PPLN Λ = 6.82 µm · 1064 → 532 nm · 2 px/µm

4.1 Light that writes itself: photorefraction

L1 Phenomenon

optical damage, 1966

A beam of visible light passing through lithium niobate leaves behind a change in refractive index that outlasts the beam. Light frees electrons from impurity and defect sites; the crystal's polar structure pushes them along the c axis without an applied voltage (the bulk photovoltaic effect); they are trapped again in the dark edges of the beam. The separated charge sets up a space-charge field, and the Pockels effect turns that field into an index change.

It was first called optical damage (Ashkin et al. 1966 reported (title)): a focused laser beam is distorted and scattered by its own track. It is reversible; uniform light or heat erases it.

Write and erase: |Esc| and Δne of a Gaussian beam trackopen circuit, local saturation
cross-section, x across beam (±2.4 w), z along beam (1 mm); |Δne| log levels 10−8..10−3 .:-=+*#%
.::---==+++***#####%%%%%%%%%%%%%%%%%%%%%%%%%%#####***+++==---::.
.::---==+++****#####%%%%%%%%%%%%%%%%%%%%%%%%#####****+++==---::.
.:::--===+++***#####%%%%%%%%%%%%%%%%%%%%%%%%#####***+++===--:::.
..::--===+++***######%%%%%%%%%%%%%%%%%%%%%%######***+++===--::..
..::--===+++****#####%%%%%%%%%%%%%%%%%%%%%%#####****+++===--::..
..::---==+++****######%%%%%%%%%%%%%%%%%%%%######****+++==---::..
..::---===+++***######%%%%%%%%%%%%%%%%%%%%######***+++===---::..
..:::--===+++****######%%%%%%%%%%%%%%%%%%######****+++===--:::..
...::--===+++****#######%%%%%%%%%%%%%%%%#######****+++===--::...
10⁻³10⁻¹10¹10³10⁵10⁷10⁰10²10⁴10⁶10⁸10⁻⁹10⁻⁷10⁻⁵10⁻³t (s)|E| (V/m)|Δne|
 1.00e+00 W/cm²
 1.00e+01 s
 1.00e−15 S/m
 1.00e+05 s
Crystal
Light
I  1.00e+00 W/cm² |Esat|  2.80e+06 V/m Δne now −4.73e−04 τ lit  2.48e+00 s τ dark  2.48e+05 s =   2.87 d  
kG(Fe) 2.8×10−9 cm/V P-comp σd range P-comp α, κ, w, 1 mm template MgO: κ ×100 reported ε33 28, ne 2.2028, r33T 32.2 pm/V P-comp model parameter, no data: α, κ

Write a scar, turn the light off

532 nm beam into a 1 mm slab; bulk photovoltaic field, open circuit, local saturation
532 nm+c →grid bend ×2400 L·∂Δn/∂y, drawnexposure 10.0 s 10⁻³ s10⁻¹ s10¹ s10³ s10⁵ s10⁷ s10⁻⁸10⁻⁶10⁻⁴|Δne| at the beam centre
1.0e+00 W/cm²
  10.0 s  
Crystal
1.0e−15 S/m
  1.16 d  
Index change now
−4.73e−04
Saturated field
2.80e+06 V/m
Scar lifetime, lit
  2.48 s  
Scar lifetime, dark
  2.87 d  
kG Fe P-comp σd range P-comp MgO: photoconductivity ×100 reported absorption, photoconductivity, beam size: model parameters, no data template ε33 28, ne 2.2028, r33T 32.2 pm/V P-comp beam: 532 nm, computed spectral color
QuantityValueConditionsSource
Lowest intensity at which photorefraction is observed10 mW/cm2blue-green light, LiNbO3:Fe; no intensity thresholdVolk & Wöhlecke 2009, p. 52P-comp
Extraordinary index decreaseup to 10−3focused Ar+ laser, 488 nm, 20 mW; ordinary change much smallerChen 1969reported
Bulk photovoltaic saturation voltage> 1000 V (about 105 V/cm)doped LiNbO3Glass et al. 1974reported

