance calibration with SMM. The upped panel shows the GUI window that allows importing the EFM/SMM approach curve to calibrate the image. The lower panel shows an example of calibrated Capacitance (ff) and Conductance (μs) images obtained using the script. The script allows the user to select the most appropriate equivalent electrical RC model (red circles) for the study performed. Calibrated image shown in the script GUI SMM calibration script

6 06 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note Calibrated capacitance and resistivity After complex impedance calibration, the data are converted into capacitance and resistance. Using a tip sample analytical model that includes tip radius, microwave penetration skin depth, and semiconductor depletion layer width, the resistivity of the DUT can be extracted from the calibrated SMM resistance [5]. The method has been tested on two doped silicon samples and in both cases the resistivity and doping concentration are in quantitative agreement with the data-sheet values over a range of 10 3 Ω cm to 10 1 Ω cm, and atoms per cm 3 to atoms per cm 3, respectively (Figure 7). The method does not require sample treatment like cleavage and cross-sectioning, and high contact imaging forces are not necessary (as it is the case in SSRM), thus it is easily applicable to various semiconductor and materials science investigations. In Figure 7, the calibrated resistance and capacitance images show lower and higher values for the highly doped areas, respectively. Using the analytical method described in [5], the SMM resistance image can be converted into a resistivity and dopant density image. Figure 7f compares the SMM resistivity and doping concentration with the data-sheet values of the sample. A quantitative agreement is obtained for the entire range of doping concentrations. The lateral resolution of SMM is limited by the probe apex radius while the bulk penetration is mainly determined by the skin-depth. The skin depth and therefore the microwave penetration depth can be changed by varying the SMM frequency. In this way, frequency dependent depth profiling can be realized. Figure 7. SMM raw data (a and b) and calibrated impedance (c and d) of a n-type doped Si sample. The sample has a flat topography (inset in a) and different n-doped areas with doping concentrations ranging from to atoms per cm 3. The 5μm wide doping areas are separated by 1μm wide bulk interface layers as observed also in the capacitance and resistance images. The resistivity and doping concentration (e) are calculated from the SMM resistance based on a tip/sample analytical model [5]. (f) Comparison of SMM resistivity and doping concentration values to the datasheet values (provided by IMEC) determined with Secondary Ion Mass Spectroscopy. Nanoscale Imaging and C-V Spectroscopy The nanoscale lateral resolution of SMM, together with the capability to change the DC tip bias, allows imaging of doped regions, interfaces, and junctions that often play a critical role in several semiconductor devices [8-11]. Figure 8 shows a bi-polar doped silicon sample imaged with SMM to obtain calibrated capacitance images at different tip DC-voltage bias [8]. Figure 8a shows the calibrated capacitance image of the differently doped n-type and p-type regions acquired with no tip bias. The capacitance depends on the doping density and reaches up to 300aF, with high capacitance values observed on highly doped silicon regions. The same region has been imaged by sweeping the DC tip bias between -2V and +2V with steps of 0.3V. For both the

7 07 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note n-type and p-type stripes with an impurity concentration of 4x10 18 atoms/cm 3 and 4x10 19 atoms/cm 3, respectively, the capacitance has been extracted at different tip bias and plotted in Figure 8b. In this way, the C-V curve is obtained directly from the calibrated capacitance images. The conductive SMM platinum tip and the thin native silicon oxide layer (~1 nm thickness) with the doped silicon substrate below form a Metal-Oxide-Semiconductor (MOS) structure. In this configuration, for n-type doped silicon regions, an increase in capacitance at high positive voltages is measured, which is in line with the standard depletion zone model. The C V curve for the p-type stripes shows high capacitance values at negative tip bias and low capacitance at positive tip bias which follows also the standard textbook model [9]. A second region of interest on the same silicon sample includes multiple p-n junction structures with doping concentration values ranging between 4x10 14 atoms/cm 3 and atoms/cm 3. The topography of the area is flat, and the structures are only visible in the electrical images (Figure 8c). Also in this case, the same area has been imaged multiple times in different DC tip bias conditions. The complex impedance calibration workflow has been applied to obtain, from S 11 amplitude and phase images, calibrated capacitance images of the p-n junction structures (Figure 8c). From the capacitance images the different polarity of the n-type and p-type region can be determined during imaging. When the tip bias is set to -1V, the n-type region shows low capacitance values and the p-type region lights up with a higher capacitance. The exact opposite is true for a tip bias of +1V. By holding the SMM tip in contact with the sample surface in a given position and sweeping