Semiconductor Defect Metrology

Photothermal Deflection Spectroscopy (PDS) & Sub-Bandgap Defect Lab

Model optical mirage effect deflection, 6-decade sub-bandgap absorption, Urbach disorder tails, deep trap densities, and 20-year metrology laboratory DCF economics.

α·d down to 10-5 Sensitivity

Pristine MAPbI3 thin film with ultra-sharp Urbach edge (15 meV) and negligible sub-gap trap density (~1.2e15 cm^-3), delivering >22% solar cell efficiency.

Urbach Energy EU

15.0 meV

Crystalline disorder
Deep Defect Nd

1.20e+15 cm⁻³

Integrated trap density
Absorption α(hν)

18.5 cm⁻¹

At operating photon energy
Mirage Deflection θ

14.2 μrad

Lock-in beam bend
Dynamic Range

5.8 Decades

10⁻¹ to 10⁵ cm⁻¹
Thermal Length μliq

107.0 μm

Fluid boundary layer

PDS Physical Parameters

1.58 eV
15.0 meV
18 cm⁻¹
Peak optical absorption from sub-gap dangling bonds and point defect states.
450 nm
10 Hz
Fluorocarbon fluids deliver ~5× greater mirage deflection than water.
1.55 eV
Sweep photon energy to inspect local absorption α, beam deflection θ, and detector response.

PDS Sub-Bandgap Absorption Spectrum α(hν)

Semi-Logarithmic (6 Decades)

Optical Mirage Beam Geometry & Thermal Gradient

Fluidic Mirage Deflection

Automated PDS Metrology Suite & Commercial Testing DCF

20-Year discounted cash flow (DCF) model for commercial automated PDS thin-film defect metrology.

IRR: 68.2% | Payback: 1.3 Yrs
$320,000
PDS rig, Xenon lamp, lock-in
$36,000/yr
Fluids, lamps, technician
900
Sample spectra/year
$550
Fee per sample scan
8%
WACC capital cost
20-Year NPV $3.74M $495k/yr Rev
Year Revenue OpEx EBITDA Taxes (21%) Free Cash Flow Discounted PV Cumulative Cash

Engineering Guide: Photothermal Deflection Spectroscopy & Defect Physics

1. Mirage Effect Thermal Diffusion Dynamics

Photothermal Deflection Spectroscopy (PDS) measures optical absorption via the mirage effect. When a monochromatic pump beam of intensity \(I_0(h\nu)\) strikes a thin film of thickness \(d\), the absorbed power per unit area is:

\(P_{\text{abs}}(h\nu) = I_0(h\nu) [1 - R(h\nu)] [1 - e^{-\alpha(h\nu) d}]\)

Non-radiative recombination converts absorbed optical energy into thermal phonons, establishing a periodic surface temperature fluctuation \(\Delta T_0 e^{i\omega t}\). This periodic heating diffuses into the surrounding liquid medium (e.g., perfluorohexane FC-72). The thermal diffusion length in the fluid is governed by:

\(\mu_{\text{liq}} = \sqrt{\frac{D_{\text{liq}}}{\pi f}}\)

where \(D_{\text{liq}} \approx 3.6 \times 10^{-4}\text{ cm}^2/\text{s}\) is thermal diffusivity and \(f\) is the optical chopping frequency (typically \(8\text{ to }15\text{ Hz}\)), resulting in a boundary layer thickness of \(\sim 100\ \mu\text{m}\).

2. Optical Beam Deflection & Detection

A highly stable He-Ne probe laser beam (\(\lambda = 633\text{ nm}\)) skims parallel to the sample surface inside the thermal boundary layer. Because liquids have large negative thermo-optic coefficients (\(dn/dT \approx -5.4 \times 10^{-4}\text{ K}^{-1}\)), the refractive index decreases near the warm sample, deflecting the probe beam outward:

\(\theta(t) = \frac{1}{n} \frac{dn}{dT} \int_{-\infty}^{\infty} \frac{\partial T(x, z, t)}{\partial z} dx\)

A split bi-cell quadrant photodetector (PSD) measures the deflection angle \(\theta\). By normalizing against a high-absorption saturated region (\(\alpha d \gg 1\)), absolute absorption coefficients \(\alpha(h\nu)\) are recovered down to \(10^{-1}\text{ cm}^{-1}\), completely immune to surface scattering.

3. Urbach Disorder Tail Analysis

Immediately below the fundamental bandgap \(E_g\), optical absorption falls exponentially according to Urbach's rule:

\(\alpha(h\nu) = \alpha_0 \exp\left(\frac{h\nu - E_g}{E_U}\right)\)

The Urbach energy \(E_U\) represents the characteristic width of localized band tail states originating from structural disorder, bond-angle distortions, and thermal lattice vibrations. In photovoltaic semiconductors (e.g., MAPbI3, CIGS, GaAs), a lower \(E_U\) correlates directly with lower open-circuit voltage (\(V_{oc}\)) deficit:

  • Pristine Halide Perovskites: \(E_U \approx 13\text{--}16\text{ meV}\) (approaching thermal limit \(k_B T \approx 25.8\text{ meV}\)).
  • Disordered/Degraded Perovskites: \(E_U > 35\text{ meV}\), signaling severe phase segregation and non-radiative recombination losses.
  • Device-grade a-Si:H: \(E_U \approx 45\text{--}55\text{ meV}\).