L2 Charge transport

Volk & Wöhlecke Eqs. 3.2-3.19; Schwesyg Eqs. 2.23-2.25

Three currents (Volk & Wöhlecke Eq. 3.2, p. 52; Schwesyg Eq. 2.23 P-comp): the bulk photovoltaic (Glass) current, diffusion of conduction-band electrons, and drift in the total conductivity.

j=jpv+jdiff+jdrift,jpv,i=βijkejekI,jdiff=kBTμe∇ne,jdrift=(σd+κI)E(1)
  • For point group 3m the symmetric photovoltaic tensor has the pattern of the piezoelectric d tensor: nonzero β311 = β322, β333, β222 = −β112 = −β121 = −β211, β113 = β223 (p. 53 P-comp). β113 drives a current that oscillates along the beam; its coherence length (half period) is lc = λ / (2|ne − no|), about 5λ in LN, and the full period λ/|ne − no| about 11λ; it matters only when the lit region is that small, as in waveguides.
  • Scalar form: jpv = kGαI with kG the Glass constant (Eq. 3.5).
  • Open circuit, steady state: drift cancels the photovoltaic current. Once σph = κI ≫ σd, the field no longer depends on intensity.
Epv=kGαIσph+σd,E(t)=−Esat[1−e−t/τ],E(t)=Esate−t/τ,τ=εε0σph+σd(2)

Writing (open circuit) and erasure (uniform light or dark) follow Eqs. 3.18-3.19 (p. 57). The dielectric relaxation time τ sets both. Index change through the linear electro-optic effect (Schwesyg Eq. 2.25 P-comp):

Δne=−12ne3r33Ez,Δno=−12no3r13Ez(3)

r33 is about 3 r13, so extraordinary light sees about three times the change (Schwesyg p. 19). The lit region shows a decrease of ne (Chen 1969 reported); the figures plot |E| and Δne < 0, since Volk & Wöhlecke and Schwesyg do not give the sign of jpv relative to +c.

A fringe pattern with modulation m and wave number K gives a diffusion field (monopolar electrons, Eq. 3.9) that is independent of intensity and shifted a quarter period from the fringes; the photovoltaic field is in phase with them (pp. 54-55):

Edif(z)=kBTeKmsinKz1+mcosKz(4)
QuantityValueConditionsSource
Glass constant kG, LiNbO3:Fe2.8×10−9 cm/VVolk & Wöhlecke 2009, p. 59P-comp
Glass constant kG, LiNbO3:Cu5.5×10−10 cm/VVolk & Wöhlecke 2009, p. 59P-comp
Photoconductivity, as-grown doped crystals10−9 to 10−10 S/mmoderate intensity (10−11 to 10−12 Ω−1cm−1)Volk & Wöhlecke 2009, p. 59P-comp
Saturated photovoltaic field106 to 107 V/mLiNbO3:Fe, LiNbO3:Cu (104 to 105 V/cm)Volk & Wöhlecke 2009, p. 59P-comp
Diffusion field amplitudeabout 1.5×105 V/mgrating period about 1 µmVolk & Wöhlecke 2009, p. 58P-comp
Dark conductivity, undoped or lightly doped CLN10−14 to 10−16 S/mroom temperature, extrapolated from high temperatureVolk & Wöhlecke 2009, p. 60P-comp
Dark storage time of gratings, low-doped LiNbO3:Feup to about 1 yearhours or seconds at high Fe contentVolk & Wöhlecke 2009, p. 60P-comp
Electron Hall mobilityabout 0.8 cm2/(V s)strongly reduced LiNbO3, photo-Hall, room temperatureVolk & Wöhlecke 2009, p. 59P-comp
Sensitivity S′1, LiNbO3:Fe0.01-0.1 cm/J; 0.05-0.3 cm/J633 nm; 488 nmVolk & Wöhlecke 2009, p. 57P-comp

