Controllable synthesis of ultrathin gold nanomembranes

Abstract

The growth of Au nanomembranes is realized using an anisotropic hexadecylglyceryl maleate (HGM) hydrogel with a one-dimensional periodic lamellar structure. Compared with other systems, the growth of Au in the HGM hydrogel is slower, such that the formation process of such 2D films can be observed clearly. The produced Au nanomembranes possess atomically smooth surfaces and exhibit excellent optical and electrical properties.

---

Nanomembranes (NMs) are a new class of materials that possess a thickness of a few tens of nanometers across macroscopic dimensions, which are not only ultra-lightweight but also robust and flexible.¹ These types of materials have single-crystalline structures, where the minimum lateral dimension is at least two orders of magnitude larger than the thickness. Because of their two-dimensional (2D) geometries, NMs are directly applied in device designs and processing approaches of the microelectronics industry, building naturally on decades of research in thin-film growth, patterning and processing.² NMs also have finite-size and quantum characteristics in their electronic, photonic and optical properties and have unique mechanical features with effects related to shape distortions and folding of sheets, which are not found in zero- and one-dimensional materials.^3–6 At present, most of the research on 2D nanomaterials focuses on layered compounds such as graphite, which consists of weakly stacked sheets forming three-dimensional bulk crystals.^7–11 However, many other materials form atomic bonds in three dimensions (for example, noble metals), reflecting the non-layered nature of their bulk crystals. Inspired by the layered ultrathin NMs, one can also anticipate that controllable synthesis of non-layer structured NMs may give rise to some unique properties and advanced functions that cannot be achieved for their counterparts in other dimensionalities. Despite the great scientific value and application potential of non-layer structured NMs, their success has been limited to only a few materials to date.^12–19 The major challenge in exploring the properties and applications of NMs of a wider range of materials is the difficulty of their synthesis. Therefore, the strategies for bottom-up growth of NMs need to be developed in order to meet the demand for diverse applications.

Hydrogels, consisting of polymers or small organic molecules, are generally thought to be formed by three-dimensional, elastic networks whose interstitial spaces are filled with water.²⁰ Crystal growth in hydrogels is a modification of crystal growth in a solution and has been adopted as a crystal growth technique for more than one century.²¹ The unique advantage of crystal growth in hydrogels comes from the compartmentalization of the solution into small cavities within the three-dimensional porous hydrogel network. This type of solution confinement will influence the nucleation and growth mechanisms, as well as the crystal morphology. Basically, most of the gelation phenomenon arises from the formation of nano- to micrometer-scale fibres or polymer chains either physically branched or entangled with each other to trap water via surface tension. In contrast to the conventional isotropic hydrogels with a fiber/polymer chains-based network, a few anisotropic hydrogels with one-dimensional periodic lamellar structures have been discovered and have evoked great curiosity.^22–24 A lamellar biological hydrogel composed of fluid membranes of lipids and surfactants with small amounts of poly (ethylene glycol)-derived polymer lipids was first reported by C. R. Safinya et al.²² Such anisotropic hydrogel can be used as the template for the growth of NMs, because the fluid membranes of lipids or other surfactants will much more easily induce the nucleation and growth of crystals along the lamellar direction, whereas the water layer will provide the confined 2D space. These two aspects both benefit the controllable growth of NMs with tuned thickness and area.

Herein, we report the in situ synthesis of ultrathin and self-standing Au NMs with large areas and well-tuned thicknesses ranging from a few to several tens of nanometers utilizing a novel hydrogel with a lamellar structure as the template. The hydrogel was composed of hundred-nanometer-thick water layers sandwiched by lamellar bilayer membranes of a self-assembled nonionic surfactant, hexadecylglyceryl maleate (HGM). The free-standing, flexible, and large area single-crystalline Au NMs with well-tuned thicknesses were obtained by adjusting the reaction time. The optical and electrical properties of a single Au NM were measured, which matched the characteristics of a 2D nanomaterial.

