How to Learn Scanning Spreading Resistance Microscopy (SSRM): From Nanoscale Tip Contact and Spreading Resistance to Dopant Profiles, Junctions and Next-Generation CMOS Metrology

## Wait, What? SSRM Does Not “See Dopant Atoms” — It Sees How Current Spreads Away From a Nanometre-Scale Contact Touch a conductive tip to a semiconductor cross section. Apply a bias. Current leaves the tiny contact and spreads into the bulk. That spreading creates a resistance. Change the local carrier concentration, mobility, contact condition or surface damage and the measured resistance changes. > **SSRM is not a chemical dopant microscope. It is a nanoscale electrical-contact experiment whose signal depends on the local resistivity plus tip geometry, force, barrier physics, surface preparation and series resistance.** ## The One-Sentence Answer **Learn SSRM by tracing conductive tip → tiny contact → current spreading → local resistance → calibrated resistivity/carrier concentration, then add contact mechanics, oxide removal, mobility, Schottky barriers, tip wear, series resistance and cross-section damage before turning a resistance map into a quantitative dopant profile.** # Beginner Layer — From Point Contact to Spreading Resistance ## Stage 1: Current Through a Tiny Contact Does Not Flow Straight Down It spreads outward into the semiconductor. ## Stage 2: Spreading Creates a Geometric Resistance For simple homogeneous conductors, constriction/spreading-resistance theory links contact radius and resistivity. ## Stage 3: Semiconductor Resistivity Depends on Carrier Concentration and Mobility Conceptually: **ρ ≈ 1/(q n μ)** for one dominant carrier type. ## Stage 4: Local Resistance Can Therefore Carry Doping Information But only after mobility and contact effects are accounted for. # Why SSRM Uses a Hard Conductive Tip ## Stage 5: A Native Oxide Can Block Current Silicon cross sections rapidly oxidize in air. ## Stage 6: High Contact Force Helps Establish Electrical Contact Hard diamond-coated probes are widely used. ## Stage 7: Force Is Therefore Part of the Measurement Too little force gives unstable contact. Too much force can damage the sample or tip. # Contact-Mechanics Layer ## Stage 8: The Electrical Contact Radius Is Not Simply the Nominal Tip Radius It depends on force, tip curvature, elastic modulus and plastic deformation. ## Stage 9: Contact Area Changes the Spreading Resistance A resistance difference can therefore arise from mechanics even if resistivity is unchanged. ## Stage 10: Force Series Are an Important Deletion Test True carrier-profile features should not disappear merely because contact force is tuned within a stable measurement regime. # Surface-Preparation Layer ## Stage 11: Cross-Sectional SSRM Creates a New Surface Samples may be cleaved, polished or ion-milled. ## Stage 12: Preparation Can Damage the Electrical Surface Polishing or FIB processing can introduce amorphization, traps, implanted ions and altered mobility. ## Stage 13: The Measured Resistance Belongs to the Prepared Cross Section It is not automatically the untouched buried device state. # Calibration Layer ## Stage 14: Raw Resistance Is Not Carrier Concentration A calibration sample with known carrier density is often required. ## Stage 15: Mobility Must Be Treated Consistently Ambient-controlled SSRM work has shown that using position-dependent mobility rather than a single average can materially improve reconstructed carrier profiles. ## Stage 16: Calibration Can Be Material Specific A silicon calibration cannot automatically be transferred to SiC, GaN or III–V compounds. # Carrier-Type Layer ## Stage 17: SSRM Measures Resistance Magnitude More Directly Than Carrier Sign p-type and n-type regions can have similar resistivity at different concentrations. ## Stage 18: Junction Identification Often Uses Device Context or Complementary SCM SSRM and SCM are naturally complementary: – SSRM → local resistance/carrier-density response; – SCM → carrier type + differential capacitance. # p–n Junction Layer ## Stage 19: A Junction Has a Real Depletion Region The local electrical transition is not infinitely sharp. ## Stage 20: Tip Bias Can Modify the Junction Local electric fields can perturb depletion and current flow. ## Stage 21: Apparent Junction