One-step fabrication of metal nanostructures by high-throughput imprinting

Direct nanoimprinting provides a simple and high-throughput route for producing uniform nanopatterns at great precision and at low costs. However, applying this technique to crystalline metals has been considered as impossible due to intrinsic limitation from grain size effect. Here we demonstrate direct superplastic nanoimprinting (SPNI) of crystalline metals well below their melting temperatures (Tm), generating ordered nanowire arrays with aspect ratio up to ~2000. Our investigations of replicating metal hierarchical nanostructures show the capability of imprinting features as small as 8 nm, far smaller than the grain size of bulk metals. Most surprisingly, the prepared metal hierarchical nanostructures were found possessing perfect monocrystalline structures. These findings indicate that nanoimprinting of crystalline metals below Tm might be from lattice diffusion. SPNI as a one-step and highly controlled high-throughput fabrication method, could facilitate the applications of metal nanostructures in bio-sensing, diagnostic imaging, catalysis, food industry and environmental conservation.

Controlled fabrication of metallic nanostructures plays a central role in much of modern science and technology [1][2][3] because the change in the dimensions of a nanocrystal enables tailor of its mechanic 4-6 , electronic 7-10 , optical [11][12][13][14][15][16][17] , catalytic 2,18,19 and antibacterial properties 20 . It was found the properties of metal nanostructures significantly depend on their shapes and aspect ratios 15,[20][21][22] . This has motivated an upsurge in research on the processing methods that allow better control of shape and size 20,[23][24][25][26] . Chemical synthesis of metal nanoparticles has been well developed for preparing metal nanocrystals with good quality 27,28 since its first documented by Michael Faraday 29 but suffers from the limited selection of precursor compounds and has general challenge in dispersion of synthesized nanocrystals in liquids. The current preparation of metal nanopatterns mainly relies on advanced nanolithography techniques 3 , such as nanosphere lithography 30,31 , electron beam lithography 32,33 .
These methods allow to fabricating homogeneously metallic nanopatterns but are costly owing to time-consuming, multistep processes and also limited in preparing nanostructures with low-aspect-ratios 34 . At present, high-throughput fabrication of metallic nanostructures in terms of controllability (e.g. resolution, precision, uniformity), material diversity, cost, and especially high-aspect-ratio remains a significant challenge [35][36][37] .
Among the variety of developed nanofabrication methods, nanoimprinting 38-41 , pioneered by Chou et al. 38 , promises high-throughput fabrication of ordered and regular nanopatterns at great precision and at low costs but only limited in polymers 3,40,41 and 3 a few of bulk metallic glasses (BMGs) with low supercooled liquid viscosities 39 . It has been generally considered infeasible to direct nanoimprinting of crystalline metals 35,39,42 because of the limitations on formability originating from fluctuations of plasticity on the nanoscale 43,44 , size effects in plasticity 4,45 , and grain size effect 42 .
Here, we show direct nanoimprinting variety of crystalline metals (e.g. Bi, Ag, Au, Cu, Pt) by superplastic forming well below their melting temperatures. This technique enables one-step and rapid fabrication of metallic nanostructures with high precision and good controllability, in particular it allows to fabricating high-aspect-ratio nanostructures. SPNI is usually performed under temperature ranges of 0.5Tm ≤ T < Tm, with unit of absolute scale of temperature. Typically, after SPNI, the mold is dissolved to release the replicated metal nanostructures. Fig. 1b shows a typical as-thermoplastic formed sample under an optical microscopy (OM, 9XB-PC, Shanghai optical instrument factory), where a piece of Au was superplastic formed into a nanoporous Al2O3 template at 500 ̊ C, well below the melting temperature of bulk Au (Tm ~ 1064 C). The Al2O3 template was purchased from Hefei Pu-Yuan Nano Technilogy Ltd. with an average pore diameter of 55 nm. The uniform rust-red color of the sample in Fig. 1b