The synthesis of Au NMs was carried out in a special iridescent hydrogel composed of lamellar bilayer membranes formed by HGM and water at low HGM concentration (2.0 wt% here) (Fig. 1a). The iridescent phenomenon was first investigated on iridescent layers, also known as Schiller layers.²⁵ Similar iridescence has also been found in dilute aqueous solutions of some kinds of surfactants. The surfactants form lamellar bilayer membranes, which were separated uniformly by water layers with equal distances on the order of the light wavelength. As a typical example, iridescent dodecyl glyceryl itaconate (DGI) solution made of nonionic surfactant DGI was first developed by Satoh and Tsujii in 1987.²⁶ HGM, as a derivative of DGI, could form lamellar bilayer membranes and also generate an iridescent solution at a temperature above its Krafft point (37 °C). Different from DGI, the HGM system has a unique thermal responsibility. It can be clearly seen that the color of the solution shifts from green to shinning blue gradually with the decreasing of temperature from 50 °C to room temperature (ESI, Fig. S1†). With the color change, the state of HGM system changes from liquid to gel. The dynamic rheology test on the gel sample was done by measuring the real (storage elasticity modulus G′) and imaginary (viscous loss modulus, G′′) parts of the dynamic moduli as a function of frequency. Over the frequency range, G′ is substantially greater than G′′ (Fig. 1c), revealing the typical characteristic of a gel. The reflection spectrum of HGM hydrogels shows a peak at 437 nm, indicating the formation of an ordered structure with a periodic distance in the order of sub-micrometers throughout the whole system. For the growth of Au NMs, a certain amount of chloroauric acid (HAuCl₄) was introduced into the iridescent HGM hydrogel. Reflection spectra were obtained for different reaction times. Compared with the pristine HGM hydrogel with no added HAuCl₄, the reflection peak exhibited a little shift to red due to the minor increase in the water volume. With the progression of the reaction, the reflection peak stayed the same through the whole process of Au growth, although its intensity decreased gradually. These results indicate that the periodically ordered lamellar structure of the hydrogel system could be maintained during the reaction process (Fig. 1d).

Fig. 1 (a, b) Schematic of HGM hydrogel structure and Au NMs synthesis procedure. (c) Dynamic elastic moduli of HGM hydrogel as a function of frequency. (d) Reflection spectra of a HGM hydrogel after addition of HAuCl₄ obtained at various times.

Morphology and structure characterizations of the synthesized Au NMs are shown in Fig. 2. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images showed the presence of nearly all hexagonal Au NMs with micrometer-scale edge lengths, and there are no obvious particles in the field of view (Fig. 2a and b). A high-resolution TEM (HRTEM) image taken from typical Au NMs reveals that it is a single-crystalline structure with a fringe spacing of 0.24 nm (Fig. 2d). The corresponding selected-area electron diffraction (SAED) pattern (inset in Fig. 2d) clearly shows the characteristics of the single crystalline structure of Au, and the hexagonal diffraction spots indicated it to be highly (111)-oriented. The presence of the forbidden 1/3{422} reflection indicated that the surface of the Au NMs was atomically flat.²⁷ Such 1/3{422} forbidden reflections of Au should be attributed to stacking faults. When the Au seeds contain such stacking faults, they will grow into thin plates with the top and bottom faces being {111} facets and the side surfaces being enclosed by a mix of {100} and {111} facets. Because of the six-fold symmetry of an fcc system, these seeds typically become thin plates with a hexagonal cross-section. Fig. 2c shows the X-ray diffraction (XRD) pattern of the Au NMs. The overwhelmingly preferred (111) diffraction peak located at 2θ = 38.2° and its secondary (222) diffraction peak can be clearly seen. The other two peaks, assigned to (200) and (220), are very weak compared to the (111) peak. These results also reveal a preferred orientation of Au NMs in the (111) direction.

Fig. 2 (a, b) Representative SEM and TEM images of Au NMs. (c) XRD pattern of Au NMs. (d) HRTEM image and (inset) SAED pattern of Au NMs.

