Fundamentals of Inorganic Glasses

By Arun K. Varshneya and John C. Mauro

Fundamentals of Inorganic Glasses, Third Edition, is a comprehensive reference on the field of glass science and engineering that covers numerous, significant advances. This new edition includes the most recent advances in glass physics and chemistry, also discussing groundbreaking applications of glassy materials. It is suitable for upper level glass science courses and professional glass scientists and engineers at industrial and government labs. Fundamental concepts, chapter-ending problem sets, an emphasis on key ideas, and timely notes on suggested readings are all included. The book provides the breadth required of a comprehensive reference, offering coverage of the composition, structure and properties of inorganic glasses.

• Clearly develops fundamental concepts and the basics of glass science and glass chemistry
• Provides a comprehensive discussion of the composition, structure and properties of inorganic glasses
• Features a discussion of the emerging applications of glass, including applications in energy, environment, pharmaceuticals, and more
• Concludes chapters with problem sets and suggested readings to facilitate self-study

• Language: English
• Publisher: Elsevier Science
• Release date: May 9, 2019
• ISBN: 9780128162262

Arun K. Varshneya

Dr. Arun K. Varshneya is Professor Emeritus of Glass Science and Engineering at Alfred University. Prior to joining the faculty at Alfred’s New York State College of Ceramics in 1982, he worked as a senior scientist for Ford and General Electric Lighting Business Group. Arun is currently president of Saxon Glass Technologies Inc. which specializes in strengthening glass for various industries. He is the invited author of the “Industrial Glass” entry in Encyclopedia Britannica and has more than 160 technical publications covering a broad range of topics in glass. Arun is a Distinguished Life Member of the American Ceramics Society, an Honorary Fellow of the Society of Glass Technology, and a recipient of the President’s Award from the International Commission on Glass for lifetime achievements. Varshneya earned a B.Sc. from Agra University in India, a B.Sc. with Honors in glass technology from the University of Sheffield (U.K.), and M.S. and Ph.D. degrees in materials science, both from Case Western Reserve University in Cleveland, Ohio

Book preview

examples.

Chapter 1

Introduction

Abstract

The word glass is derived from a late-Latin term glæsum to mean a lustrous and transparent material. Glassy substances are also called vitreous, originating from Latin word vitrum (clear). The history of glass as glazed stone beads goes back perhaps as much as 12,000 years. As independent objects, glassware was available ~ 5000 years ago. The most important technological developments in glass, perhaps as the glass window, were sponsored by the Christian Church during the Middle Ages on the European continent. Although transparency, luster, and durability against elements of nature are neither sufficient nor necessary to describe glass, they remain some of the key characteristics of glass that are important to large-scale commercialization. More than 95% of the commercial tonnage is oxide glasses, of which the vast majority is silica based. Vitreous silica, soda lime silicate, borosilicate, lead silicate, aluminosilicate, and optical glasses are the primary glass families. Of the nonoxide glasses, those of significant commercial interest are the heavy metal fluoride glasses (HMFG), the amorphous semiconductor and chalcogenide group, and glassy metals. Of these, the amorphous semiconductors and the chalcogenides form the basis of miniaturization of the computer as switching and memory devices, solar cell (photovoltaics), and the xerographic process (photoconductivity). Glass is also found in nature. The more important and interesting examples are volcanic glass (obsidians), lunar glass, and tektites (generally thought to be fused ejecta from a meteorite impact).

Keywords

Glass; Silica; Natural glass

1.1 Brief history

The word glass is derived from a late-Latin term glæsum used to refer to a lustrous and transparent material. Other words often used to refer to glassy substances are vitreous, originating from the Latin word vitrum (= transparent or clear), and amorphous, originating from Greek amorphe (= without form or shape). Near-transparency, luster, or shine, and in particular, its durability when exposed to the elements of nature, were probably the most significant properties of glass recognized by early civilizations. Glazed stone beads from Egypt date back to 12,000 BC. Several of the artifacts unearthed from the tombs of the pharaohs exhibit excellent glass inlay work in a variety of colors. As independent objects, glassware perhaps existed for roughly 5000–6000 years. The technology of glass windows, exploiting the property of optical transparency, had developed around the birth of Christ, and was developed to new heights of artistry by the Christian Church during the Middle Ages. Many of these beautifully stained windows, which can still be viewed in a number of churches over the European continent, show the deep commitment of the church to preserve the history of mankind and religious teachings through the medium of glass.

Many of the uses of glass in the modern-day world continue to exploit the transparency, luster, and durability of glass. Containers, windows, lighting, insulation, fiber, stemware, and other handcrafted art objects are typical of these traditional uses. At this point, it is worth noting that for a material to be used in a product it must have certain desirable properties that determine its use. In our later discussions, it will become clear that the properties of transparency, luster, and durability are neither sufficient nor necessary to describe glass. Similarly, being amorphous does not have the same meaning as being a glass. Through the application of basic sciences to the study of glass, newer properties of glasses have been developed, and hence, newer products have been conceived.

