Compound Semiconductor Epitaxy & Nanometrology

Reflection High-Energy Electron Diffraction (RHEED) Epitaxy Lab

Model relativistic electron de Broglie wavelengths (\(\lambda = h / \sqrt{2m_0eV(1 + eV/2m_0c^2)}\)), grazing-angle incidence optics (\(\theta_i \approx 0.5^\circ\text{–}3.0^\circ\)), Ewald sphere reciprocal lattice rods, real-time specular intensity oscillations (\(T_{\text{osc}} = 1/\text{Flux}\)), atomic monolayer layer-by-layer growth, Stranski-Krastanov 3D island transitions, and 20-year cleanroom fab DCF economics.

Epitaxial System & Device Presets

Electron Wavelength 0.0987 Å Relativistic de Broglie \(\lambda\)
Ewald Sphere Radius 63.66 ů¹ Wavevector \(k = 2\pi/\lambda\)
Epitaxial Growth Rate 2.41 Å/s \(R = d_{\text{ML}} / T_{\text{osc}}\) (0.87 μm/h)
Oscillation Period 1.18 s/ML \(T_{\text{osc}} = 1/\text{Flux}\) Monolayer Time
Streak Separation 11.1 mm 0th Laue Zone Screen Spacing
Penetration Depth 0.68 nm Grazing Surface Sensitivity
20-Yr Cleanroom NPV +$22.8M 8% Discount Rate
Discounted Payback 1.31 yrs CapEx Break-Even

RHEED Electron Gun & Growth Kinetics

Material system and lattice constant
Electron energy (10–30 kV)
Incidence grazing angle
Growth rate flux (ML/s)
Substrate pyrometer temp
Roughness decay time
Sample to phosphor screen

Cleanroom Capital Budgeting (DCF)

Gun, screen, CCD, software
Avoided out-of-spec wafers

Figure 1: RHEED grazing-incidence geometry inside the MBE chamber. The relativistic electron beam strikes the wafer at \(\theta_i \approx 0.5^\circ\text{–}3^\circ\), restricting penetration to the topmost monolayer, with the massive Ewald sphere intersecting reciprocal lattice rods.

Figure 2: Real-time RHEED specular beam intensity oscillations \(I(t)\). Intensity maxima denote completed atomically smooth monolayers (\(\theta = 1.0\text{ ML}\)), while minima denote maximum step-edge diffuse scattering at half-coverage (\(\theta = 0.5\text{ ML}\)).

Figure 3: Phosphor screen diffraction view. Atomically smooth 2D surfaces produce sharp vertical streaks along the 0th Laue zone; lattice-mismatched quantum dots spontaneously transition into 3D transmission Bragg spots upon reaching critical thickness.

20-Year Compound Semiconductor Fab Cleanroom DCF Analysis

Cash flow projection evaluating in situ RHEED electron metrology CapEx, filament and vacuum bakeout OpEx ($28,000/yr), and yield scrap avoidance in high-value MQW lasers, GaN RF power HEMTs, and nanosheet superlattices (8% WACC).

Timeline Scrap Benefit ($) Maintenance OpEx ($) Net Cash Flow ($) Discounted Flow (8%) Cumulative NPV ($)

Cleanroom Metrology Knowledge Assessment

Validate your understanding of relativistic electron optics, reciprocal space rods, specular intensity oscillations, and epitaxial growth modes.

1. Why does the RHEED specular beam intensity reach a minimum at half-monolayer coverage (\(\theta = 0.5\text{ ML}\))?

2. In reciprocal space, why does an atomically flat 2D single-crystal surface produce diffraction streaks rather than sharp spots?

3. What does the abrupt transition from a streaky RHEED pattern to a spotty transmission pattern signify during heteroepitaxial growth?

Frequently Asked Questions: RHEED Epitaxy Metrology

RHEED is an in situ surface diagnostic metrology tool used primarily in Molecular Beam Epitaxy (MBE) and chemical beam epitaxy systems. A collimated, high-energy electron beam (typically 10 to 30 keV, yielding de Broglie wavelengths of 0.07 to 0.12 Angstroms) strikes the growing crystal wafer at an extreme grazing angle (0.5 to 3.0 degrees). Because the normal component of the electron momentum is very small, penetration is restricted to the top 1 to 2 atomic monolayers. Diffracted electrons strike a phosphor screen on the opposite side of the chamber, creating a diffraction pattern whose streak separation reflects in-plane surface lattice spacing and whose specular beam intensity oscillates in real time with each newly deposited atomic layer.