It is believed that the morphologies of the products synthesized can reflect the microstructure of the template. Freeze-fracture transmission electron microscopy (FF-TEM) was employed to demonstrate the microstructures of the HGM solution and HGM hydrogel (ESI, Fig. S2†). The image of HGM solution is filled with well-ordered, parallel, and gently waving lamellae spanning a few microns (ESI, Fig. S2a†). This image is consistent with the periodically ordered lamellar structure as we expected for the iridescent solution system. As for the HGM hydrogel, the lamellae become more curved (ESI, Fig. S2b†). As discussed in our previous study, these curved structures could resist the water flow along the direction normal to the membrane layers, which is believed to be the origin of the hydrogel.²⁸ Because the fluidity is restrained, the HGM bilayer membranes are almost immobilized in the HGM hydrogel, resulting in the formation of the stable confined 2D space. Both theoretical and experimental results demonstrated that the viscosity of water increases in this confined 2D space. Therefore, the diffusion of reactive ions in the water layer is slowed down greatly and constrained, which effectively decreases the growth rate of Au, particularly in the direction vertical to the HGM bilayer membrane. Thus, it is beneficial to control the thickness and growth orientation of Au NMs. In order to reveal the advantages of the HGM hydrogel systems over other systems, a comparison experiment was executed, where HGM solution was used as a template for the growth of Au. It can be seen that the product is mainly composed of broken nanosheets and dendritic particles (ESI, Fig. S3†). This is because that the reduction of Au³⁺ in the fluid phase is much faster than that in the gel phase and it is difficult to control the anisotropic growth of Au in the HGM solution.

Compared with other systems, the growth of Au in the HGM hydrogel is slow enough to obtain plenty of information in the process of growth. Our results demonstrated that the HGM bilayer membranes here play an important role as the reduction and 2D template, which affected the formation of Au NMs effectively. Each HGM molecule contains two hydroxyl groups at the hydrophilic end that are in direct contact with the water layer in the system, which can gently reduce Au³⁺ dispersed in the water layer to Au⁰. This process also oxidizes some HGM molecules and damages the homogeneity of the HGM bilayer membrane. As seen from the reflection spectra (Fig. 1d), the intensity of the reflection peak decreased gradually with the progress of the reaction. In order to clarify the formation mechanism of the Au NMs, HRTEM was employed to observe the whole formation process, as shown in Fig. 3. At the initial stage, the nucleation is the dominant process. Au³⁺ near the HGM bilayer membrane was first reduced by the hydroxyl groups and forms small Au particles, which adhere to the lamellar bilayer membranes. It can be found that the Au particles first gathered together on 2D areas, as shown in Fig. 3a, and distinct interstitials between particles can be seen from the corresponding HRTEM image (inset in Fig. 3a). In the follow-up process, it was found that the crowded particles gradually melt to become a film, even though individual particles can still be observed, as shown in Fig. 3b. As time evolved, we found that all the particles had immersed and disappeared, as shown in Fig. 3c, resulting in the formation of an uneven 2D film. Interestingly, some ambiguous hexagonal patterns can be observed. On the edge of the hexagons, some defect-like steps can be seen and the color contrast represents the uneven thickness of the 2D film. Moreover, the hexagonal patterns become clearer and clearer in the next process, as shown in Fig. 3d. Finally, the 2D film became a rather smooth film without distinct steps or other defects, as shown in Fig. 2d. Owing to their small thickness, the films are flexible and form into a curled shape, similar to thin polymer membrane or graphene, which are the so-called Au NMs (ESI, Fig. S4†). As far as is known to us, this is the first time the clear formation process of such 2D films has been observed. The formation mechanism of the Au NMs was deduced that for the initial state, the free energy of the small nanoparticles is rather high, attributed to their high specific surface area, resulting in the thermodynamically unstable state of the particles. In order to release the total free energy of the system, a trend is observed wherein the adjacent nanoparticles integrate with each other and finally form a flat 2D film as the steady state. In this process, some of the particles will overlap with others or overlap with the interstitials, resulting in the uneven thickness of the formed 2D film, as shown in Fig. 3c and d. By comparing Fig. 3c and d, it can be found that the hexagonal patterns become larger and arrange themselves more regularly as time evolves, which allows for the release of free energy from the system. Finally, the fact that all the defects resembling steps disappeared to reduce the entropy of the system also contributed to the minimization of the free energy, resulting in the formation of the smooth 2D Au NM. Theoretically speaking, the Au NMs can have the same area as the HGM bilayer membranes, which means that the lateral dimension can reach several centimeters. However, due to the existence of gravity, the Au NM will precipitate when its lateral dimension reaches beyond the critical size (ESI, Fig. S5†).