As may be expected, much of glass science developed based on major commercial uses of glass. More than 95% of the commercial tonnage of glass consists of oxide compositions. By far the largest percentage of these is silica-based. This includes both commodity glass products and highly specialized applications of glass, such as in microelectronic packaging, where the annual volume of sale may be low but glass is a key value-adding component, that is, the application of glass is either a critical component or enhances the value of the assembly after the incorporating process. It is not surprising that when the term glass is used in scientific conversation, oxide glasses are usually implied. Over the past few decades, however, a great many studies of nonoxide glasses have been triggered by the possibility of some exotic uses of glass in the fields of healthcare and information technology. It is well, therefore, to review our thoughts on the various families of glasses, their compositions, and their uses before we delve into the detailed science of glass. Inorganic glass families and their composition ranges developed through the 1970s have been extensively discussed by Kreidl [1], Rawson [2], and Vogel [3].

1.2 Glass families of interest

Table 1.1 summarizes the chemical composition of several common commercially available inorganic glass products. All of the glasses listed here are silica-based. Many of these are reclassified in Table 1.2 into five major composition families showing the key properties responsible for the commercial interest.

Table 1.1

Table 1.2

One may note that, besides silica, other major constituents (amount generally greater than ~ 3%) are alkali oxides, alkaline earth oxides, alumina, boric oxide, and lead oxide. Many of the same oxides may also be present as minor constituents. Compounds such as arsenic oxide, tin oxide, and colorants are often intentionally added in trace amounts (less than ~ 0.3%). Not shown are the presence of trace impurities such as iron oxide and hydroxyl, which are common contaminants in the raw materials used in the glass melting process. The reasons for many of these additions of different components will become clear as we continue. In more recent years, the use of toxic compounds such as lead oxide and arsenic has come under pressure due to concerns over worker safety and environmental protection. Several of the various glass families are discussed below.

1.3 Vitreous silica

Vitreous silica is the most refractory glass in commercial usage. In addition to its ability to withstand exceptionally high temperatures, it has a high resistance to chemical corrosion (particularly to acids), a very low electrical conductivity, a very low (~ 5.5 × 10− 7/°C) coefficient of thermal expansion (CTE), and good ultraviolet (UV) transparency. However, because of the high cost of manufacture, the uses of vitreous silica are mostly limited to astronomical mirrors, optical fibers, crucibles for melting high-purity silicon, and high-efficacy lamp envelopes. In one technique, the glass is obtained by melting high-purity quartz crystals or beneficiated sand at temperatures in excess of 2000°C. In a second technique, SiCl4 is sprayed into an oxy-hydrogen flame or water vapor-free oxygen plasma, reacting to form SiO2 vapors, which deposit on a substrate and are subsequently consolidated at ~ 1800°C.

1.4 Soda lime silicate glass

Soda lime silicate glass is the most widely used of all commercial glasses. Most beverage containers, glass windows, and incandescent and fluorescent lamp envelopes are made from soda lime silicate owing to its low cost, good chemical durability, and high optical transmission in the visible region. Because of its relatively high CTE (~ 95 × 10− 7/°C), it is prone to thermal shock failure, and this prevents its use in a number of applications. Large-scale continuous melting of inexpensive batch materials such as soda ash (Na2CO3), limestone (CaCO3), and sand (SiO2) at 1400–1500°C makes it possible to form the products inexpensively and with high throughput.

1.5 Borosilicate glass

Small amounts of alkali added to silica and boron oxide yield a family of glasses commonly utilized for their low thermal expansion coefficient (~ 50 × 10− 7/°C) and especially high resistance to chemical attack. Laboratory glassware, pharmaceutical glassware, household cookware, and automobile headlamps are prime examples of borosilicate glass usage. A more important application is the use of a Type I borosilicate glass cartridge to contain epinephrine in autoinjector devices for emergency antidote to prevent anaphylactic shock from severe allergies to bee stings, peanuts, and shell foods, which otherwise could potentially lead to death. The glass cartridge is chemically strengthened to reduce glass fracture probability due to applied force during emergency administration. Borosilicate glasses can be made commercially in a manner similar to soda lime glasses but require higher temperatures for melting (~ 1550–1650°C). The higher melting temperatures and higher cost of B2O3 raw materials make borosilicate glasses much less competitive compared to soda lime silicate for common products.