the tip DC bias, a pointwise C-V curve can be acquired. In both the n-type and p-type regions, two spots (shown with an x in Figure 8c) with different doping concentrations have been chosen and the pointwise C-V experiment has been conducted (Figure 8d). Again the same type of C-V behavior is obtained for the p-type and n-type doping as in Figure 8b. These results show that the C-V curves can be either determined pointwise at particular positions or also from the entire capacitance image acquired at different DC tip-bias voltages. Figure 8. Nanoscale SMM imaging and capacitance-voltage (C-V) spectroscopy experiments. (a) Calibrated capacitance and topography (inset), of a bipolar doped Si staircase sample with n- and p-implanted areas (produced by Infineon Technologies). (b) A DC tip bias voltage has been applied during the scan in a range from -2 V to 2 V, in steps of 0.5 V. C-V curves generated from vertical cross-sections of the capacitance image show the capability of the SMM to detect the impurity polarity directly from the capacitance image. (c) S 11 amplitude with topography (inset), and zoomed-in calibrated capacitance of a flat p-n junction structure located on a different area of the Infineon doped silicon sample. The three capacitance images have been acquired at -1V, 0V, and +1V tip bias, respectively. The capacitance of the n-doped (p-doped) region increases (decreases) when the DC tip bias voltage is increased from -1V to +1V. (d) Pointwise C-V spectroscopy curves in the four different spots labeled with crosses in (c). The curves have been acquired by keeping the tip on the cross labels shown in (c) and by sweeping the tip bias from -1V to +2 V in steps of 0.1V.

8 08 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note Augmented SMM: SMM Combined with Other Keysight Products Keysight s broad test and measurement products portfolio can be leveraged to augment SMM with other high precision instruments, leading to advanced nanoscale measurement solutions unique in the industry. SMM + Source Meter Unit (SMU) for Voltage/Current Spectroscopy SMM can be interfaced with the Keysight s high precision Source Meter Unit (SMU) to simultaneously and accurately measure the DC and RF impedance characteristics of the DUT. SMM allows measuring the calibrated capacitance of the DUT at 1-20GHz as a function of the tip bias voltage. In this augmented SMM solution, the SMU (Keysight B2900) ensures well defined bias conditions, offering both an accurate DC source as well as precise measurement channels for voltage and current (with a resolution of S11-plane, 18 GHz Increase capacitance VNA calibration plane Silicon S11,a Im (S11) Gold pad Black box calibration plane Silicon Metallic sample holder Re(S11) 100nV and 100 fa, respectively). Figure 9 shows micrometric metal-oxide-semiconductor (MOS) capacitors and Schottky diodes characterized by combining the SMM with the SMU [12]. In this experimental configuration, by controlling the biasing conditions, the SMM tip is used to bias the Schottky contacts between reverse and forward mode. In reverse bias direction, the Schottky contacts show mostly a change in the imaginary part of the admittance while in forward bias direction the change is mostly in the real part of the admittance. By leveraging the high precision of the SMU and the nanoscale capabilities of the SMM, calibrated capacitance and 1/C 2 spectroscopy curves can be acquired on the DUT as a function of the DC bias voltage. Furthermore, calibrated capacitance images of both the MOS and Schottky contacts can be acquired with nanoscale resolution at different tip-bias voltages. SMM + Modular Function Generator for C-V Spectroscopy For capacitance voltage measurements SMM can be connected with the Keysight s U2761A USB Modular Function/Arbitrary Waveform Generator [13]. The generated ramp signal is directly sent to the SMM-tip via a BNC divider and the head electronic box (HEB). The same signal is also recorded in the SMM and displayed in the voltage channel (CSAFM/Aux BNC channel). By combining the PNA amplitude signal with the tip-bias voltage data, capacitance voltage curves can be generated. Depending on the sawtooth frequency and scan speed typically C V curves can be constructed from one single scan line. Therefore, depending on the number of lines, thousands Figure 9. Experimental SMM/SMU setup and equivalent circuitry for the Schottky contact. (a) The reflection coefficient S 11,m of the AFM tip connected to the coaxial resonator cable which depends on the tip-sample impedance is measured by the VNA. The bias voltage is generated by a Source Meter Unit (SMU). (b) SMM reflection coefficient S 11 measurements at 18GHz shown in the complex plane. The S 11 -voltage trajectories of four MOS reference capacitors and four Schottky contacts are shown. The bias voltage is swept from -6 V to +6 V. The Schottky contacts (#1 1μm diameter; #2 2μm diameter; #3 3μm diameter; #4 4 μm diameter) show a specific 90 change in their trajectories at zero bias voltage when the impedance changes from dominant capacitive to dominant resistive. The MOS contacts #1 (200 nm oxide), #2 (150 nm oxide), and #3 (100 nm oxide) stay constant during the bias sweep and only #4 (50 nm oxide) shows some deviation from ideal behavior at larger negative biases. For the capacitance calibration procedure only the zero bias capacitance of the MOS contacts are used (marked with circles).

