Nanoporous Gold

From an Ancient Technology to a High-Tech Material

By A. Wittstock, J. Biener, J. Erlebacher, M. Bäumer

The Royal Society of Chemistry

Copyright © 2012 Royal Society of Chemistry
All rights reserved.
ISBN: 978-1-84973-374-8

Contents

Chapter 1 Introduction to Nanoporous Gold Arne Wittstock, Jürgen Biener and Marcus Bäumer, 1,
Chapter 2 Fundamental Physics and Chemistry of Nanoporosity Evolution During Dealloying J. Erlebacher, R. C. Newman and K. Sieradzki, 11,
Chapter 3 Mechanistic Studies of Initial Dealloying Frank Uwe Renner, 30,
Chapter 4 Mechanical Properties of Nanoporous Gold Andrea M. Hodge and Thomas John Balk, 51,
Chapter 5 Microfabrication of Nanoporous Gold Oya Okman and Jeffrey W. Kysar, 69,
Chapter 6 Optical Properties and Applications of Nanoporous Metals X. Y. Lang and M. W. Chen, 97,
Chapter 7 Actuation with High-Surface-Area Materials L.-H. Shao, H.-J. Jin and J. Weissmüller, 137,
Chapter 8 Surface Chemistry and Catalysis Arne Wittstock, Jürgen Biener and Marcus Bäumer, 167,
Chapter 9 Electrocatalytical Properties of Nanoporous Gold Houyi Ma and Yi Ding, 199,
Chapter 10 Nanoporous Gold in Sensor Applications I-Wen Sun and Po-Yu Chen, 224,
Subject Index, 248,

CHAPTER 1

Introduction to Nanoporous Gold

ARNE WITTSTOCK, JÜRGEN BIENER AND MARCUS BÄUMER

1.1 Nanoporous Gold

Nanoporous gold is a corrosion-derived bulk nanostructured material. It is generated by the corrosion of an alloy of Au and a less noble metal, such as Ag or Cu. By electrochemical removal (dealloying) of the less noble constituent, the remaining gold undergoes a self-organization process forming a three-dimensional bicontinuous porous network of interconnected ligaments (Figure 1.1). Depending on the preparation conditions, the resulting pores and ligaments can be as small as 5 nm, but are typically around 30 to 40 nm. By IUPAC definition, the as-prepared material is mesoporous. Due to its high porosity and small feature size, this material has a specific surface area in the range of 10 m2 g-1. The void or pore volume in the resulting material mostly depends on the concentration of the less noble metal (e.g. Ag) in the starting compound. Because of fundamental limitations for bulk dealloying, such as the ‘parting limit’ (see Chapter 2) and the stability of the evolving porous network, alloys containing between 60 at.% and 80 at.% Ag are most viable. Processing (dealloying) of according alloys results in pore volumes between about 60% and 80%.

Early experimental work on corrosion-derived nanoporous Au by Pickering and Swann in the 1960s and by Forty in the 1970s focused on the corrosion aspect using this material and its starting alloys, respectively, as a model system for studies on the molecular mechanism of alloy corrosion. With the onset of nanotechnology in the late 1990s and the early 2000s, researchers revealed and developed the potential of this material for a variety of techno- logical aspects. As a consequence, the number of publications dealing with nanoporous Au has increased steeply by about 40% per year, from about 11 publication in the year 2001 to more than 150 in the year 2010. One of the reasons for the success of this material is the comparatively simple preparation of this nanomaterial using bench-top corrosion techniques to generate bulk samples several millimeters in size and even larger. By avoiding financially demanding techniques, such as electron beam lithography, this material became available to a variety of research groups working on the optical or mechanical properties, the catalysis or the electrochemistry of the material.