1.6 Lead silicate glass

This family of glasses contains PbO and SiO2 as the principal components, with small amounts of soda or potash. These glasses are utilized for their high degree of brilliance (as stemware or the misleadingly named lead crystal), large working range (useful to make art objects and intricate shapes without the need to reheat the glass), and a high electrical resistivity (e.g., for electrical feedthrough components). PbO additions increase the fluidity of glass and its wettability to oxide ceramics. Hence, high lead borosilicate glasses (generally without any alkali additions) are used extensively in microelectronics (e.g., for conductor, resistor, and dielectric pastes). Because of the toxicity of lead, alternate formulations for a variety of products are being explored.

1.7 Aluminosilicate glass

The electrical resistance of the alkali-free alkaline earth aluminosilicates often approaches that of vitreous silica. The most important 21st-century application of the alkali-free aluminosilicates is in liquid crystal display flat thin glass sheets. Many of the glasses are intermediates between the soda lime silicate and vitreous silica for refractoriness and thermal expansion (~ 30–50 × 10− 7/°C). Another commercial use of these characteristics is in the high-efficacy lamps involving tungsten-halogen cycle (for instance, the lamp inside an automobile halogen headlamp) where the glass can be sealed directly to molybdenum electrical leads. Aluminosilicates generally have high values of elastic moduli and a high resistance to chemical corrosion. Compositions commonly referred to as E-glass and S-glass are used as the load-bearing fiber component in fiber-reinforced plastics.

Alkali-containing aluminosilicate glasses form the basis for a number of ultrahigh strength glass products where the strengthening is achieved chemically using an ion exchange process. Through the ion exchange process, a compressive stress layer of around 1 GPa can be developed at the glass surface to protect against the introduction and propagation of damage. Chemically strengthened aluminosilicate glasses are commonly used as aircraft windshields and as thin display covers for smartphones and other personal electronic devices.

1.8 Bioactive glasses

Bioactive glasses, for example, glasses that stimulate some desired effect in the human body, were originally discovered in the 1960s. Bioactive glasses include a variety of silicate and borate compositions that are designed to be surface reactive when implanted in the body [4]. The surface reaction contributes the needed chemicals to stimulate healing of hard or soft tissues in the body. For example, bioactive glasses are now used to promote regeneration of bone tissue or the healing of flesh wounds. Bioactive glasses will be covered in detail in Section 23.8.

1.9 Other silica-based oxide glasses

Many other families of silica-based glasses exist, including optical glasses used in optical components and devices, such as ophthalmics. The inclusion of BaO, ZnO, La2O3, Nd2O3, or other oxides, often as major constituents, in soda lime silicate, lead silicate, or borosilicate glasses yields compositions with useful refractive index and optical dispersion properties.

1.10 Other nonsilica-based oxide glasses

Oxide glasses not having silica as a principal component have significantly less commercial usage. B2O3-based and P2O5-based glasses are readily attacked by water. However, their studies have been extremely important toward enhancing our understanding of glass structure. Some nonsilicate oxide glasses that have commercial interest include the alkaline earth aluminophosphates (for laser hosting), boroaluminates (e.g., cabal glasses with electrical resistivities exceeding that of silica), alkaline earth aluminates (as a high-temperature sealant and infrared (IR)-transmitting glass), low-melting temperature V2O5-based glasses, and bioactive B2O3-based glasses for soft tissue repair. Many of these glasses have very low liquidus temperatures and are quite fluid compared to silicate glasses. Hence, some of these glasses have found use in sealing of electronic components. Moreover, the especially high refractive index of tellurite glasses (in excess of 2.0) makes them useful in optical systems.

1.11 Halide glasses

Because of high solubility in water, simple halide glasses such as BeF2, ZnCl2, and their mixtures with alkali halides have little commercial interest despite the fact that their ready glass formability has been known since the 1920s. Of the known halide glasses, those based entirely on heavy metal inorganic fluorides have attracted the most attention during the 1980s. It has been shown that the theoretical attenuation in these heavy metal fluoride glasses (HMFGs) can be as little as 10− 3 dB/km at around the 3.5 μm wavelength. [Note: 1 dB = − 4.343 ln (I/I0) where I0 and I are the incident and the exiting intensities, respectively.] For a 6000 km propagation distance, a light signal would suffer a 6 dB loss, and so, a quarter of the original intensity would still be transmitted. This indicates the possibility of utilizing HMFGs for repeaterless, continuous glass fiber cable to carry transcontinental or transoceanic telecommunications. Such repeaterless communications would not be possible in oxide glasses, for instance, in silica glass the attenuation minimum is shown to be not less than about 0.15 dB/km (at 1.55 μm wavelength).

