Recombination Parameter Extraction is the analytical process of fitting experimental minority carrier lifetime data measured as a function of injection level (tau vs. delta_n curves) to recombination physics models to determine the identity, energy level, capture cross-sections, and concentration of electrically active defects in silicon — the quantitative bridge between measurable electrical signals and the atomic-scale defect properties that control device performance.
What Is Recombination Parameter Extraction?
- Input Data: The primary input is an injection-level-dependent lifetime curve, tau_eff(delta_n), measured by QSSPC, transient µ-PCD at multiple injection levels, or time-resolved photoluminescence. This curve contains the signatures of all active recombination mechanisms competing in the material: SRH (defect) recombination, radiative recombination, and Auger recombination.
- SRH Model: Shockley-Read-Hall recombination through a single trap level is described by: tau_SRH = (tau_p0 (n_0 + n_1 + delta_n) + tau_n0 (p_0 + p_1 + delta_n)) / (n_0 + p_0 + delta_n), where tau_n0 = 1/(sigma_n v_th N_t) and tau_p0 = 1/(sigma_p v_th N_t) are the fundamental capture time constants. The parameters n_1 and p_1 are functions of the trap energy level E_t relative to the Fermi level.
- Extracted Parameters: Fitting the measured tau_SRH(delta_n) to the SRH equation yields: E_t (trap energy level, typically expressed as E_t - E_i in eV), k = sigma_n/sigma_p (capture cross-section symmetry parameter), and tau_n0/tau_p0 (related to N_t and capture cross-sections). These three parameters uniquely characterize a defect's electrical activity.
- Defect Fingerprinting: Each defect species has a characteristic (E_t, k) signature. Iron: E_t = E_i + 0.38 eV (FeB pair), k = 37. Chromium-Boron pair: E_t = E_i + 0.27 eV. Gold acceptor: E_t = E_i - 0.06 eV. Comparing extracted parameters to the literature database identifies the physical origin of the lifetime-limiting defect without chemical analysis.
Why Recombination Parameter Extraction Matters
- Non-Destructive Defect Identification: Traditional defect identification requires destructive techniques (SIMS for chemical identity, DLTS for electrical characterization requiring contacts and cryogenic measurements). Recombination parameter extraction from QSSPC data requires only a contactless photoconductance measurement, identifying defects in minutes without any sample preparation or damage.
- Process Root Cause Analysis: When a batch of silicon wafers exhibits unexpectedly low lifetime, recombination parameter extraction determines whether the cause is iron (furnace contamination), chromium (chemical contamination), boron-oxygen complexes (light-induced degradation in p-type Cz silicon), or structural defects (dislocations, grain boundaries). This identification drives targeted process corrective action.
- Quantification of Competing Mechanisms: Real silicon often contains multiple defects simultaneously. Advanced fitting routines (Transient-mode QSSPC, DPSS — Defect Parameter Solution Surface analysis) separate contributions from multiple trap levels to quantify each defect's contribution to total recombination activity.
- Solar Cell Simulation Calibration: Solar cell device simulation requires accurate bulk lifetime as a function of injection level. Extracted SRH parameters provide the physically accurate lifetime model for simulation tools (Sentaurus, PC1D, Quokka), enabling predictive simulation of how changes in silicon quality will affect cell efficiency.
- DPSS (Defect Parameter Solution Surface) Analysis: For a single measured tau(delta_n) curve, multiple combinations of (E_t, k) can produce similar fits. DPSS analysis maps all combinations consistent with the data as a surface in (E_t, k) parameter space, revealing the uniquely identifiable defect parameters and their uncertainties. When data at multiple temperatures is available, the intersection of DPSS surfaces at different temperatures narrows the solution to a unique defect identification.
Practical Workflow
1. Measure: Obtain tau_eff(delta_n) by QSSPC on symmetrically passivated sample (minimize surface recombination).
2. Separate: Subtract Auger contribution (known silicon intrinsic Auger coefficients) and radiative contribution (known intrinsic radiative coefficient) to isolate tau_SRH(delta_n).
3. Fit: Minimize chi-squared between measured tau_SRH and SRH model using non-linear least squares over the parameter space (E_t, k, N_t).
4. Identify: Compare best-fit (E_t, k) to literature database of known defect signatures.
5. Validate: Confirm identification by temperature-dependent measurements (tau_SRH changes predictably with temperature for a given defect) or by correlation with chemical analysis (DLTS, SIMS).
Recombination Parameter Extraction is defect forensics at the atomic scale — decoding the injection-level signature encoded in a lifetime curve to identify the specific atom species, its energy level position, and its concentration without touching the sample, transforming a macroscopic electrical measurement into a quantitative atomic-level defect census.