Metal nanowires with controlled aspect-ratio
SPNI relies on direct mechanical deformation of imprint materials and can therefore achieve great replication precision in lateral dimensions. We now investigate the controllability of superplastic nanoimprinting along vertical direction. In general, the length of prepared metal nanowires by SPNI is a function of the forming pressure, processing temperature and time, which provides the way to control aspect ratio of We first cut five Au short rods (61.8 ± 0.5 mg) from an Au wire with diameter of 2 mm.
Subsequently, we superplastic formed the prepared five Au rods into the Al2O3 templates at 500 C, under an applied force of 10 KN and held for 1, 4, 8, 20, 60 min, respectively. The length of Au nanowires at the center of each sample were measured under SEM, which has been used to calculate the aspect ratio of prepared Au nanowires (black dots in Figure 2a). Theoretically, we can generally quantify the length of replicated nanowires (L) under conditions of applied stress (σ) and temperature (T) as where d and t are the diameter of nanopore and forming time, respectively. By applying classic Norton-Bailey's creep power law 46 , we have where the constant L0 approximates the length of metals flowing into the nanopores before load reaching the maximum force and it is in general a function of temperature, nanopore size and loading rate. R and Q are gas constant and active energy for creep, respectively. A is a constant related to nanopore size and material properties such as elastic modulus, n is the stress exponent of creep rate. In the above superplastic nanoimprinting experiments, the temperature is kept constant, the calculated mean Tab. 1), which can also be approximated as a constant. Thus the length and herein the aspect ratio of replicated nanowires simply obeys where Bi (i = 1, 2) are constants related to nanopore size, material properties, processing temperature and forming pressure. Eq. (3) indicates that the aspect ratio of prepared metal nanowires continues to increase as the forming time increases, which can be adopted to fabricate high-aspect-ratio metallic nanostructures. This is very different with nanoimprinting of BMGs, where the processing time above glass transition temperature (Tg) is limited by the crystallization time due to the metastable nature of BMGs 39 . In addition to varying holding time to control the length of replicated Au nanowires, we also independently varied the processing temperature to study its effect on the aspect ratio of prepared Au nanowires (Extended Data Fig. 1). In this series of experiments, we firstly prepressed four Au short rods (29.5 ± 0.5 mg) at 415 C to get flat discs with thickness of 0.35 ± 0.05 mm. Subsequently we superplastic formed the Au flat discs into 100 nm Al2O3 templates at 415, 450, 496, 722 C, respectively, where all the samples were loaded to 3 KN and held for 100 s. The measured length of Au nanowires at the center of each sample is shown in Extended Data Fig. 1. It is clear that the length of nanowires increases with the rising of temperature, which can be understood from the Arrhenius-type temperature relation in eq. (2), the rising of temperature will increase the activity of atoms and/or defects.

Replication of metal nanoarchitectures and TEM characterization
At present, fabrication of metal nanostructures smaller than 10 nm is very challenging, even with advanced nanolithography techniques. To demonstrate the powerful of our 9 SPNI technique to replicate extremely small features, we adopted a smallest nanomold we could geta hierarchical Al2O3 template with multiply branched nanopores in its surface layer, where the branched nanopore size gradually increases from ~ 8 nm in the outmost surface to 200 nm in the base layer (Extended Data Fig. 2). Figure 3a shows SEM images of replicated Au hierarchical nanostructures by using the hierarchical Al2O3 template (see Methods section). The aggregation of Au nanostructures into bundles makes it difficult to examine the replicated hierarchical structures. We therefore transferred the prepared Au nanostructures onto a transmission electron microscope (TEM) mesh grid (see Methods section). SEM imaging of Au hierarchical nanostructures on the TEM mesh grid clearly shows the primary stems abruptly multiplied by several small branches (Fig. 3b). The replicated smallest branch is ~ 8 nm (Extended Data Fig. 3), in consistent with the hierarchical Al2O3 template we used, indicating the high replicating fidelity of our SPNI technique.  (Figure 4). Fig. 4c shows a typical diffraction pattern for the selected Au nanostructure in Fig. 4b, which exhibits symmetrical and clear electron diffraction spots for single face-centered cubic (fcc) crystal. The growing axis of the single crystal is determined along <111> crystallographic orientation. High-resolution TEM images (Fig. 4d-g) and fast Fourier transformations (insets of Fig. 4d-g) at the selected regions denoted by A, B, C, D in the nanostructure in Fig. 4b confirm its perfect single-crystallinity again. Most surprisingly, we observed that even around the regions where the branches abruptly converged (Fig. 4e and f), the crystallinity is not disturbed at all. Among the typical creep deformation mechanisms of viscous flow, dislocation motion, lattice or grain boundary diffusion, dislocation motion is attributed to the main creep mechanism of crystalline metals at high temperatures, i.g. T > 0.4Tm 47 . In view of the fact that most of metals possess grain sizes larger than 1 μm, direct nanoimprinting of crystalline metals below Tm has been considered as impossible 35,39,42 . However, we noted that both viscous flow and lattice diffusion are independent of grains, which provides the possible mechanisms for direct nanoimprinting of metals. Considering that our investigations of the effect of holding time on length of replicated nanowires has ruled out the viscous flow dominated mechanism as discussed before (Fig. 2), it is quite possible that the deformation mechanism of superplastic nanoimprinting crystalline metals below Tm originates from lattice diffusion.