Besides the availability of the material for different research groups, another crucial factor fuelling interest in this material is its structural and chemical flexibility (see Figure 1.1). Microfabrication of the material using fast ion bombardment has been used to generate various micrometer-sized patterns and structures of interest for mechanical tests, for example. Temperature-activated ripening of the nanostructures opens the door to materials with pores and ligaments in the size regime between about 30 nm and several micrometers, without losing the typical bicontinuous structure of the material. By using templating techniques, such as slip casting of alloy-coated polystyrene beads and subsequent removal of the template, hierarchical nanoporous Au can be generated as well with relative densities as low as 2 to 3%. In addition to these structural variations, the materials surface can be chemically modified with metals, organic entities or metal oxides bringing forward its applications in electrochemistry (e.g. fuel-cell applications), sensorics, and catalysis.

Although the term ‘nanotechnology’ is rather new, the use of nanomaterials can be dated back several hundreds or even thousands of years. The first reports on the use of corrosion to generate nanoporous gold can be related to pre-Columbian civilizations, such as the Incans (see Chapter 2). Here, the superficial dealloying and subsequent burnishing of a comparably cheaper Au–Cu alloy (removal of the Cu from the alloy surface) was used to generate a shiny gold surface, giving the work piece the allure of pure gold. Undoubtedly, this must have caused severe frustration in the Spanish conquistadores when melting the looted, apparently pure, gold pieces back in Spain. However, artisans throughout the centuries have used this superficial enrichment of Au alloys as a means of gilding. The technique was thus dubbed depletion gilding or ‘mis-en-colour’, accordingly. For these reasons, when dealing with nanoporous gold, we speak of an ancient material yet with a novel technological impact.

1.2 Gold — Some Facts

Gold is an element that has inspired mankind at all times. Around 700 BC, the famous Greek poet Hesiod described the five ages of mankind in his poem Works and Days. The first age he calls the golden age of mankind, free from later gradual deterioration of moral values. Indeed, gold was the first metal recognized by humans even before bronze and iron. Traces of gold can be found in early human settlements (~8000 BC) in the Euphrat and Tigris river system, an area that is today part of the Iraq. Archeological findings of gold from later high civilizations such as Egypt and Mesopotamia can be dated back as early as 4000 BC. Back in those days, gold was already used as a means of payment, in the form of rings (about 2700 BC) and later in the form of coins (since 600 BC). The earliest craftsmanship, such as the funeral mask of the Egyptian Pharaoh Tutankhamun (1223 BC) or Solomon’s famous temple in Jerusalem (build around 950 BC), allegedly overlaid with gold, bears testimony to this early and lasting fascination with gold.

The belief in gold as the embodiment of value continued throughout the centuries. Today, the drastically increasing demand for gold as a safe investment very much reflects this fact. One reason for investing in gold as a safe haven of treasure and investment is its nobleness and obvious inability to corrode like iron. Gold stays in its metallic form, apparently unaffected by dirt and corrosion. Another reason is that gold is rare. Elements heavier than iron (56Fe) cannot be generated by fusion reactions that occur in the sun but result from neutron-capture reactions as in supernovae, a comparatively rare astrophysical event. In the galaxy, elements such as gold are thus inherently rare.

The fact that we still have unexpectedly high amounts of gold in the Earth’s crust might be one fortunate cosmic coincidence. In the process of the formation of the Earth, the iron segregated into the core and was surrounded by a silicate mantle. In this process, so-called ‘iron-loving’ metals such as gold as well as other precious metals were buried in the deep interior of our planet. The presence of unexpectedly large amounts of gold in the Earth’s crust, though, is thought to originate from meteoritic material deposited after the formation of the core mantle. Willbold and co-workers very recently provided experimental data that the presence of gold in the Earth’s crust is due to a meteor bombardment about 3.9 billion years ago. Quite literally, one could say that our gold fell from the sky.