Derived numbers

QuantityValueFormula and inputsStatus
Δne at the saturated photovoltaic field1.7×10−4 to 1.7×10−3½ ne3 r33 E; ne = 2.2028 at 632.8 nm, r33T = 32.2 pm/V (Weis & Gaylord Tables 2, 4); E = 106 to 107 V/mderived
Dielectric relaxation time, dark6.9 h to 29 dεε0/σd; ε33 = 28; σd = 10−14 to 10−16 S/mderived
Dielectric relaxation time, light0.25 s to 2.5 sε33 = 28; σph = 10−9 to 10−10 S/mderived
Diffusion field per unit modulation, 1 µm fringes1.6×105 V/m(kBT/e)(2π/Λ), T = 295 K; agrees with p. 58derived

Conflict. The computed dark relaxation times (6.9 h to 29 d) disagree with the 5 to 500 days printed on p. 159 of the same compilation for the same conductivity range; the upper Δne bound matches the 10−3 reported by Chen 1969.

L3 Mechanism and measurement

dark conduction, transport schemes, holography
  • Dark conductivity in undoped LN is protonic: one activation energy of about 1.1 eV covers 200 to 1000 °C and the conductivity scales with hydrogen concentration; Li+ transport has an activation energy near 0.4 eV, which does not match the measured conductivity (pp. 59-60 P-comp). At 0.2 to 0.25% Fe the activation energy drops to 0.3 to 0.4 eV and electron tunneling between Fe sites takes over (pp. 60-61).
  • Mixed proton-electron dark conduction gives stretched-exponential decay, δΔn ∝ exp[−(t/τ)β]; purely ionic conduction gives a single exponential (p. 61). The write-and-erase figure uses a single exponential.
  • Transport schemes: the one-center model (Fe2+ donor, Fe3+ trap) fails to fit several data sets; the two-center model adds the Nb antisite (NbLi) as a shallow electron trap (Sec. 3.2.2-3.2.3, pp. 62-73). n-type photoconductivity is confirmed only for congruent LN doped with Fe and Cu (p. 59).
  • Full band-transport model: Kukhtarev et al. reported (title) (Crossref issue date 1978, often cited as 1979). The figures use the reduced single-time-constant form and a 1D grating solver.
  • Measurement: two-beam holographic recording with diffraction efficiency η = exp(−αL/cos θ) sin2(πΔnL/(λ cos θ)) (Eq. 3.11); light-induced birefringence between crossed polarizers; beam distortion and far-field scattering (fanning). In waveguides and resonators: resonance shift versus power and time (4.3).
  • Error sources: dark and photoconductivity vary by orders of magnitude between samples, so absolute write and erase times are crystal-specific.
Fringes to field: 1D grating along c, two periods, steady staterows normalized to their amplitude
I(z) − Ī5.00e+03 W/m²ρ(z)3.28e+03 C/m³space-charge E(z)2.05e+06 V/mΔnₑ(z) − mean3.53e−04 0 Λ0.5 Λ1 Λ1.5 Λ2 Λ
phase, Δn vs I
  +3.3 deg
Edif per unit m
1.60e+05 V/m
|E1|, fundamental
1.50e+06 V/m
Λ
 1.00 µm
m
 0.50

Photovoltaic field: local, in phase with the fringes. Diffusion field: from bright to dark, a quarter period off. Set kG to 0 to see 90 deg; diffusion off gives 0 deg.

 1.00 µm
 0.50
1.00 × Fe
 295 K
Diffusion
dashed: closed form Eq. 3.9, diffusion only P-compkG(Fe) P-compI0, κ, σd, α: model parameters, no data templatephase: fundamental Fourier components of Δn and I

Fringes to field

1D grating along c, two periods, steady state
light, 532 nmcharge+++++++++−−−−−−−−−+++++++++−−−−−−−−−index change  +3.3°
1.00 × Fe
 1.00 µm
0.50
295 K
Diffusion current
Index peak behind fringe peak
  +3.3°
Diffusion field per unit contrast
1.60e+05 V/m
Field amplitude, fundamental
1.50e+06 V/m

Photovoltaic field: local, in phase with the fringes. Diffusion field: from bright to dark, a quarter period off. Set kG to 0 to see 90 deg; diffusion off gives 0 deg.

