Limits...
Resolving transitions in the mesoscale domain configuration in VO2 using laser speckle pattern analysis.

Seal K, Sharoni A, Messman JM, Lokitz BS, Shaw RW, Schuller IK, Snijders PC, Ward TZ - Sci Rep (2014)

Bottom Line: The configuration and evolution of coexisting mesoscopic domains with contrasting material properties are critical in creating novel functionality through emergent physical properties.However, current approaches that map the domain structure involve either spatially resolved but protracted scanning probe experiments without real time information on the domain evolution, or time resolved spectroscopic experiments lacking domain-scale spatial resolution.Our straightforward analysis of laser speckle patterns across the first order phase transition of VO2 can be generalized to other systems with large scale phase separation and has potential as a powerful method with both spatial and temporal resolution to study phase separation in complex materials.

View Article: PubMed Central - PubMed

Affiliation: 1] Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA [2] Department of Physics &Astronomy, University of Tennessee, Knoxville, TN 37996, USA.

ABSTRACT
The configuration and evolution of coexisting mesoscopic domains with contrasting material properties are critical in creating novel functionality through emergent physical properties. However, current approaches that map the domain structure involve either spatially resolved but protracted scanning probe experiments without real time information on the domain evolution, or time resolved spectroscopic experiments lacking domain-scale spatial resolution. We demonstrate an elegant experimental technique that bridges these local and global methods, giving access to mesoscale information on domain formation and evolution at time scales orders of magnitude faster than current spatially resolved approaches. Our straightforward analysis of laser speckle patterns across the first order phase transition of VO2 can be generalized to other systems with large scale phase separation and has potential as a powerful method with both spatial and temporal resolution to study phase separation in complex materials.

No MeSH data available.


Related in: MedlinePlus

Subtracted autocorrelation functions for different temperatures under 800 nm illumination along the y (a) and x (b) axes with red highlighted regions showing transition between 58.1°C and 62.2°C. The shifts in the x and y axes, dx and dy, are in pixels. Subtracted autocorrelation functions across the same temperature range under 633 nm illumination along y (c) and x (d) with red highlighted regions showing transition between 56.8°C and 58.9°C.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4151099&req=5

f4: Subtracted autocorrelation functions for different temperatures under 800 nm illumination along the y (a) and x (b) axes with red highlighted regions showing transition between 58.1°C and 62.2°C. The shifts in the x and y axes, dx and dy, are in pixels. Subtracted autocorrelation functions across the same temperature range under 633 nm illumination along y (c) and x (d) with red highlighted regions showing transition between 56.8°C and 58.9°C.

Mentions: While first order intensity statistics allow one to extract the roughness of the dielectric landscape, second order intensity statistics such as autocorrelation analysis is very sensitive to small variations in speckle sizes. Speckle correlation function coefficients depend on the distribution of the dielectric constant at the surface, with the spatial resolution for roughness features being constrained by the diffraction limit17232425. The autocorrelation function thus provides an independent spatial analysis of the dominant length scales present in each speckle pattern which are related to the morphology of the dielectric landscape on the VO2 surface. We calculate the 2D autocorrelation plot from each image as well as the cross correlation of consecutive pairs of image recorded with increasing temperature. To effectively analyze this, first the 2D autocorrelation, CTn, from each image (at each temperature) was obtained using , where δI(x,y) = (I(x,y) − 〈I(x,y)〉)/〈I(x,y)〉. Next, the x and y cross sections or of each image were obtained and subtracted from their consecutive counterpart recorded at the next temperature, . The maximum value of the shift in the x and y directions (dx and dy, respectively) were about half the image size (~250 pixels). The results for wavelengths 633 nm and 800 nm are plotted in Fig. 4(a)–(d). The subtracted autocorrelation cross sections, Csub, are essentially flat at most temperatures. This is especially true for the area of the central maximum at all temperatures, which confirms that the average speckle size stays constant with temperature at each wavelength. Near the MIT transition temperature, a secondary peak develops away from the central maximum region, as seen most prominently in Fig. 4(a) and (b) at T = 57°C for an illumination wavelength 633 nm and at a slightly higher temperature of 59°C for 800 nm. The secondary maxima are a signature of long range correlation, which indicates a decreased scattering mean free path25 and increased scattering strength. Additional small maxima at different positions are also noticeable at temperatures 61°C for 800 nm and 62°C for 633 nm. We attribute these maxima to changes in the metallic domain size and the resulting effective dielectric contrast near the transition. Coupled with the peak in effective dielectric roughness variation observed in Fig. 3(c), this indicates that the dielectric contrast is maximum at the MIT as expected from the percolation model. At shorter wavelengths the maxima occur across a wider temperature range. This is due to the fact that detected domain sizes are smaller and will therefore be detected further away from the transition point, similar to the larger width in the variance for shorter wavelengths (see Fig. 2).