Fabricating variety of metal nanostructures by SPNI
Our SPNI technique includes but not limited to fabricating Au nanostructures. To show the generalization of this SPNI technique for preparing crystalline metal nanostructures, we also fabricated Bi, Ag, Cu, and Pt nanowire arrays ( Figure 5). Fig. 5a shows the fabricated high-aspect-ratio Bi nanowire arrays with aspect ratio of AR ~ 300, which is achieved by superplastic nanoimprinting a piece of Bi at 260 C, closing to its melting temperature (Tm ~ 273 C). We observed that Bi can completely fill the template within can also be nanoimprinted at ~ 820 C by using our method (Fig. 5d)

Surface-enhanced Raman scattering from Au nanowires
Reproducible and robust metal nanostructures that strongly enhance the electromagnetic field are most desirable for surface-enhanced Raman scattering (SERS) 14 but are difficult to achieve 36,48 . We have shown above that direct thermoplastic nanoimprinting of crystalline metals offers a rapid and controllable method to fabricate uniform metallic nanostructures, which provides an ideal way to fabricate robust SERS substrates. We now explore how effective thermoplastic nanoimprinted metal nanowire arrays as SERS substrates. As examples, we prepared five Au flat discs, four of them were imprinted with Au nanowire arrays by SPNI and the remaining one is used as a reference sample (see Methods section). For simplicity, we denote the reference sample as bulk Au and the four samples attached with Au nanowire arrays as s1, s2, s3 and s4, corresponding to nanowire size of 200, 90, 55, 20 nm, respectively. The optical micrographs of the as-thermoplastic formed Au/Al2O3 template combinations show clearly size-dependent colors (Extended Data Fig. 6a-d): the surface color changes from pinkish (s1), rust red (s2) to olive-green (s3) and finally tends to grey (s4). Such a absorption redshift is well understood from the size-dependent surface plasmon oscillation 21 and it indicates the effectiveness of the prepared samples as SERS substrates.
Extended Data Fig. 6i shows measured Raman signals at the center of the prepared five samples by using 1.0×10 -5 M crystal violet (CV) as a sensitive SERS analyte to detect the electromagnetic enhancement (RENISHAW Raman microscope, INVIA, see Methods section). Although CV molecular absorbed on the bulk Au shows rather weak Raman signals, five Raman lines located at 442 cm -1 , 802 cm -1 , 1174 cm -1 , 1384 cm -1 and 1620 cm -1 can still be recognized and these Raman shifts agree well with literatures 15 report (Extended Data Tab. 2). On the contrary, almost all of the reported Raman lines for CV between 400 and 1800 cm -1 are drastically intensified by samples s1-s4 (Extended Data Fig. 6i and Extended Data Tab. 2), which demonstrates that our SPNI technique is very suitable to fabricate metallic SERS substrates. Extended Data Fig. 6i also demonstrates that the Raman shifts decrease as the size of nanowires increased from 90, 200 nm to infinity (bulk Au). Such a size dependent SERS signals is in good agreement with theoretical prediction 21,22 . However, when the size of nanowires continues to decrease below 90 nm, the Raman signals become slightly weaker rather than continuous increasing, which we attribute to the aggregation of small nanowires to form big bundles due to mechanical instability (Extended Data Fig. 6g-h).

Conclusions
Direct superplastic nanoimprinting of crystalline metals well below Tm has enabled us to fabricate variety of metal nanowires with aspect ratio up to ~ 2000. By adopting a hierarchical Al2O3 template with gradient nanopore size, we are able to replicate Au hierarchical nanostructures. The perfect monocrystalline structure of the prepared Au hierarchical nanostructures, together with the fact that the nanopore size at the entrance of the Al2O3 template is only ~8 nm, far smaller than the grain size of the bulk Au, we argue that direct superplastic nanoimprinting of crystalline metals might originate from lattice diffusion dominated mechanism. Finally, we show our SPNI technique is very suitable to fabricate metallic substrates for SERS application. SPNI is inherently highthroughput due to its parallel printing, and it requires only one step and simple equipment set-up, leading to low-cost. We propose our technique should facilitate the applications of metal nanostructures in catalysis, nanoelectronics, sensors and plasmonics.