X-ray standing wave technique with spatial resolution: In-plane characterization of surfaces and interfaces by full-field x-ray fluorescence imaging

The present paper describes a powerful extension of the x-ray standing-wave technique that combines it with the capability of full-field x-ray fluorescence imaging, so that the impurity depth profiles at all different parts of a nanolayer material can be measured in parallel. The imaging capability improves the robustness and reliability of the x-ray standing-wave technique in investigating inhomogeneous two-dimensional materials and soft interfaces. It can also visualize the three-dimensional element-specific structure at a buried interface with a depth resolution at the nanometer level.

Figure 1

(a) Conventional XSW analysis. It measures the full-field XRF intensity profile $Y\left(\theta \right)$ and analyzes the average depth distribution $\rho \left(z\right)$ of the impurity over the entire probing area. (b) XSW imaging technique. It measures $Y_{x,y}\left(\theta \right)$ independently at every $(x,y)$ point and reveals their respective impurity depth distribution ${\rho }_{x,y}\left(z\right)$.

Figure 2

XRR pattern of the nickel/carbon periodic multilayer.

Figure 3

Examples of the XSW imaging experiment data. (a) Full-field XRF spectra measured over the entire probing area. (b) An XRF image of Fe $\mathrm{K}\alpha$ taken at $\mathrm{\Delta }\theta =+0.{02}^{\circ }$.

Figure 4

XRF intensity profile of Si $\mathrm{K}\alpha ,\beta$ from the silicon substrate. (a) Full-field profiles of the entire probing area, measured by either the Si-PIN detector or the pinhole camera. The experimental profiles match well with the simulated profile the vertical contrast of which has been reduced by half. (b) Microregion profiles of C3, D4, E5, and F6, respectively. All of these microregions are along the diagonal of the image.

Figure 5

XRF intensity profiles of Fe $\mathrm{K}\alpha$ from the iron impurity. (a) Full-field profiles of the entire probing area, measured by either the Si-PIN detector or the pinhole camera. The experimental profiles match well with the simulated profile when the depth of the iron impurity is assumed to be at the center of each nickel layer $\left(Z=7\AA \right)$ and the vertical contrast of the simulated profile has been reduced to 20%. (b) Microregion profiles of C3, D4, E5, and F6, respectively. All of these microregions are along the diagonal of the image.

Figure 6

Simulation of the iron impurity depth distribution and the corresponding Fe $\mathrm{K}\alpha$ intensity profile. (a) Model of the iron impurity depth distribution. (b) Simulated profiles. The shape of the profiles changes greatly as the depth $Z$ of the iron impurity moves from 0 to $49\AA$.

Figure 7

Illustration of how to convert an XRF intensity profile into a feature vector. (a) The simulated profile corresponds to the model in which the depth of the iron impurity is at the center of every nickel layer $\left(Z=7\AA \right)$. In this case, $S_1<0$ and $S_3>0$. (b) The simulated profile corresponds to the model in which the depth of the impurity is at the center of every carbon layer $\left(Z=32\AA \right)$. In this case, $S_1>0$ and $S_3<0$.

Figure 8

Feature map for Fe $\mathrm{K}\alpha$ intensity profiles. (a) The feature points corresponding to all the simulated profiles. In the simulation model, the center of the iron impurity in the depth direction moves from $Z=0$ to $49\AA$. (b) The two feature points corresponding to the full-field intensity profiles of the entire probing area, measured by the pinhole camera and the Si-PIN detector, respectively. (c) The four feature points corresponding to the microregion profiles of C3, D4, E5, and F6, respectively.