Resolving transitions in the mesoscale domain configuration in VO2 using laser speckle pattern analysis.

Seal K, Sharoni A, Messman JM, Lokitz BS, Shaw RW, Schuller IK, Snijders PC, Ward TZ - Sci Rep (2014)

Subtracted autocorrelation functions for different temperatures under 800 nm illumination along the y (a) and x (b) axes with red highlighted regions showing transition between 58.1°C and 62.2°C. The shifts in the x and y axes, dx and dy, are in pixels. Subtracted autocorrelation functions across the same temperature range under 633 nm illumination along y (c) and x (d) with red highlighted regions showing transition between 56.8°C and 58.9°C.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4151099&req=5

f4: Subtracted autocorrelation functions for different temperatures under 800 nm illumination along the y (a) and x (b) axes with red highlighted regions showing transition between 58.1°C and 62.2°C. The shifts in the x and y axes, dx and dy, are in pixels. Subtracted autocorrelation functions across the same temperature range under 633 nm illumination along y (c) and x (d) with red highlighted regions showing transition between 56.8°C and 58.9°C.
Mentions: While first order intensity statistics allow one to extract the roughness of the dielectric landscape, second order intensity statistics such as autocorrelation analysis is very sensitive to small variations in speckle sizes. Speckle correlation function coefficients depend on the distribution of the dielectric constant at the surface, with the spatial resolution for roughness features being constrained by the diffraction limit17232425. The autocorrelation function thus provides an independent spatial analysis of the dominant length scales present in each speckle pattern which are related to the morphology of the dielectric landscape on the VO2 surface. We calculate the 2D autocorrelation plot from each image as well as the cross correlation of consecutive pairs of image recorded with increasing temperature. To effectively analyze this, first the 2D autocorrelation, CTn, from each image (at each temperature) was obtained using , where δI(x,y) = (I(x,y) − 〈I(x,y)〉)/〈I(x,y)〉. Next, the x and y cross sections or of each image were obtained and subtracted from their consecutive counterpart recorded at the next temperature, . The maximum value of the shift in the x and y directions (dx and dy, respectively) were about half the image size (~250 pixels). The results for wavelengths 633 nm and 800 nm are plotted in Fig. 4(a)–(d). The subtracted autocorrelation cross sections, Csub, are essentially flat at most temperatures. This is especially true for the area of the central maximum at all temperatures, which confirms that the average speckle size stays constant with temperature at each wavelength. Near the MIT transition temperature, a secondary peak develops away from the central maximum region, as seen most prominently in Fig. 4(a) and (b) at T = 57°C for an illumination wavelength 633 nm and at a slightly higher temperature of 59°C for 800 nm. The secondary maxima are a signature of long range correlation, which indicates a decreased scattering mean free path25 and increased scattering strength. Additional small maxima at different positions are also noticeable at temperatures 61°C for 800 nm and 62°C for 633 nm. We attribute these maxima to changes in the metallic domain size and the resulting effective dielectric contrast near the transition. Coupled with the peak in effective dielectric roughness variation observed in Fig. 3(c), this indicates that the dielectric contrast is maximum at the MIT as expected from the percolation model. At shorter wavelengths the maxima occur across a wider temperature range. This is due to the fact that detected domain sizes are smaller and will therefore be detected further away from the transition point, similar to the larger width in the variance for shorter wavelengths (see Fig. 2).

Bottom Line: The configuration and evolution of coexisting mesoscopic domains with contrasting material properties are critical in creating novel functionality through emergent physical properties.However, current approaches that map the domain structure involve either spatially resolved but protracted scanning probe experiments without real time information on the domain evolution, or time resolved spectroscopic experiments lacking domain-scale spatial resolution.Our straightforward analysis of laser speckle patterns across the first order phase transition of VO2 can be generalized to other systems with large scale phase separation and has potential as a powerful method with both spatial and temporal resolution to study phase separation in complex materials.

View Article: PubMed Central - PubMed

Affiliation: 1] Materials Science and Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA [2] Department of Physics &Astronomy, University of Tennessee, Knoxville, TN 37996, USA.

ABSTRACT
The configuration and evolution of coexisting mesoscopic domains with contrasting material properties are critical in creating novel functionality through emergent physical properties. However, current approaches that map the domain structure involve either spatially resolved but protracted scanning probe experiments without real time information on the domain evolution, or time resolved spectroscopic experiments lacking domain-scale spatial resolution. We demonstrate an elegant experimental technique that bridges these local and global methods, giving access to mesoscale information on domain formation and evolution at time scales orders of magnitude faster than current spatially resolved approaches. Our straightforward analysis of laser speckle patterns across the first order phase transition of VO2 can be generalized to other systems with large scale phase separation and has potential as a powerful method with both spatial and temporal resolution to study phase separation in complex materials.

No MeSH data available.


Related in: MedlinePlus