Glass constant at 1 × the Fe value P-compdiffusion field, Eq. 3.9 P-compintensity, conductivities: model parameters, no data templatephase: fundamental Fourier components of Δn and I

Dated steps

YearStepSource
1966optical damage reported in LiNbO3 and LiTaO3Ashkin et al. 1966reported (title)
1967annealing in an electric field made LiTaO3 resistant up to 500 W/cm2; helped LN lessLevinstein et al. 1967reported
1968holograms written into LN and erased thermallyChen et al. 1968reported
1969charge-drift model of the index changeChen 1969reported
1974bulk photovoltaic current as the driverGlass et al. 1974reported
1980, 1984MgO doping: conference abstract, then journal report (4.2)Zhong et al. 1980S Bryan et al. 1984reported

Sources

  1. Ashkin et al. 1966A. Ashkin, G. D. Boyd, J. M. Dziedzic et al., Optically-induced refractive index inhomogeneities in LiNbO3 and LiTaO3. Applied Physics Letters 9 72-74 (1966), doi:10.1063/1.1754607.Read: title
  2. Chen et al. 1968F. S. Chen, J. T. LaMacchia, D. B. Fraser, Holographic storage in lithium niobate. Applied Physics Letters 13 223-225 (1968), doi:10.1063/1.1652580.Read: abstract
  3. Chen 1969F. S. Chen, Optically Induced Change of Refractive Indices in LiNbO3 and LiTaO3. Journal of Applied Physics 40 3389-3396 (1969), doi:10.1063/1.1658195.Read: abstract
  4. Levinstein et al. 1967H. J. Levinstein, A. A. Ballman, R. T. Denton, A. Ashkin, J. M. Dziedzic, Reduction of the Susceptibility to Optically Induced Index Inhomogeneities in LiTaO3 and LiNbO3. Journal of Applied Physics 38 3101-3102 (1967), doi:10.1063/1.1710070.Read: abstract
  5. Glass et al. 1974A. M. Glass, D. von der Linde, T. J. Negran, High-voltage bulk photovoltaic effect and the photorefractive process in LiNbO3. Applied Physics Letters 25 233-235 (1974), doi:10.1063/1.1655453.Read: abstract
  6. Kukhtarev et al. 1978N. V. Kukhtarev, V. B. Markov, S. G. Odulov, M. S. Soskin, V. L. Vinetskii, Holographic storage in electrooptic crystals. I. Steady state. Ferroelectrics 22 949-960 (1978), doi:10.1080/00150197908239450.Read: title
  7. Zhong et al. 1980G. G. Zhong, J. Jin, Z. K. Wu, Measurements of optically induced refractive-index damage of lithium niobate doped with different concentrations of MgO. 11th International Quantum Electronics Conference (IQEC), 1980; J. Opt. Soc. Am. 70, 631 (1980) abstract (1980).Read: not read
  8. Bryan et al. 1984D. A. Bryan, R. Gerson, H. E. Tomaschke, Increased optical damage resistance in lithium niobate. Applied Physics Letters 44 847-849 (1984), doi:10.1063/1.94946.Read: abstract
  9. Volk & Wöhlecke 2009Tatyana Volk, Manfred Wöhlecke, Lithium Niobate: Defects, Photorefraction and Ferroelectric Switching. Springer Series in Materials Science 115 (2009), doi:10.1007/978-3-540-70766-0.Read: selected chapters: Ch. 2.2 (pp. 14-19), Ch. 3 (pp. 51-61), Ch. 4 (pp. 75-77, 82-85, 94-97), Sec. 7.1.3 (p. 168)
  10. Weis & Gaylord 1985R. S. Weis, T. K. Gaylord, Lithium niobate: Summary of physical properties and crystal structure. Applied Physics A Solids and Surfaces 37 191-203 (1985), doi:10.1007/BF00614817.Read: full
  11. Schwesyg 2011Judith R. M.-L. Schwesyg, Interaction of light with impurities in lithium niobate crystals. Dr. rer. nat. dissertation, Universität Bonn (2011).Read: selected: Sec. 2.2.2-2.3.2 (pp. 10-21), Ch. 5 Sec. 5.1 (pp. 69-72), Sec. 5.4-5.5 (pp. 99-101)