In two-dimensional layer-by-layer (Frank-van der Merwe) growth, deposition begins on an atomically flat terrace. At zero coverage (θ = 0), the surface is smooth and reflective, producing maximum specular intensity. As atomic adatoms arrive and nucleate into 2D islands, the perimeter of atomic step edges expands. At half-monolayer coverage (θ = 0.5 ML), the step-edge density and surface roughness reach a maximum. Diffuse scattering off step edges and destructive phase interference between electrons reflecting from the upper and lower terrace levels cause the specular intensity to plunge to a minimum. As the layer completes (θ = 1.0 ML), the surface flattens again and specular intensity reaches a new peak. Exactly one oscillation cycle corresponds to the deposition of one atomic monolayer.

For an atomically smooth 2D surface crystal, there is no periodicity perpendicular to the surface (z-direction). In reciprocal space, this causes the 3D reciprocal lattice points to elongate into continuous 1D reciprocal rods oriented normal to the surface plane. Because high-energy electrons possess large wavevectors (\(k = 2\pi/\lambda \approx 60\text{ to }90\text{ \AA}^{-1}\)), the Ewald sphere has an enormous radius compared to the rod spacing. At grazing incidence, the Ewald sphere intersects these reciprocal rods almost tangentially along their length, projecting vertical streaks onto the phosphor screen rather than isolated diffraction spots.

In lattice-mismatched heteroepitaxy (such as InAs grown on GaAs with 7.2% lattice mismatch), growth initially proceeds as a smooth 2D strained wetting layer producing sharp RHEED streaks. Once the accumulated elastic strain energy exceeds the critical layer thickness (approx 1.7 monolayers for InAs/GaAs), the film relaxes spontaneously by forming 3D islands (quantum dots). When this occurs, grazing electrons can penetrate directly through the sides and facets of the 3D islands, transforming the diffraction mechanism from 2D reflection into 3D transmission electron diffraction. The RHEED screen instantaneously transitions from elongated streaks into bright, spotty Bragg transmission diffraction patterns, providing an exact in situ trigger for shutter closure.

A research- or production-grade RHEED metrology suite (electron gun, high-voltage supply, phosphor screen, CCD camera, and real-time oscillation tracking software) requires an initial CapEx of $380,000 to $620,000 with annual operating expenses of $25,000 to $35,000 for cathode filaments and vacuum viewport bakeouts. In high-value multi-quantum well (MQW) laser diodes, GaN 5G/6G RF power HEMTs, and VCSEL epitaxy, in situ calibration of effusion cell fluxes and atomic layer thickness eliminates ex situ trial runs, saving $2.2M to $3.5M annually in avoided scrap wafers and accelerated tool qualification, yielding a discounted payback period of 1.2 to 1.5 years and a 20-year net present value exceeding $20M.

Related Thin-Film & Nanometrology Simulations

HRXRD Thin-Film Epitaxy Lab

Model high-resolution X-ray diffraction, rocking curves, superlattice satellite fringes, reciprocal space mapping, and fab DCF.

ToF-SIMS Surface Chemical Lab

Model Time-of-Flight Secondary Ion Mass Spectrometry, static SIMS limits, cluster ion yield enhancement, and isobaric resolution.

Spectroscopic Ellipsometry Lab

Simulate thin-film optical constants (\(\Psi\), \(\Delta\)), Cauchy/Tauc-Lorentz dispersion, multi-layer dielectric thickness, and fab yield.

LEED Surface Crystallography Lab

Model Low-Energy Electron Diffraction, retarding field grids, Wood's notation, I(V) curve fitting, and surface fab DCF.

ARPES Band Structure Lab

Model Angle-Resolved Photoemission Spectroscopy, electronic band dispersion, Dirac cones, Fermi surfaces, and lab DCF.