The most studied composition in the HMFG family is the ZBLA glass, which comprises 57ZrF4·36BaF2·4LaF3·3AlF3 (mol%). Improvements have been sought by changing various constituents or by substituting, for instance, LaF3 by YF3 (the ZBYA glass) or by adding varying amounts of other constituents such as fluorides, chlorides, and even small concentrations of oxides. In almost all cases, the starting raw materials (generally fluorides) must have better than 5–6 N (99.999%–99.9999%) purity. They are melted in nonreactive crucibles typically made of platinum or vitreous carbon at 800–1000°C. A reactive atmosphere, such as CCl4, SF6, or NF3, is maintained during the melting to remove oxygen and OH− impurities which otherwise would degrade the transmission properties to the point of rendering the glass useless.

The HMF glasses, in general, are extremely prone to crystallization. As a result, despite 30 years of intense research, researchers are yet to report glasses better than ~ 0.2 dB/km loss. Because these glasses can also be readily attacked by water, their practicality in telecommunication technology relative to that of the vitreous silica fibers is no longer a viability. It is likely that the application of HMFGs will be limited to near- to mid-IR transmitting short-haul sensors.

1.12 Amorphous semiconductors

Elements such as Si, Ge, P, As, compounds such as CdGexAs2 (x = 0 to 1.2, a family AIIBIVCV2 called tetrahedral glasses) and Si1 − xHx (where x = 0.1–0.2), and mixtures of Si, Ge, As, Sb, etc., with S, Se, and Te (chalcogen-based) retain their semiconducting behavior (observed in the crystalline state) even in the noncrystalline form. It was discovered in 1960s that many of these noncrystalline materials displayed switching between high and low electrical conductivity states (while remaining semiconducting), potentially making them the best possible choice for computer memory [5]. The enthusiasm decayed considerably after it was realized that the switching occurred as a result of transition between localized melting and the onset of crystallization, a phenomenon that could not be fully controlled. The switching could be carried out optically using short laser pulses. Devices based on electrical and optical pulse switching behavior are in commercial production for computer memories. Amorphous semiconductors are also used in photovoltaics, enabling solar cell technology with lower cost compared to crystalline silicon (c-Si). Amorphous silicon (a-Si) can be produced more inexpensively in the form of thin films by vapor deposition on cold substrates. (Useful abbreviations are a-, g-, and c-for the amorphous, glassy, and the crystalline states, respectively.) The hydrogenation of a-Si (by glow discharge plasma decomposition of SiH4 silane) apparently cleans up the poisoning defect sites in a-Si and, thus, produces a-Si1 − xHx (x = 0.1–0.2), which is a more efficient photoreceptor for use in the solar cells. As indicated in Section 2.1, amorphous solids such as a-Si are excluded from being classified as glasses.

1.13 Chalcogenide and chalcohalide glasses

Glasses obtained by melting chalcogen elements (group 16: S, Se, and Te) with one or more of groups 15 and 14 elements are called chalcogenide glasses. Compositions modified by adding halogens are called chalcohalides. The primary interest in these glasses comes from their semiconducting (switching) behavior, photoconductivity, and IR-transmitting properties. The photoconductivity property is utilized in xerography (photocopying). Interest has also been high in the more recent years upon the recognition that, although these glasses generally appear gray/black and are opaque in ordinary visible light, they begin to transmit from near-IR to as far as the 18–20 μm wavelengths. This then makes them a potential fiber material candidate for transporting CO2 laser wavelength (10.6 μm) in such applications as laser-assisted microsurgery. Chalcogenide glass fibers could also be used as fiber amplifiers based on stimulated Raman scattering and their high optical nonlinearity and wide range of tunability. Fiber amplifiers stand to share some of the revolution in telecommunication of optical signals. Instead of converting light to electricity, amplifying, and then converting back to light for onward transmission, the optical fiber amplifiers amplify light signals directly, thereby reducing conversion losses.

Compounds of sulfur, for instance, with boron, alkalis, and silver, have found potential application as fast ion conducting glasses. The compound B2S3 is an analog of B2O3 where the oxygen has been replaced by sulfur. As a result, an entire group of oxygen-free thioborate glasses based on B2S3 as the primary glass constituent is now being developed [6].

Like the halides, the chalcogenides also require O2- and OH-free melting conditions to assure good transmission behavior. Raw materials (pure elements) must be distilled or sublimed to increase purity level to 6N. Glasses are generally melted in evacuated silica glass ampoules at 800–1000°C inside a rocking furnace to achieve a high degree of homogeneity. Chalcogenide glasses are much more resistant to water attack compared to halide glasses. For this reason, it is possible that chalcogenides, rather than the halides, may be the material of choice for mid-IR sensor applications.