Fig. 3 (a) TEM and (inset) HRTEM image of a dendritic Au nanostructure formed after Au particles fused with one another. (b) HRTEM of Au NMs with individual spherical particles on the surface. (c) HRTEM of Au NMs with ambiguous hexagonal patterns. (d) HRTEM of Au NMs with clearly hexagonal patterns.

The tunable thickness of the Au NMs was realized by adjusting the reaction time. In our experiment, the Au NMs with well-tuned thicknesses in the range of 3–5 nm, 5–10 nm, and 10–15 nm were synthesized (ESI, Fig. S6†). Fig. 4a shows three representative atomic force microscopy (AFM) images of Au NMs obtained with different reaction times and the corresponding height profiles. With prolonging the reaction time, the NMs become much thicker. To demonstrate the NMs' 2D characteristics, transparency and electrical conductivity properties of a single Au NM with different thicknesses were examined, as shown in Fig. 4d and e. The transmittance of the Au NMs quickly decreased with increasing thickness. For a 5 nm thick Au NM, its transparency is as high as 83% at 550 nm, which is consistent with theoretical simulation of the transmittance of a single-crystalline Au film with similar thickness based on the dielectric constant of bulk Au. Electrical conductivity was measured by drop-casting an Au NM onto gold electrodes. To eliminate the contact resistance arising from the adsorbed HGM molecule on Au membrane surfaces, the Au membrane bridged electrode was annealed at 400 °C for 4 h before measurement. Under 400 °C annealing, HGM could be removed from the Au surface, which was confirmed by thermogravimetry (TG) measurement of HGM powder (ESI, Fig. S7†). As for the Au NM, the heat impact was not strong enough to cause melting and subsequent cracking when the annealing temperature is no more than 400 °C (ESI, Fig. S8†). Maybe because of its perfect single crystalline and atomically flat surface, the electrical conductivity of the Au NMs remains the same with different thicknesses. The straight line corresponding to the resistance of 16 Ω reveals the good conductivity of such a thin Au NM. These results suggest that the ultrathin and large size Au NMs have potential applications for next-generation electronic devices.

Fig. 4 (a–c) Representative AFM images of Au NMs obtained after different reaction times and the corresponding height profile curves (scale bar 5 μm). (d) Transmittance spectra of single Au NMs with different thicknesses. (e) I–V curve of single Au NMs with different thicknesses.

In conclusion, ultrathin, free-standing and single-crystalline Au NMs with tunable thicknesses have been successfully synthesized in a unique anisotropic hydrogel. This hydrogel has a one-dimensional periodic lamellar structure with a water layer sandwiched by lamellar bilayer membranes of a self-assembled HGM molecule. The diffusion of reactive ions in the water layer was slowed down greatly and constrained in the 2D confined space, making the growth of 2D nanomaterials controllable, especially in the direction vertical to the HGM bilayer membrane. The produced Au NMs possessed atomically smooth surfaces and exhibited excellent optical and electrical properties. The effectiveness of the template directed growth of 2D nanomaterials in the unique hydrogel system has been proved. Moreover, this provides a promising route for producing various 2D nanomaterials with non-layered structures.

Acknowledgements

The authors are gratefully for the financial support from the National Natural Science Foundation of China (no. 51401107 and 21401100), the Natural Science Foundation of Jiangsu (no. BK20140384), the China Postdoctoral Science Foundation (no. 2015M571759), the Jiangsu Planned Projects for Postdoctoral Research Funds (no. 1402007A), the Fundamental Research Funds for the Central Universities (no. 30916011346) and the NUPT Scientific Foundation (NY215013).

