Acceptance bandwidth

Acceptance bandwidth quantifies the range of input conditions over which an optical process (nonlinear conversion, grating coupling, filter operation) maintains useful efficiency. It is typically specified as a full-width at half-maximum (FWHM) of the efficiency versus the relevant input parameter.

Three principal acceptance bandwidths in optical-system design:

1. Spectral acceptance bandwidth ($\Delta \lambda$): wavelength range over which the device is efficient
2. Angular acceptance bandwidth ($\Delta\theta$): incidence angle range over which the device is efficient
3. Temperature acceptance bandwidth ($\Delta T$): temperature range over which the device stays in spec (for nonlinear crystals where index varies with temperature)

Spectral acceptance bandwidth for phase-matched nonlinear processes.

For a phase-matched second-harmonic generation in a crystal of length $L$, the SHG efficiency falls off with input wavelength shift as $\text{sinc}^2(\Delta k L / 2)$. The FWHM in wavelength:

$$\Delta \lambda_\text{FWHM} \;\approx\; \frac{0.886 \lambda^2}{2 L \, |n_g(2\omega) - n_g(\omega)|},$$

where $n_g$ is group index. For 10-mm PPLN at 1064 nm pump: $\Delta \lambda \approx 0.2$ nm. For 1-mm PPLN: $\Delta \lambda \approx 2$ nm. Shorter crystal → broader bandwidth but lower peak efficiency.

Spectral acceptance bandwidth for gratings.

For a uniform-period Bragg grating of length $L$ in fiber or waveguide, the reflection bandwidth is:

$$\Delta \lambda \;\approx\; \frac{\lambda_B^2 \kappa}{\pi n_g}, \quad \text{or} \quad \Delta \lambda \;\approx\; \frac{\lambda_B \kappa L}{2 \pi n_g} \text{ for weak gratings},$$

where $\kappa$ is the coupling coefficient. For typical telecom fiber Bragg gratings: $\Delta \lambda = 0.05 - 0.5$ nm. Wider bandwidth → more reflectivity per unit length → shorter gratings possible at the cost of reduced peak reflectivity.

Angular acceptance bandwidth.

For grating couplers, the angular acceptance is the angle range over which the coupling efficiency is within 1 dB of peak:

$$\Delta\theta \;\approx\; \frac{\lambda}{N \Lambda \cos\theta_0},$$

where $N$ is the number of grating periods illuminated. For a silicon photonic grating coupler with $N = 20$ periods and $\Lambda = 600$ nm at $\theta_0 = 10°$: $\Delta\theta \approx 7°$. Longer grating → narrower angular acceptance.

Temperature acceptance bandwidth in nonlinear crystals.

Lithium niobate's refractive indices depend on temperature; thermal expansion changes the QPM period:

$$\Delta T_\text{FWHM} \;\approx\; \frac{0.886 \lambda}{L \, |\partial \Delta k / \partial T|},$$

For 10-mm PPLN at 1064 nm: $\Delta T \approx 3$°C. Temperature stabilization to 0.1 – 1°C is therefore required for stable operation.

Trade-offs.

Application	Wide bandwidth desirable	Narrow bandwidth desirable
Frequency conversion (SHG, SFG)	Broadband ultrafast input	Single-wavelength CW
WDM filter	Multi-channel spread	Single ITU channel
Fiber Bragg grating sensor	Wide measurement range	High strain/temperature resolution
Grating coupler	Wavelength-flexible	Spectral selectivity
Tunable laser external mirror	Wide tuning range	Long coherence length

Bandwidth engineering techniques.

• Crystal length / grating length: shorter → wider bandwidth (and lower efficiency)
• Apodization: smooth the structure's envelope for cleaner spectral response
• Chirped designs: periodically-poled crystals or gratings with spatially-varying period give intentional broadband response
• Multi-section designs: cascaded sections with different periods give multi-band response
• Group-velocity dispersion compensation: matched dispersion in subsequent elements compensates initial dispersion for broadband ops

Comparison: temporal bandwidth vs spectral acceptance bandwidth. A nonlinear conversion of an ultrafast pulse requires bandwidth matching: the pulse spectrum must fit within the crystal's spectral acceptance bandwidth. For a transform-limited 100 fs pulse at 1064 nm: spectral bandwidth $\sim 16$ nm. This requires a $< 1$ mm crystal with broad SHG acceptance bandwidth, dropping single-pass SHG efficiency dramatically.

Acceptance bandwidth product. For a fixed crystal material and wavelength, the product (efficiency × bandwidth × interaction length) tends to be a conserved quantity — the famous Manley-Rowe-like constraint in optical engineering. Broadening the bandwidth proportionally reduces the peak efficiency.

References: Boyd, Nonlinear Optics, Ch. 2; Agrawal, Nonlinear Fiber Optics, Ch. 10; Othonos & Kalli, Fiber Bragg Gratings, Ch. 4 for grating bandwidth treatment.