1.10 Domain walls up close
L1 Thin wall, wide halo
An ideal 180° wall in LiNbO3 is only one or two unit cells wide P(abs), yet around a freshly written wall the crystal stays strained and slightly birefringent over micrometres, enough to see walls in a polarizing microscope P-comp.
If a wall is tilted away from c, the polarization on its two sides no longer meets it edge-on and leaves a sheet of bound charge; mobile electrons gather to screen it, and the wall conducts (charged wall). In MgO-doped LN such tilted walls conduct more than 1013 times better than the surrounding crystal P.
10 µm 1 µm 100 nm 10 nm 1 nm 0.1 nm +·····-·--+·····-·--+·····-·--+·····-·--+·····-·--+ A B C D E F ^ s 10.0 µm
A X-ray shear-strain region: up to 10 µm across the wall P-comp
X-ray topography; attributed to defect clusters pinning the wall
B Birefringent region: within 2 µm; Δn 10−5–10−4, congruent P-comp P(abs)
polarizing microscopy; about 10× less after annealing, much less near stoichiometry
C Wall width by PFM: 65–150 nm P-comp
probe-limited: the tip, not the wall, sets this width
D 10 nm: wall width assumed by Werner et al. 2017 P
no measured wall feature here; the assumption converts current to conductivity (1 nm gives about 10× more)
E Ideal wall: 1–2 lattice units, about 0.5–1.0 nm P(abs) derived
approximate: drawn with a = 0.5148 nm; the source does not name the lattice length
F Wall width by TEM: not exceeding 0.28 nm, LiTaO3 P-comp
measured on LiTaO3, not LiNbO3
|------| 10.0 µm view 80.0 µm A [=====] B = C . D . E . F . · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · P up P down | ^ ^ ^ ^ ^ ^ ^ | v v v v v v v | ^ ^ ^ ^ ^ ^ ^ | v v v v v v v | ^ ^ ^ ^ ^ ^ ^ | v v v v v v v | ^ ^ ^ ^ ^ ^ ^ | v v v v v v v | ^ ^ ^ ^ ^ ^ ^ | v v v v v v v |
| A | X-ray shear strain, up to 10 µm | P-comp |
| B | birefringent region, within 2 µm | P-comp |
| C | PFM width 65–150 nm, probe-limited | P-comp |
| D | 10 nm, assumed (Werner et al. 2017) | P |
| E | ideal, 1–2 a ≈ 0.5–1.0 nm | derived |
| F | TEM ≤ 0.28 nm, LiTaO3 | P-comp |
| Scale bar / view | 10.0 µm 80.0 µm | template |
| Wall tilt from c | 0.00 deg | |
| σb = 2Ps sin θ | 0.0000 C/m² | derived |
| σb / e | 0 e/cm² | derived |
| Mean charge spacing | none | derived |
| Wall, θ ≈ 1°, MgO:CLN | 0.02 (Ω cm)−1 | P |
| Bulk dark conductivity | << 10−15 (Ω cm)−1 | P |
0.14 | | ··········· | ··········· 0.07 | ············ | ··········· 0 | o····· --+---------W---------+---------+---------+---------+ 0 1 2 3 4 5 deg σb C/m², Ps 0.71 o current W Werner 2017, ~1 deg
A X-ray shear-strain region: up to 10 µm across the wall P-comp
X-ray topography; attributed to defect clusters pinning the wall
B Birefringent region: within 2 µm; Δn 10−5–10−4, congruent P-comp P(abs)
polarizing microscopy; about 10× less after annealing, much less near stoichiometry
C Wall width by PFM: 65–150 nm P-comp
probe-limited: the tip, not the wall, sets this width
D 10 nm: wall width assumed by Werner et al. 2017 P
no measured wall feature here; the assumption converts current to conductivity (1 nm gives about 10× more)
E Ideal wall: 1–2 lattice units, about 0.5–1.0 nm P(abs) derived
approximate: drawn with a = 0.5148 nm; the source does not name the lattice length
F Wall width by TEM: not exceeding 0.28 nm, LiTaO3 P-comp
measured on LiTaO3, not LiNbO3
- P-comp
- P-comp
- P-comp
- P
- derived
- P-comp
- σb = 2Ps sin θ derived
- 0.0000 C/m2
- Screening electrons derived
- 0 cm−2
- One charge every derived
- none
Zoom from 10 µm to 0.1 nm: the micrometre halo that a polarizing microscope sees narrows to the probe-limited PFM width, then to the seam one to two lattice units wide where lithium sits above its oxygen plane on one side and below it on the other. Tilt the wall by 1° and a bound-charge sheet of 0.0248 C/m2, 1.55 × 1013 electrons per cm2 to screen it (Ps = 0.71 C/m2; 1.74 × 1013 for 0.80), appears along the wall: one charge per 2.54 nm square. Conductivity enters only as the single Werner et al. 2017 point.