1.14 Metallic glasses

Metallic glasses (sometimes also referred to as glassy metals or, inappropriately, as amorphous metals) are noncrystalline materials composed of either pure metals or combinations of metals and metalloids. They are metals in the sense that their electrical, magnetic, and optical properties are typical of metals. (However, their electrical resistivity often decreases with temperature.) They are neither semiconductors nor are they optically transparent. An example of a glassy metal is Fe40Ni40P14B6 sold under the trademark Metglas.

These glasses are often made in the form of thin tapes or fibers using very high-speed quenching techniques, for instance, by passing liquid streams between high thermal diffusivity rollers or by melt spinning. The cooling rates are generally in the order of 10⁵–10⁸ °C/s. In pulsed laser quenching, a ~ 100 nm thin surface layer of a metal is melted by an incident picosecond pulsed laser beam. During the off portion of the pulse, the large cold substrate causes the melted layer to cool at rates exceeding 10¹² °C/s. Amorphization of some metals is also possible by mechanical methods such as ball milling. As with a-Si, there are lingering questions as to whether some of these substances should be called amorphous metals as opposed to metallic glasses [7].

Because metallic glasses are readily produced as thin ribbons, the primary commercial use of amorphous Fe-based ferromagnets is in flexible magnetic shielding and power transformer core laminations. These materials have very low magnetic hysteresis curve losses. In addition, they have about three times higher electrical resistivity than their crystalline counterparts. As a result, their use as power transformer core laminations can lead to as much as 30% power savings. Metallic glasses have extremely high mechanical strengths, approaching theoretical values. Hence, as fiber-reinforced composites, they could be candidate materials for high strength-to-weight ratio applications (such as high-speed aircraft, space vehicles, etc.). There has been some interest in their superconducting properties for applications in thermonuclear fusion superconducting magnetic torus, primarily because their amorphous nature makes them less susceptible to radiation damage. However, the superconducting transition temperatures are quite low (about 9 K). Because of their unique magnetostrictive characteristics, they are potential candidates for application in remote sensing, product labeling, and electromagnetic shielding. More recently, new compositions with improved glass-forming ability have been developed, enabling the formation of bulk metallic glasses (BMGs) with three-dimensional (3D) morphologies [8].

1.15 Glass-like carbon

Controlled carbonization of water-dispersed cellulose or of slow-aged thermosetting resins such as furfuryl alcohol and phenolformaldehyde at temperatures between 1200°C and 3000°C yields a nongraphitizable black mass which is called glass-like carbon and sold under trade names such as Cellulose Carbon, Glassy Carbon, and Vitreous Carbon. In addition to the properties common to carbon, such as high electrical and thermal conductivities, glass-like carbons have high strength, high hardness, high resistance to oxidation and reaction with chemicals even at high temperatures, and very low permeability to gases despite the presence of finely distributed porosity [9]. Because of the many desirable physical properties, glass-like carbon is used mostly as a crucible material for high-purity melting and as high-temperature mandrels and jigs. In reality, the only resemblance glass-like carbon has to glass is its shiny, conchoidal surfaces when broken. It is best classified as an agranular, macro-isotropic, microcrystalline solid and, hence, will not be included in our discussions henceforth.

1.16 Mixed anion glasses

As the name implies, these are hybrid families where the oxygen is substituted in part by halogen (usually F and Cl), N, or C. Many are potentially high-performance materials but remained in research laboratories so far. Stable glasses containing small amounts of the substituent (generally ~ 1%–4%) may be prepared by melting together halides, nitrides, or carbides with oxides over narrow range of compositions. The melt viscosities and the electrical resistivities of the oxyhalides are generally lower than those of the oxides. Caution must, however, be exercised when using halogen containing glasses in picoelectronic packaging because of the potential to form corrosive HF and HCl from reaction with atmospheric water over long periods. Bulk oxynitride glasses containing silicates, phosphates, and 5–18 at.% nitrogen can be prepared by ammoniating the oxide melts. Oxynitride glasses can also be prepared [10] by the direct melting of common glass-making oxides along with AlN or Si3N4 under N2/Ar atmosphere at 1500–1800°C. Oxycarbide glasses with high (20%–40%) carbon content can only be prepared currently in the form of thin specimen forms by a sol-gel route [11]. These glasses may, in fact, be nanoscale composites of carbon and oxide glasses giving them a black appearance. Nitriding and carbiding of glasses enhances the high-temperature mechanical and rheological properties greatly; hence, these are of considerable commercial interest.

1.17 Metal-organic framework glasses

The newest family of glass-forming chemistry comes from a hybrid organic-inorganic class of material known as metal-organic frameworks, which consist of metal cations linked by organic ligand molecules. In their crystalline form, metal-organic frameworks are known for their high degree of porosity. Very recently, it has been discovered that some of these metal-organic framework crystals can be melted into a stable liquid and then quenched into the glassy state [12, 13]. When this happens, the porous crystal collapses into a porosity-free glass having significantly higher density than the parent crystal. While this field of metal-organic framework glasses is still in its infancy, they already show an interesting combination of optical, mechanical, and chemical properties, which may lead to potential future applications.