4. Deep Defect Quantification (Jackson-Amer Method)

Below the Urbach tail, excess absorption \(\alpha_{\text{excess}}(h\nu)\) emerges from sub-bandgap deep trap states (such as dangling bonds or point vacancies). The integrated defect density \(N_d\) is evaluated via the Jackson-Amer relation:

\(N_d = C \int_{\text{sub-gap}} \left[ \alpha(h\nu) - \alpha_{\text{Urbach}}(h\nu) \right] d(h\nu)\)

where \(C \approx 7.9 \times 10^{15}\text{ cm}^{-2}\text{eV}^{-1}\) is the calibration constant calibrated against electron spin resonance (ESR), allowing rapid, non-destructive defect quantification down to \(10^{14}\text{ cm}^{-3}\).

Frequently Asked Questions About Photothermal Deflection Spectroscopy (PDS)

Photothermal Deflection Spectroscopy (PDS), often termed the optical mirage effect, is an ultra-sensitive photothermal characterization technique designed to measure extremely low optical absorption in semiconductor thin films. Conventional UV-Vis spectrophotometers calculate absorption indirectly by measuring transmission (T) and reflection (R) via A = 1 - T - R; this makes them incapable of resolving absorption coefficients below alpha * d ~ 10^-2 due to diffuse surface scattering and substrate interference. In contrast, PDS directly detects the thermal wave generated solely by non-radiative absorption. Consequently, PDS is completely insensitive to elastic surface scattering and achieves a sensitivity limit of alpha * d down to 10^-5, providing over 5 to 6 orders of magnitude of dynamic range.

The Urbach energy (E_U) characterizes the exponential slope of the sub-bandgap absorption edge, described by the relation alpha(h*nu) = alpha_0 * exp((h*nu - E_g) / E_U). Originating from structural disorder, bond-angle deviations, microstrain, and thermal phonon fluctuations that cause band-edge electronic state tailing into the forbidden gap, E_U serves as a definitive figure of merit for semiconductor crystalline quality. Pristine, high-efficiency lead-halide perovskite single crystals and films exhibit sharp Urbach energies between 13 and 16 meV, whereas disordered, defective, or moisture-degraded films display broadened tails exceeding 30 to 50 meV, leading directly to severe open-circuit voltage deficits in photovoltaic devices.

When a sample immersed in a high dn/dT deflecting fluid (such as perfluorohexane FC-72, dn/dT = -5.4e-4 K^-1) absorbs a chopped monochromatic pump beam, non-radiative de-excitation heats the sample surface periodically at frequency f. This heat diffuses outward into the fluid, establishing a time-dependent thermal gradient dT/dz within a thermal boundary layer of thickness mu_liq = sqrt(D / (pi * f)). A stable He-Ne probe laser beam grazing the film surface encounters this index gradient and bends away from the hot surface by deflection angle theta = (1/n) * (dn/dT) * integral(dT/dz dx). A quadrant position-sensitive photodiode (PSD) coupled to a lock-in amplifier detects this deflection with sub-microradian precision.

Deep in the sub-bandgap regime below the exponential Urbach tail, absorption reaches an excess shoulder or plateau caused by optical transitions from occupied defect states (such as silicon dangling bonds in a-Si:H, halide vacancies in perovskites, or copper vacancies in CIGS) into the conduction band. By subtracting the extrapolated Urbach exponential tail from the total measured PDS absorption spectrum, the excess defect absorption alpha_excess(h*nu) is isolated. Applying the Jackson-Amer calibration integral, N_d = C * integral(alpha_excess d(h*nu)) where C ~ 7.9e15 cm^-2 eV^-1, yields the absolute deep-level trap density in cm^-3.

A research-grade automated PDS workstation—comprising a high-stability Xenon arc lamp with automated grating monochromator, precision optical chopper, fluidic quartz cuvette chamber, 5-axis motorized kinematic stages, He-Ne probe laser, bi-cell position sensitive detector, and dual-phase digital lock-in amplifier—incurs a capital expenditure (CapEx) of approximately $280,000 to $380,000. Annual operational expenses (OpEx) for Xenon lamp replacement, perfluorohexane fluids, and laboratory technician time average $30,000 to $45,000. Commercial semiconductor and photovoltaic defect metrology runs billing at $450 to $700 per sample over 800 to 1,200 annual tests generate $400,000 to $700,000 in revenue, achieving full payback in 1.2 to 1.8 years and an IRR exceeding 55%.

PDS & Defect Spectroscopy Knowledge Check

Related Advanced Characterization & Metrology Labs

MOKE Spintronics Lab

Model Polar/Longitudinal Kerr hysteresis, STT-MRAM wafer testing, and skyrmion imaging.

s-SNOM Nanophotonics Lab

Model tip-enhanced near-field polarizability, 2D polariton interference, and optical harmonic demodulation.

Scanning SQUID Lab

Model spatial magnetic flux mapping, 2D Biot-Savart current inversion, and IC defect isolation.

Thermal Desorption Lab

Model Polanyi-Wigner desorption kinetics, Redhead activation energy, and hydrogen trapping.

TDTR Thermal Lab

Model picosecond thermoreflectance, Kapitza conductance G, and heat diffusion.