Tailoring hierarchical nanoporous gold on dual length scales

Dealloyed nanoporous metals with a hierarchical structure provide model systems for low-density structural and functional nanomaterials. It has been suggested that these materials are distinguished by particularly stringent design principles, with precisely defined characteristic length scales, and with geometrically similar structures on each hierarchy level, and that the length scales can be independently tuned on each level. Studying nanoporous gold made by two-step dealloying, we here demonstrate the tunability of the microstructure, independently for the upper and the lower hierarchy level. Small-angle (SAXS) and ultrasmall-angle x-ray scattering (USAXS) revealed sharp interference peaks corresponding to each of the two levels, confirming the stringent structural definition. Exploiting USAXS, we resolve and study upper-hierarchy-level ligament spacings of up to 600 nm. The length scales inferred from the peak positions correlate excellently with structure sizes determined by analysis of electron micrographs. This suggests a scaling factor that allows for size conversion between the two approaches. Furthermore, the analysis of the small-angle scattering enables a characterization of the volume-specific surface area, in good agreement with the estimate based on the ligament size and the leveled-wave model as an approximate description of the material's microstructure.

Figure 1

Scanning electron micrograph of a focused-ion-beam-cut nanoporous gold sample for small-angle scattering. The center of the vertical slice is probed in the scattering experiment.

Figure 2

Optical microscope images of samples in various stages of preparation. (a) Stage (i) sample after $1\mathrm{st}$ dealloying to a composition of ${\mathrm{Ag}}_{75}{\mathrm{Au}}_{25}$; (b) stage (ii) sample after annealing at 400 ${}^{\circ }\mathrm{C}$; (c) stage (iii) sample after $2\mathrm{nd}$ dealloying with a potential of 0.80 V.

Figure 3

Scanning electron micrographs showing impact of annealing treatment on microstructures of samples in stages (ii) and (iii). Top line: Stage (ii) nanoporous silver-gold. Bottom line: Stage (iii) hierarchical nanoporous gold prepared by dealloying at potential $E_{\mathrm{D}}=0.80\mathrm{V}$. Columns: Same annealing temperature, $T_{\mathrm{A}}$. (a), (d): $T_{\mathrm{A}}=300{}^{\circ }\mathrm{C}$; (b), (e): $T_{\mathrm{A}}=400{}^{\circ }\mathrm{C}$; (c), (f): $T_{\mathrm{A}}=500{}^{\circ }\mathrm{C}$. Note that $L^{\mathsf{U}}$ is maintained during the second dealloying. Note also nearly identical lower-hierarchy level ligament size.

Figure 4

Scanning electron micrographs showing impact of second dealloying potential $E_{\mathrm{D}}$ on microstructure in stage (iii). (a) Stage (ii) nanoporous silver-gold starting material, annealing temperature $400{}^{\circ }\mathrm{C}$, and upper hierarchy level ligament size $L^{\mathsf{U}}=130$ nm, as in Fig. 3. (b) $E_{\mathrm{D}}=0.76\mathrm{V}$. (c) $E_{\mathrm{D}}=0.83\mathrm{V}$.

Figure 5

Compilation of small-angle scattering (SAXS) and ultrasmall-angle scattering (USAXS) graphs of samples with different preparation conditions. The two synthesis variables, annealing temperature $T_{\mathrm{A}}$ and dealloying potential $E_{\mathrm{D}}$, are indicated in legends. (a) Nanoporous silver-gold (NPSG) samples prepared with different $T_{\mathrm{A}}$; (b) hierarchical nanoporous gold (HNPG) samples prepared with different $T_{\mathrm{A}}$ but identical $E_{\mathrm{D}}=0.80$ V; (c) HNPG samples prepared with same $T_{\mathrm{A}}=400{}^{\circ }\mathrm{C}$ but different $E_{\mathrm{D}}$.

Figure 6

Evaluation of ligament spacings, $\tilde{L}$, by analysis of interference peak positions. (a) Exemplary experimental scattering curve (bold blue line) with fit (orange line) superimposed. The peak position according to the fit is marked by an arrow. (b) Results for samples from Figs. 5 and 5 for stage (ii) nanoporous silver-gold (NPSG) and stage (iii) hierarchical nanoporous gold (HNPG) samples. Different annealing temperature $T_{\mathrm{A}}$ and identical dealloying potential. Ligament spacings at upper (${\tilde{L}}^{\mathsf{U}}$) and lower (${\tilde{L}}^{\mathsf{L}}$) hierarchy levels determined from small-angle scattering (SAXS) and ultrasmall-angle scattering (USAXS), as indicated in legend. Note tuning of upper hierarchy level ligament size at constant size on lower level. (c) As in (b), but for samples from Fig. 5, HNPG samples with identical $T_{\mathrm{A}}$, and different dealloying potential $E_{\mathrm{D}}$. Note tuning of lower hierarchy level ligament size at constant size on upper level.

Figure 7

Comparing different approaches to metrics for the microstructure of nanoporous silver-gold (NPSG) and hierarchical nanoporous gold (HNPG). (a) Ligament spacing $\tilde{L}$ derived from the scattering data along with Eq. (1), plotted against ligament size $L_{\mathrm{SEM}}$ determined by analysis of scanning electron micrographs. Error bars indicate the standard deviation in $L_{\mathrm{SEM}}$. If both a SAXS and a USAXS measurement exist, the ligament spacing value is the average of both values. Line: Slope 1 (linear correlation) straight line of best fit, with a slope of 2.7. (b) Specific surface area ${\alpha }_{\mathrm{Porod}}$ derived from the analysis of the Porod region [Eq. (4)] in small-angle scattering versus the specific surface area, $\widehat{\alpha }$, of the leveled-wave model, using data for ligament spacing and solid fraction in Eq. (5) or (6). Line: Slope 1 (linear correlation) straight line of best fit, with a slope of 0.90.