Position Is Therefore Measurement-Condition Dependent The metallurgical dopant crossing and the electrically observed resistance boundary are not always identical. # Series-Resistance Layer ## Stage 22: Very Highly Doped Regions Can Become Difficult Once local spreading resistance becomes small, parasitic series resistance elsewhere in the circuit can dominate. ## Stage 23: “Outwitting the Series Resistance” Is a Real SSRM Problem Advanced methods separate local contact/spreading resistance from amplifier/cabling resistance and sample backside contact. ## Stage 24: A Flat Low-Resistance Image Can Be Instrument Limited It does not automatically mean perfectly uniform heavy doping. # Contact-Barrier Layer ## Stage 25: Tip–Semiconductor Contact Can Be Schottky-Like Current may depend on barrier height, bias polarity and surface states. ## Stage 26: Local I–V Spectroscopy Helps Diagnose Non-Ohmic Contact A quantitative resistance map should not rely on one bias if the contact is strongly nonlinear. # Spatial-Resolution Layer ## Stage 27: Tip Radius Is Only One Limit Electrical resolution also depends on contact radius, current spreading, depletion width and sample geometry. ## Stage 28: Resolution Can Reach Few-Nanometre Scale Under Favorable Conditions Recent SiC work reports approximately 5-nm-class SSRM cross-sectional mapping. ## Stage 29: Pixel Size Is Not Physical Resolution Dense sampling cannot undo a broad electrical interaction volume. # Wide-Bandgap Semiconductors ## Stage 30: 4H-SiC Is a Major Modern SSRM Application Power MOSFET cells contain source implants, body regions, JFET regions and drift layers. ## Stage 31: SSRM + SCM Separates Complementary Information Recent work has demonstrated cross-sectional 2D carrier profiles in 4H-SiC MOSFETs using both methods. ## Stage 32: Wide-Bandgap Surfaces Are Especially Preparation Sensitive Surface states and contact resistance can dominate if the cross section is not controlled. # Next-Generation CMOS Frontier ## Stage 33: SSRM Is Re-Entering a Critical Semiconductor-Metrology Phase A 2025 *Applied Physics Reviews* article describes SSRM’s continued role as CMOS architectures move from FinFETs toward nanosheet FETs, CFETs and increasingly three-dimensional device stacks. ## Stage 34: 3D Architectures Make “One Cross Section” Less Representative Future metrology must combine serial sections, tomography-like reconstruction, process simulation and correlative microscopy. ## Stage 35: The More Three-Dimensional the Device, the More the Inverse Problem Matters A resistance pixel may be influenced by nearby buried conductive paths. # SSRM Versus Conductive AFM ## Stage 36: C-AFM Asks “How Much Current Flows Here?” ## Stage 37: SSRM Asks “What Local Spreading Resistance Corresponds to Semiconductor Carrier Density?” The hardware can overlap. The scientific job and calibration differ. # SSRM Versus SCM ## Stage 38: SCM Uses a Local MOS Capacitance Receiver ## Stage 39: SSRM Uses Direct Contact Resistance Neither should silently absorb the other’s canonical job. # Modelling Layer ## Stage 40: Finite-Element or TCAD Modelling Can Reproduce Current Spreading The model can include realistic geometry, contact radius, doping, mobility and barriers. ## Stage 41: Quantitative Inversion Is Usually More Reliable Than One Universal Analytical Formula Especially in nanoscale devices. # Machine-Learning Layer ## Stage 42: ML Can Accelerate Resistance-to-Doping Inversion ## Stage 43: It Can Also Learn Tip Wear or Preparation Damage Training must include multiple tips, force, surface conditions and device geometry. ## Stage 44: Physics Validation Must Reproduce Measured I–V and Force Dependence # Professional Layer ## Stage 45: Separate Five Objects 1. true electrically active carrier/dopant distribution; 2. tip–sample contact mechanics/barrier; 3. current-spreading field; 4. measured local resistance; 5. inferred carrier profile. ## Stage 46: Professional SSRM Is a Contact–Spreading–Carrier Inverse Problem > **Which carrier concentration or junction profile remains identifiable after contact radius, tip force, mobility, surface damage, Schottky behaviour, series resistance and three-dimensional current spreading are all allowed to explain the same resistance map?