Notes and references

1. J. A. Rogers, M. G. Lagally and R. G. Nuzzo, Nature, 2011, 477, 45–53.
2. W. Lu and C. M. Lieber, Nat. Mater., 2007, 6, 841–850.
3. B. RadisavIjevic, A. Radenovic, J. Brivio, V. Giacometti and A. Kis, Nat. Nanotechnol., 2011, 6, 147–150.
4. C. F. Zhu, Z. Y. Zeng, H. Li, F. Li, C. H. Fan and H. Zhang, J. Am. Chem. Soc., 2013, 135, 5998–6001.
5. D. Voiry, H. Yamaguchi, J. W. Li, R. Silva, D. C. B. Alves, T. Fujita, M. W. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nat. Mater., 2013, 12, 850–855.
6. M. L. Tsai, S. H. Su, J. K. Chang, D. S. Tsai, C. H. Chen, C. Wu, L. J. Li, L. J. Chen and J. H. He, ACS Nano, 2014, 8, 8317–8322.
7. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191.
8. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666–669.
9. M. Chhowalla, H. S. Shin, G. Eda, L. J. Li, K. P. Loh and H. Zhang, Nat. Mater., 2013, 5, 263–275.
10. M. S. Xu, T. Liang, M. M. Shi and H. Z. Chen, Chem. Rev., 2013, 113, 3766–3798.
11. V. Nicolosi, M. Chhowalla, M. G. Kanatzidis, M. S. Strano and J. N. Coleman, Science, 2013, 340, 1226419.
12. C. L. Tan and H. Zhang, Nat. Commun., 2015, 6, 7873.
13. C. Schliehe, B. H. Juarez, M. Pelletier, S. Jander, D. Greshnykh, M. Nagel, A. Meyer, S. Foerster, A. kornowski, C. Klinke and H. Weller, Science, 2010, 329, 550–553.
14. S. Gao, Y. Lin, X. C. Jiao, Y. F. Sun, Q. Q. Luo, W. H. Zhang, D. Q. Li, J. L. Yang and Y. Xie, Nature, 2016, 529, 68–72.
15. Z. Q. Sun, T. Liao, Y. H. Dou, S. M. Hwang, M. S. Park, L. Jiang, J. H. Kim and S. X. Dou, Nat. Commun., 2014, 5, 3813.
16. Z. Y. Tang, Z. L. Zhang, Y. Wang, S. C. Glotzer and N. A. Kotov, Science, 2006, 314, 274–278.
17. W. L. Cheng, M. J. Campolongo, S. J. Tan and D. Luo, Nanotoday, 2009, 4, 482–493.
18. Y. Chen, Z. Ouyang, M. Gu and W. L. Cheng, Adv. Mater., 2013, 25, 80–85.
19. K. J. Si, D. Sikdar, Y. Chen, F. Eftekhari, Z. Q. Xu, Y. Tang, W. Xiong, P. Z. Guo, S. Zhang, Y. R. Lu, Q. L. Bao, W. R. Zhu, M. L. Premaratne and W. L. Cheng, ACS Nano, 2014, 8, 11086–11093.
20. L. A. Estroff and A. D. Hamilton, Chem. Rev., 2004, 104, 1201–1218.
21. E. A. Smith, H. Y. Li, E. C. Keene, Z. W. Seh and L. A. Estroff, Adv. Funct. Mater., 2012, 22, 2891–2914.
22. H. E. Warriner, S. H. J. Idziak, N. L. Slack, P. Davidson and C. R. Safinya, Science, 1996, 271, 969–973.
23. M. A. Haque, G. Kamita, T. Kurokawa, K. Tsujii and J. P. Gong, Adv. Mater., 2010, 22, 5110–5114.
24. Y. J. Kang, J. J. Walish, T. Gorishnyy and E. L. Thomas, Nat. Mater., 2007, 6, 957–960.
25. P. Bergmann, P. Low-Beer and H. Z. Zocher, J. Phys. Chem. A, 1938, 181, 301–314.
26. N. Satoh and K. Tsujii, J. Phys. Chem., 1987, 91, 6629–6632.
27. C. C. Li, W. P. Cai, B. Q. Cao, F. Q. Sun, Y. Li, C. X. Kan and L. D. Zhang, Adv. Funct. Mater., 2005, 16, 83–90.
28. J. Niu, D. Wang, H. L. Qin, X. Xiong, P. L. Tan, Y. Y. Li, R. Liu, X. X. Lu, J. Wu, T. Zhang, W. H. Ni and J. Jin, Nat. Commun., 2014, 5, 3313.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra07140f

---