Yet, the gold in the Earth’s crust is not evenly distributed. About 40% of all gold mined within the last 120 years has come from just one area, the Witwatersrand Basin in South Africa (Transvaal and Orange Free State). Since the average concentration of gold in the Earth’s crust is only about 2–5 parts per billion (by weight), mining of traces in many areas of the world is not economically viable. Hence, gold such as that in the Witwatersrand Basin is mined from deposits where it is largely enriched. The average gold content of the ore in these so-called conglomerate deposits is about 12 ppm. Although these numbers still sound very small, the gold in those deposits is enriched by a factor of more than 1000, and mining becomes economically viable. The explanation for this enrichment is that hot fluids (~400 °C) in the Earth’s crust are able to dissolve gold traces in the soil, yet at shear zone faults, the pressure and temperature drop, and gold precipitates, forming veins (primary deposits). Later sedimentation and compaction of sand and shingle deposits can lead to conglomerate deposits, i.e. gold deposits finely dispersed in rock.

The mining of gold strongly profits from the development of production processes to remove the gold from deposits and bringing it into a concentrated and pure form. The total production of gold up to the end of the Roman Empire is estimated to be about 10 000 t, slowing down in the medieval ages to a total of about 2000 to 3000 t. With the advent of the Californian Gold Rush in 1848, the production of gold increased steeply to about 40 t per year. However, these numbers are very small when compared to the current mining and production of gold. In 2010, the worldwide annual production (mining) of gold was 2500 t, more than 20 times the annual production from 1848. In recent years, the production of gold from South Africa, mainly from the abovementioned famous Witwatersrand Basin, has been continuously declining from 464 t in 1998 to only 190 t in 2010, making it still the fourth largest producer of gold. Today, the largest producer of gold is China, with an annual production of 345 t, followed by Australia (255 t) and the USA (230 t). Importantly, the recycling of gold has becomes increasingly relevant. In 2010, the US Department of Geological Survey reported that in the USA, 205 t of gold was generated from recycling of new and old scrap, a number that was even higher than the actual consumption. This reflects the fact that gold is a very sustainable resource.

The mining and production of gold are of considerable economic value. Already, in 1997, about 500 000 people worldwide were employed in the mining and production of gold. The value associated with gold primary production at this time was already in the range of ~$30 billion. Since then, it has increased drastically to about $100 billion in 2010 (based on an average price of gold of $40/g). Most gold is used in the form of jewellery (50%) or as a means of exchange (coins and money) and monetary assets (40%). The remaining 10% of the world’s gold production goes into the fields of industry and technology. Here, it is mostly used in the field of electronics, mainly because of its high electrical conductivity and corrosion resistivity. Another typical application of gold is its use as an ‘inert’ coating. For example, gold alloyed to silver and palladium (Pallacid) forms a very resistant coating against mineral acids, especially at higher temperatures.

Yet, gold, as a nanomaterial, is also playing an emerging and fascinating role in modern technologies such as biomedicine, water purification, fuel cells, exhaust-gas purification, energy-efficient glazing coatings, touch-sensitive screens, and even solar cells, to name just a few. At first glance, some of these applications seem to contradict the apparent inertness of gold, as they involve, for example, chemical activity at the gold surface. However, it is the unique combination of gold and its nanostructure that is opening doors to these high- tech applications. With the anticipated ongoing increase in worldwide research funding on nanotechnology to be over $12 billion per year in 2015 (according to Cientifica Ltd), gold as a nanomaterial certainly is opening up a variety of promising new perspectives also in the applied and industrial sector.

1.3 What Makes ‘Nano’ Special?

As mentioned above, a crucial factor for various applications of nanoporous gold is its nanostructure. The term ‘nano’ or ‘nanomaterial’ implies that the critical features and structures of the material are several nanometers (10-9 m) in size, typically below 100 nm. This is less than one-hundredth of the diameter of a human hair. When dealing with ‘nano’-sized materials, new material properties and applications emerge. But why, and to what extent, should the properties of a material such as gold ‘change’ when they become nanosized? One important factor is the reduced characteristic length and the increased surface-to-volume ratio of the materials. When moving from objects with characteristic length-scales (e.g. diameter, edge length) in the order of centimeters or millimeters to the nanometer scale, the surface-to-volume ratio is increased by a factor of more than a million. The proportion of surface atoms in ligaments (pillars) with diameters in the range of tens of nanometers is in the order of several percent. These hence constitute a sizable proportion of the total number of atoms in the material (Figure 1.2).