** # Evidence: What Makes an SSRM Claim Strong? Stronger evidence combines calibrated doping standards, force series, bias/I–V series, repeat tips, cross-section preparation controls, backside-contact checks, known mobility model, SCM/SIMS comparison, TCAD/FEM current-spreading model and pre/post tip-condition verification. # Misconceptions Worth Hunting – SSRM directly counts dopant atoms. – The measured resistance is simply resistivity times a constant. – Tip force only affects image stability. – A diamond tip never wears. – Native oxide is irrelevant because the tip is conductive. – Low resistance always means high carrier concentration. – p-type and n-type regions are always distinguishable from resistance alone. – Pixel size equals electrical resolution. – A cross-sectional map reproduces the untouched buried device. – Series resistance only matters in low-conductivity samples. – One silicon calibration works for every semiconductor. – A 3D device can always be interpreted with a 2D spreading-current model. # Transfer Check A nominally uniform region becomes lower resistance when contact force increases. Did carrier concentration rise? **No. Contact area or barrier resistance probably changed.** A heavily doped source region appears uniformly saturated at the amplifier’s low-resistance limit. Is the detailed dopant profile known? **No. Series/instrument limits may have erased contrast.** A SiC device gives a 5-nm-wide electrical transition. Does that automatically mean the metallurgical junction is 5 nm wide? **No. Contact geometry and depletion physics remain part of the response.** A polished cross section disagrees strongly with SIMS only in the first few nanometres from the prepared surface. Could preparation damage explain it? **Yes.** # How We Know the Learning Has Held A learner should be able to explain spreading resistance, connect resistivity to carrier density and mobility, explain hard-tip/high-force operation, identify oxide/contact effects, explain calibration, identify series resistance, distinguish SSRM from SCM and C-AFM, explain junction broadening, explain cross-section preparation artifacts, explain SiC and advanced CMOS applications and identify FEM/TCAD inversion limits. # Model Limits SSRM requires a sufficiently conductive path from the probe to the sample contact and a stable nanoscale electrical interface. It does not directly measure chemical dopant concentration, carrier sign, mobility or junction chemistry without other information. Professional SSRM keeps **tip material + force + contact radius + bias + surface preparation + carrier mobility + series resistance + sample geometry + calibration + orthogonal dopant evidence** visible together. # Teaching Guide Teach in this order: **tiny contact → current spreading → spreading resistance → carrier resistivity → hard probe → contact mechanics → oxide/surface preparation → calibration → junctions → series resistance → barriers/I–V → resolution → SiC → advanced CMOS → FEM/TCAD → validation.** # Connect This to the eduKate Learning Estate – Atomic Force Microscopy — topography/contact mechanics. – Conductive AFM — general local current mapping. – Scanning Capacitance Microscopy — dC/dV carrier profiling. – KPFM — work function and electrostatic potential. – Semiconductors and Transistors — device mechanism owner. # Research Foundations and Further Learning – NIST review of electrical scanning probe microscopy for semiconductor metrology. – Bruker SSRM/SSRM-HR technical framework. – Ambient-controlled SSRM measurement and modelling. – Series-resistance analysis in quantitative SSRM. – 4H-SiC MOSFET carrier profiling by SSRM and SCM, 2024. – Laskar et al., *The enduring legacy of scanning spreading resistance microscopy: Overview, advancements, and future directions*, *Applied Physics Reviews* 12, 041305 (2025). # The Quiet Ending The beginner asks, “Which region had lower resistance?” The developing semiconductor scientist asks, “How did current spread away from the tip?” The advanced learner asks, “How much of the map belongs to carrier concentration, and how much to force, contact barrier or mobility?” And the professional asks: > **Which nanoscale dopant profile survives after the tip–sample electrical contact itself is treated as part of the semiconductor device?**