9 09 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note of individual C V curves can be acquired from a single scan. Assuming 512 lines per image and 80 C V curves per line roughly 40,000 C V curves can be constructed from one scan. dc/dv is the preferred method for dopant profiling of semiconductors, while scanning sawtooth capacitance spectroscopy is suggested to be the preferred method for acquiring C V curves of materials to study their physical properties including trapped charges in oxides, metastable interface states and hysteresis effects. The system is calibrated in situ and no modeling is required to construct the C V curves [13]. Nanoscale Modeling with EMPro for Data Interpretation The experimental SMM results can be compared to quantitative modelling using two software packages from Keysight, ADS and EMPro (Figure 10). With respect to ADS, the experimental SMM frequency response function between 1-20GHz can be modelled using the electronic circuit simulation package [14]. ADS is an electric circuit simulation tool that provides optimisation, tuning, network parameters, and broadband frequency response for standard electronic components. ADS was used to model the broadband frequency S 11 response curve of the coaxial cables and the impedance matching network, as well as the effect of the AFM tip in contact with different dopant samples. Figure 10 shows the Smith Chart analysis of the ADS simulation results. The Smith Chart is used typically for high frequency network analysis, in which both the scattering parameter S 11 and the impedance Z, respectively the admittance Y, are plotted. The variation of S 11 is shown at two different frequencies, 2GHz and 18GHz. The typical experimental values obtained with SMM are close to Y = 0, with tip/sample capacitance values as low as 1aF. The values of the tip-sample admittance acquired with the SMM have been compared to the ones obtained from ADS simulations and a quantitative agreement has been achieved [14]. Figure 10. ADS circuitry diagram and Smith Chart analysis. The SMM high frequency parts are modelled including VNA port as source and sensor, coaxial cable, shunt resistor, resonator, and the tip-sample system. The tip-sample interaction is modelled using EMPro (red box) and the results are directly transferred into the ADS model. The Smith Chart presents a study of the range of values obtained in the admittance domain at two different frequencies, 2.2GHz and 18GHz.

10 10 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note Furthermore, calibrated SMM impedance images can be compared to 3D microwave finite element modelling (FEM) using the EMPro software package for proper data interpretation [15, 16]. 3D electromagnetic full-wave simulations using EMPro can validate the calibrated complex impedance SMM measurements (Figure 11). EMPro supports two simulation engines, Finite Element Modelling (FEM) and Finite Difference Time Domain (FTDT). In FEM, the structure is included in CAD and meshing is done until the solutions for the Maxwell s equations in the mesh tetrahedrons converge to a certain threshold. Integrating the obtained electric and magnetic fields, the voltage and current is calculated in each mesh point. The ratio of voltage to current calculated by EMPro at the port gives the impedance, which corresponds to S 11. Both EMPro and ADS modelling are used to gain insights into the sub-surface imaging capabilities of oxides and semiconductor materials at different frequencies and different experimental conditions. As such the modelling packages are useful for proper experimental SMM planning as well as data interpretation at a quantitative level. Figure 11. (a) 3D E-field distribution of the tip-sample system done with EM-Pro modelling. The E-field is lower in the doped region with highconductivity (shown in blue), than in the Si bulk of the sample (shown in green). (b) Cross-section of the 3D E-field distribution. (c) EM-Pro model geometry (bottom), with a zoom on a shallow doped structure (top). The shallow stripe is 300nm high, 2.5µm wide, with a conductivity of 10 3 S/m. The silicon bulk is modelled with a conductivity of 1.4 S/m. (d) Cross-section of the E-field distribution showing the tip-sample region. The E-Field is highest close to the tip and decreases inside the sample. Acknowledgements This work was done by Keysight Labs Linz Enrico Brinciotti, Manuel Kasper, and Ferry Kienberger (ferry_kienberger@keysight.com; Principal Investigator), with support from Johannes Kepler University of Linz (JKU Linz) Silviu Sorin-Tuca and Georg Gramse.