Nanosized objects are to a large extent determined by the physicochemical properties of their surfaces, as distinguished from their bulk properties. In a way, nano-sized materials bridge the realm of isolated atoms on the subnanometer scale (Å = 10-10 m) and bulk materials. Atoms at the surface have a lower number of neighboring atoms. Considering a cut through a bulk crystal, the atoms along this cut have fewer neighbors and accordingly a lower coordination number (CN). Since gold crystallizes in a face-centered cubic lattice, the CN of an atom in the bulk is 12, meaning that every gold atom is surrounded by 12 nearest neighbors. An Au atom at the surface has a reduced coordination number; depending on whether it is located on a terrace, edge, or kink site, the CN is reduced to 9, 7, and 6, respectively (Figure 1.2). In the following, effects related to surfaces and low coordination of atoms are exemplarily discussed.

One effect is that the charge redistribution due to low coordination of surface atoms gives rise to surface stress (see Chapter 7). The excess charge from unsaturated bonds is redistributed into in-plane bonds, which are strengthened or weakened, depending on whether the additional charge is distributed into bonding or antibonding states. Compared to the bond length of atoms within the bulk, the distance between surface atoms can thus be altered, leading to surface stress. For compressive stress, the surface tends to expand compared to the bulk, whereas for tensile stress, the surface tends to shrink. A typical consequence of surface stress is a reconstruction of the surface. As an example, the Au(111) surface reconstructs in the commonly known Herringbone reconstruction. The tensile surface stress of the Au(111) surface is compensated by the incorporation of about 4% additional atoms into the surface as compared to the bulk. Recently, several investigations have shown that in the case of nanoporous metals (npAu, npPt), a change in surface stress can lead to macroscopically detectable strain of the entire material, an effect formerly known only for piezo ceramics.

An additional effect is used in catalysis (see Chapters 8–10). The low coordination of surface atoms enhances the reactivity of (metal) surfaces. The desorption enthalpy of CO on gold surfaces is (among others) a function of the roughness that is the coordination number of surface atoms (Figure 1.2). Theoretically, there are two ways in which the coordination of a surface atom and the geometry of the surface, respectively, can influence the interaction with molecules and the activation barrier for the reaction; one is electronic, and the other is geometric. The adsorption and bonding of a molecule on a metal surface are determined by the electronic structure of the substrate. A concept describing the interaction (adsorption) of various molecules with transition metals was introduced by Nørskov and co-workers. In transition metals, the so-called d-bands determine the reactivity of the substrate, as they are the highest-lying electronic states. The d-band center is defined as the first moment (‘average’) of the density of d-states. The position (i.e. energy) of the d-band center with respect to the Fermi level affects the ability of a metal surface atom to form a bond with an adsorbate. For example, transition metals tend to have higher d-states in case of low coordination numbers (kink- and edge-sites). Accordingly, these atoms interact more strongly with adsorbates than atoms in a close-packed surface. Additionally, the geometry of the substrate provides specific adsorption sites that can be crucial for the activation and reaction of the adsorbed molecules. The impact of the latter geometrical effect depends on the transition state, i.e. the optimal local geometry of atoms to form and break bonds. Certainly, the electronic and the geometrical effect are hardly distinguishable, as the local geometry of a surface atom always affects its electronic structure. Yet, both effects contribute to the catalytic performance of a metal. In the case of gold, the presence of low coordinated atoms even determines whether the surface shows any catalytic activity.

(Continues…)Excerpted from Nanoporous Gold by A. Wittstock, J. Biener, J. Erlebacher, M. Bäumer. Copyright © 2012 Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.