1.18 A brief note on glasses found in nature

Despite the various erosion mechanisms, there is quite an abundance of natural glasses on the earth [14]. Obsidian, which is volcanic in origin, is perhaps the most familiar example. The chemical composition of a typical California obsidian is 75SiO2·13.5Al2O3·1.6FeO/Fe2O3·1.4CaO·4.3Na2O·4.5K2O·0.7MnO (wt%), which makes the glass just another good member of the alkali aluminosilicate family. Most obsidians are generally less than 65 million years old but would tend to devitrify (i.e., crystallize) over longer periods. Obsidians are not dry glasses: they may contain as much as ~ 1% water. This is likely the reason for their foaming behavior when heated.

Examples of other glasses found on the earth are fulgurites, which are made by fusion when lightning strikes soil, glasses of meteoritic origin, impact glasses (impactites) which might have been formed either during a meteoritic impact event by shock transformation (diaplectic glasses) or by fusion of local minerals and rocks due to absorbed heat, and tektites [15]. Not to be confused with these are the manmade glasses from industrial wastes (fragments of glass containers, sheets, highway building materials, reflector beads from road signs, etc.), and from nuclear explosions.

Scientifically, the most intriguing and hotly debated natural glass is the tektite. There are many tektite-strewn fields: the more studied ones are the Australasian tektites on land and the associated microtektites (in deep-sea deposits in the Indian Ocean, Philippine Sea, and Western Equatorial Pacific Ocean), the moldavites of Central Europe, and the Libyan Desert glass (from Western Egypt). The Australasian tektites are black to dark brown in color, typically 75SiO2·13Al2O3·4FeO/Fe2O3·3.5MgO/CaO·4Na2O/K2O·0.7TiO2 (wt%), and have been dated to be roughly 700,000 years old using ⁴⁰ K–⁴⁰Ar method. Specimens are liquid splash forms (spheres, teardrops, dumbbells, and buttons) and are generally 1–2 cm in dimensions, although, a specimen as large as 10–20 cm and weighing 12.8 kg has also been found. Estimates of the total quantities of microtektites alone are as high as about 10¹⁰ metric tons. The microtektites appear clear to light yellow/brown in color and are generally less than 1 mm in diameter. These glasses are remarkably very homogeneous [16]. After calculating possible trajectories, O’Keefe concluded that the Australasian tektites had to be of lunar volcanic origin as opposed to being the result of a terrestrial meteoritic impact in order to be as homogeneous [15]. The moldavites are pale yellow to yellowish green, and typically having a composition about 75–80SiO2·9–12Al2O3·1–3FeO/Fe2O3·0.3Na2O·3.5K2O·2–3CaO (wt%). They are estimated to be 15 million years old and are acknowledged to be fused ejecta associated with a meteorite impact which formed the Ries crater in Germany. The Libyan Desert glass specimens are pale yellow to yellowish green, contain ~ 97% SiO2, with nearly all of their 0.5% iron as Fe2O3, and dated to be about 28 million years old. Estimates of the total mass as it exists today range up to about 1500 tons, and specimens as large as 7.5 kg have been recovered. There have been suggestions that the glass has been formed by a sol-gel process, since an associated meteoritic impact crater site is yet to be found.

Also studied and analyzed extensively (by reputable laboratories around the globe) are the lunar glasses returned from the Apollo expeditions. Irregular fragments, spherules, teardrops, and dumbbells of various sizes, as little as a few microns, have been found. Glasses are known to be both of lunar volcanic and meteoritic impact origin. The compositions of some of the particles returned from the Apollo 12 expedition have been found [17] to be 38–50SiO2·10–27Al2O3·4–20FeO·0.4Na2O/K2O·9–14CaO·6–13MgO·2–4TiO2 (wt%), although there were regions of inhomogeneities having as much as 89% SiO2. Note the high FeO and the low alkali. A specimen of known lunar origin has been discovered in Antarctica [18].

Natural glasses are relics of the past; they bear the marks of time. They (naturally) present opportunities for studying chemical durability as a function of glass composition. Several of the compositions have shown remarkable resistance to weathering. Perhaps, aside from looking for the origin of the earth and the universe, some important lessons learned from the study of natural glasses are in the area of waste immobilization to help us clean up our precious planet.