11 11 Keysight Scanning Microwave Microscopy Solutions for Quantitative Semiconductor Device Characterization - Application Note References 1. H.P. Huber et al., Calibrated nanoscale capacitance measurements using a scanning microwave microscope, Review of Scientific Instruments 81, , H.P. Huber et al., Calibrated nanoscale dopant profiling using a scanning microwave microscope, J. Appl. Phys. 111, (2012). 3. G. Gramse et al., Calibrated complex impedance and permittivity measurements with scanning microwave microscopy, Nanotechnology 25 (2014) (8pp). 4. T. Schweinböck and S. Hommel, Quantitative Scanning Microwave Microscopy: A calibration flow, Microelectronics Reliability, Volume 54, Issues 9 10, 2014, E. Brinciotti et al., Probing resistivity and doping concentration of semiconductors at the nanoscale using scanning microwave microscopy, Nanoscale, 2015, 7, G. Gramse and E. Brinciotti et al., Quantitative sub-surface and non-contact imaging using scanning microwave microscopy, Nanotechnology 26 (2015) (9pp) E. Brinciotti et al., Frequency analysis of dopant profiling and capacitance spectroscopy using Scanning Microwave Microscopy, submitted in June S. M. Sze and K. Ng Kwok, Physics of Semiconductor devices, John Wiley and Sons, Inc., Hoboken, New Jersey, A. Imtiaz et al., Frequency-selective contrast on variably doped p-type silicon with a scanning microwave microscope, J. Appl. Phys. 111, (2012). 11. L. Michalas et al., De-embedding techniques for nanoscale characterization of semiconductors by scanning microwave microscopy, Journal of Microelectronic Engineering, Volume 159 Issue C, June 2016 Pages M. Kasper et al., Metal-oxide-semiconductor capacitors and Schottky diodes studied with scanning microwave microscopy at 18GHz, Journal of Applied Physics 116, (2014). 13. M. Moertelmaier et al., Continuous capacitance voltage spectroscopy mapping for scanning microwave microscopy, Ultramicroscopy 136(2014) P.F. Medina et al., Transmission and reflection mode scanning microwave microscopy (SMM): experiments, calibration, and simulations, Proceedings of the 45th European Microwave Conference, 7-10 Sept 2015, Paris, France, M. Kasper et al, Electromagnetic Simulations at the nanoscale: EMPro modeling and comparison to SMM experiments, Keysight Application Note EN, A. O. Oladipo, et al., Three-dimensional finite-element simulations of a scanning microwave microscope cantilever for imaging at the nanoscale, Appl. Phys. Lett. 103, (2013). AFM Instrumentation from Keysight Technologies Keysight Technologies offers high precision, modular AFM solutions for research, industry, and education. Exceptional worldwide support is provided by experienced application scientists and technical service personnel. Keysight s leading-edge R&D laboratories are dedicated to the timely introduction and optimization of innovative, easy-to-use AFM technologies. For more information on Keysight Technologies products, applications or services, please contact your local Keysight office. The complete list is available at: Americas Canada (877) Brazil Mexico United States (800) Asia Pacific Australia China Hong Kong India Japan 0120 (421) 345 Korea Malaysia Singapore Taiwan Other AP Countries (65) Europe & Middle East Austria Belgium Finland France Germany Ireland Israel Italy Luxembourg Netherlands Russia Spain Sweden Switzerland Opt. 1 (DE) Opt. 2 (FR) Opt. 3 (IT) United Kingdom For other unlisted countries: This information is subject to change without notice. Keysight Technologies, 2016 Published in USA, July 15, EN