L2 Widths and bound charge
| Quantity | Value | Status |
|---|---|---|
| Ideal wall width | 1–2 lattice units (about 0.5–1.0 nm with a = 0.5148 nm; approximate) | P(abs) derived |
| Wall width by piezoresponse force microscopy (PFM) | 65–150 nm (probe-limited) | P-comp |
| Wall width by TEM, LiTaO3 | not exceeding 0.28 nm | P-comp |
| Switching-induced birefringence near walls, congruent | 10−5 to 10−4; reduced about 10× by annealing; much smaller near stoichiometry | P-comp |
| Width of the birefringent region | within 2 µm; shear-strain region by X-ray topography up to 10 µm | P-comp |
| Wall energy, from wall bowing between pinning centres | 0.2–0.4 J/m2 (calculated) | P-comp |
| Bound charge on a wall inclined by θ to c | σb = 2Ps sin θ; at θ = 1°: 0.0248 C/m2 = 1.55 × 1013 e/cm2 (Ps = 0.71 C/m2), 0.0279 C/m2 = 1.74 × 1013 e/cm2 (0.80 C/m2) conflict | derived |
Charged-wall conduction: one sourced example
The numbers below describe one crystal and one writing method; they are a data point, not a material constant (domain-wall conduction).
| Quantity | Value | Status |
|---|---|---|
| Crystal | z-cut congruent LN + 5 mol% MgO, 300 µm | P |
| Wall writing | room-temperature tip (“calligraphic”) poling with current feedback at about 30 nA; lines about 10 µm wide, mm long | P |
| Wall inclination | about 1° | P |
| Wall conductivity (wall width 10 nm assumed) | about 0.02 (Ω cm)−1; bulk dark conductivity well below 10−15 (Ω cm)−1 | P |
| Current | single wall about 0.55 µA at 50 V; 732 parallel 1 mm walls approach 10 mA | P |
| Stability | stable at ≤ 70 °C (about 1% fluctuation over 35 h at 30 °C); activation energy 0.10 ± 0.01 eV below 70 °C | P |
| Earlier demonstration | mm-thick LN single crystals; wall conduction switched by super-bandgap light and tuned by wall tilt | P(abs) |
L3 Probes and assumptions
Why walls are “thick” in congruent crystal. The extended strain and space-charge fields are attributed to defect clusters pinning the wall P-comp P(abs): extrinsic point defects and their clustering control the local wall structure and change macroscopic properties by orders of magnitude. PFM profiles are symmetric across walls in stoichiometric crystal and asymmetric in congruent crystal, read indirectly as a trace of the internal field P-comp. The mechanism remains qualitative.
Measuring walls. Selective etching (the Miller-Savage wall-velocity method), polarizing microscopy of wall birefringence, electro-optic imaging, near-field optics, PFM, SHG microscopy, X-ray topography P-comp. The measured width depends on the probe; the 0.28 nm TEM value is for LiTaO3, not LiNbO3 P-comp.
Conduction numbers depend on the assumed wall width. 10 nm gives 0.02 (Ω cm)−1; 1 nm gives about ten times more P. In Schröder et al. 2012, wall transport in undoped crystal was switched by super-bandgap illumination P(abs).
How the figure is computed. Each scale feature is drawn at its sourced full width around the wall; the feature named at a scale is the one whose representative length is nearest the scale bar on a log axis template. The atom strip repeats one octahedral column of the Hsu et al. 1997 cell along a1, with the refined structure (s = +1) left of the wall and its inversion (s = −1) right of it; Li sits 71.3 pm above or below its O3 face derived. The bound charge is σb = 2Ps sin θ with Ps = 0.71 or 0.80 C/m2 conflict derived; the microscope halo and the PFM profile shape are display renderings template template.
Sources
- Volk & Wöhlecke 2008T. Volk, M. Woehlecke, Lithium Niobate: Defects, Photorefraction and Ferroelectric Switching, Springer Series in Materials Science 115 (Springer, Berlin Heidelberg, 2008). doi:10.1007/978-3-540-70766-0
- Gopalan et al. 2007V. Gopalan, V. Dierolf, D. A. Scrymgeour, "Defect-domain wall interactions in trigonal ferroelectrics", Annu. Rev. Mater. Res. 37, 449-489 (2007). doi:10.1146/annurev.matsci.37.052506.084247
- Werner et al. 2017C. S. Werner, S. J. Herr, K. Buse, B. Sturman, E. Soergel, C. Razzaghi, I. Breunig, "Large and accessible conductivity of charged domain walls in lithium niobate", Sci. Rep. 7, 9862 (2017). doi:10.1038/s41598-017-09703-2
- Hsu et al. 1997R. Hsu, E. N. Maslen, D. du Boulay, N. Ishizawa, "Synchrotron X-ray studies of LiNbO3 and LiTaO3", Acta Cryst. B 53, 420-428 (1997). doi:10.1107/S010876819600777X
- Gopalan et al. 1998V. Gopalan, T. E. Mitchell, Y. Furukawa, K. Kitamura, "The role of nonstoichiometry in 180 ° domain switching of LiNbO3 crystals", Appl. Phys. Lett. 72, 1981-1983 (1998). doi:10.1063/1.121491
- Kitamura et al. 1998K. Kitamura, Y. Furukawa, K. Niwa, V. Gopalan, T. E. Mitchell, "Crystal growth and low coercive field 180 ° domain switching characteristics of stoichiometric LiTaO3", Appl. Phys. Lett. 73, 3073-3075 (1998). doi:10.1063/1.122676
- Schröder et al. 2012M. Schroeder, A. Haussmann, A. Thiessen, E. Soergel, T. Woike, L. M. Eng, "Conducting domain walls in lithium niobate single crystals", Adv. Funct. Mater. 22, 3936-3944 (2012). doi:10.1002/adfm.201201174