1.19 Glass greats: Antonio Neri and Norbert J. Kreidl

Antonio Neri (1576–1614) was a Roman Catholic priest from Florence who published the first book on glassmaking in 1612, near the end of the Renaissance. Neri's monograph, L’Arte Vetraria (or The Art of Glass), became the standard reference for glassmakers for the next 200 years (Fig. 1.1). Here, Neri disclosed many of the techniques of Venetian glassmaking, which had previously been kept as closely held secrets. In the 133 chapters written by Neri, he guided the reader through experiments making different types of glass, including many types of colored glass for which he had an especial affection. Neri's book demonstrated his true love for glass, calling it both noble and a fruit of the art of fire.

Fig. 1.1 Cover of one volume of L’Arte Vetraria (or The Art of Glass ) by Antonio Neri, published in 1612.

Norbert J. Kreidl (1904–94, see Fig. 1.2) was one of the more admired members of the modern glass professional community. Kreidl was born in Vienna in 1904 and completed his PhD in physics in 1927 from the University of Vienna. Following his PhD, Kreidl worked in the glass industry in Czechoslovakia for 10 years. Fleeing Hitler's reign of terror, he and his young wife, Melanie, came to the United States in 1938 with only 13 dollars and a baby. Norbert found a teaching job at The Pennsylvania State University. From 1943 to 1966, Kreidl worked at Bausch and Lomb in Rochester, New York, becoming director of materials research and development. After retiring from Bausch and Lomb, Kreidl pursued his passion for teaching, first at Rutgers University and then at University of Missouri–Rolla. Kreidl is known as one of the original gurus of glass science, having published extensively on glass structure, properties, and technology. He is the author or coauthor of several books on glass science, including Glass: Science and Technology [1], coauthored with D. R. Uhlmann. A lecture award for best graduate research in glass is presented in his honor by the American Ceramic Society annually.

Fig. 1.2 Norbert J. Kreidl (1904–94), one of the most influential founders of modern glass science. Source: Am. Ceram. Soc. Bull. 73 (9) (1994).

Summary

The word glass is derived from a late-Latin term glæsum to mean a lustrous and transparent material. Glassy substances are also called vitreous, originating from Latin word vitrum (clear). The word amorphous (from Greek amorphe meaning without shape) does not have the same meaning as being a glass.

The history of glass as glazed stone beads goes back perhaps as much as 12,000 years. As independent objects, glassware was available ~ 5000 years ago. The most important technological developments in glass, such as the glass window, were sponsored by the Christian Church during the Middle Ages on the European continent.

Although transparency, luster, and durability against elements of nature are neither sufficient nor necessary to describe glass, they remain some of the key characteristics of glass that are important for large-scale commercialization. More than 95% of the commercial tonnage is oxide glasses, of which the vast majority is silica-based.

Vitreous silica, soda lime silicate, borosilicate, lead silicate, aluminosilicate, and optical glasses are the primary glass families. Vitreous silica is the basis of much of the optical fibers for telecommunications. Most of the consumer glassware such as the architectural windows, beverage containers, household lamps, fluorescent lamps, and tumblers are made from soda lime silicate glass.

Of the nonoxide glasses, those of significant commercial interest are the HMFG, the amorphous semiconductor and chalcogenide group, and metallic glasses. Of these, the amorphous semiconductors and the chalcogenides form the basis of miniaturization of the computer as switching and memory devices, solar cell (photovoltaics), and the xerographic process (photoconductivity).

Glass is also found in nature. The more important and interesting examples are volcanic glass (obsidians), lunar glass, and tektites.

Glass is currently the preferred host for the fixation of nuclear and nonnuclear hazardous waste to provide mankind an environmentally safer planet to live.

Online resources

(1)Corning Museum of Glass Video Series: https://www.cmog.org/glassmaking/studio/video-series.

(2)Corning Incorporated Inspired by Glass Video Series: https://www.corning.com/worldwide/en/innovation/the-glass-age/inspiration/inspired-by-glass-video-series.html.

(3)Center for Research, Technology and Education in Vitreous Materials (CeRTEV) YouTube Channel from Edgar Zanotto: https://www.youtube.com/channel/UCvGLC2cszadfTXeLcyl_bVw.

Exercise

(1)Name four of the major technological achievements in glass which have had the most profound impact on mankind.

(Ans: See Fig. 23.1 in Chapter 23.)

References

[1] Kreidl N.J. Inorganic glass-forming systems. In: Uhlmann D.R., Kreidl N.J., eds. Glass: Science & Technology. New York: Academic Press; 107–300. 1983;vol. 1.

[2] Rawson H. Inorganic Glass-Forming Systems. New York: Academic Press; 1967.

[3] Vogel W. Chemistry of Glass. American Ceramic Society; 1985.

[4] Jones J., Clare A., eds. Bio-Glasses: An Introduction. John Wiley & Sons; 2012.

[5] Ovshinsky S.R. Phys. Rev. Lett. 1968;21:1450.

[6] Martin S.W., Bloyer D.R. J. Am. Ceram. Soc. 1991;74:1003.

[7] Greer A.L. Science. 1995;267:1947.

[8] Schroers J. Phys. Today. 2013;66(2):32.

[9] Noda T., Inagaki M., Yamada S. J. Non-Cryst. Solids. 1969;1:285.

[10] Das T. Bull. Mater. Sci. 2000;23:499.

[11] Soraru G.D., D’andrea G., Campostrini R., Babonneau F., Mariotto G. J. Am. Ceram. Soc. 1995;78:379.

[12] Bennett T.D., Yue Y., Li P., Qiao A., Tao H., Greaves G.N., Richards T., Lampronti G.I., Redfern S.A., Blanc F., Farha O.K. J. Am. Chem. Soc. 2016;138:3484.

[13] Qiao A., Bennett T.D., Tao H., Krajnc A., Mali G., Doherty C.M., Thornton A.W., Mauro J.C., Greaves G.N., Yue Y. Sci. Adv. 2018;4:eaao6827.

[14] Pye L.D., O'Keefe J.G., Frechette V.D., eds. Natural Glasses. North Holland Publ.; 1984.

[15] O’Keefe J.G. Tektites. The University of Chicago Press; 1963.

[16] Varshneya A.K., Cooper A.R. J. Geophys. Res. 1969;74:6845.

[17] Greene C.H., Pye L.D., Stevens H.J., Rase D.E., Kay H.F. Proc. 2nd Lunar Sci. Conf. Cambridge, MA: MIT Press; 2049. 1971;vol. 3.

[18] Fudali R.F., Kreutzberger M., Kurat G., Brandstaetter F. In: Pye L.D., O’Keefe J.G., Frechette V.D., eds. Natural Glasses. North Holland Publ.; 1984:383–396 in Ref. [14].

Chapter 2

Fundamentals of the glassy state

Abstract

Glasses are a class of noncrystalline or atomically disordered material combining liquid- and solid-like features. Glasses are not merely inorganic products of fusion, although cooling of a molten liquid to essentially a rigid state is the most common way of producing glasses of commercial interest. Glass may be defined as a nonequilibrium, noncrystalline state of matter that appears solid on a short-time scale but continuously relaxes toward the liquid state. Amorphous solids are another distinct class of noncrystalline solid that, unlike glasses, do not continuously convert to the liquid state. On cooling, a molten liquid may avoid crystallizing because the mass does not possess sufficient nucleation or crystal growth rates. The supercooled liquid so obtained continues to behave like a liquid until the rapidly increasing viscosity causes it to depart from an extrapolated behavior and freeze into a solid-looking mass called glass. The departure depends on the cooling rate employed; faster cooling causes the departure sooner. Glass transition is the smooth region that connects the liquid line to the glassy state. A glass transition temperature Tg is often used to mark the end of transition into the solid state from the liquid side. The structure of glass at room temperature is presumed to be similar to that of the liquid at fictive temperature Tf. Glass invariably tends to approach the supercooled liquid with time. Due to increased mobility of atoms, lesser time is needed at higher temperatures. On a volume-temperature diagram, glass never retraces its cooling path in the transition region on reheating.

Keywords

Glass; Thermodynamics; Kinetics; Crystallization; Volume

2.1 What is glass?

Our impressions of the nature of glass are affected by everyday experience with the material. To the layman, glass is a transparent solid that breaks easily. Yet, a number of glass types, in particular the chalcogenides and metallic glasses, are opaque in the visible spectrum. Also, high-strength glasses can be made by a variety of techniques, some being used in bullet-resistant security glazing. At one time, the American Society for Testing Materials (ASTM) defined glass as "an inorganic product of fusion which has been cooled to a rigid condition without crystallizing." Even this definition is too restrictive as many organic glass systems are known, and fusion is not the only means of making a glass. The sol-gel process of making a glass is a chemical process that avoids the normally high temperatures employed for the fusion of glass. Chemical vapor deposition is yet another technique which completely avoids fusion of constituent materials.

The outward appearance of glasses is essentially solid like. The density, the mechanical properties, and the thermal properties of glasses are similar to those of the corresponding crystals. However, unlike crystals, glasses do not have a sharp, well-defined melting point. Unlike most crystals, glasses do not cleave in the preferred directions. In the absence of applied forces and internal stresses, glasses are essentially isotropic. The isotropicity of physical properties makes glasses resemble liquids. It follows that the atomic arrangements in glass must display long-range disorder typical of liquids which is evidenced by X-ray diffraction (XRD) analysis. Glass is, therefore, a noncrystalline solid. The term noncrystalline solid actually comprises two subclasses: glass and amorphous solid [1]. According to Varshneya and Mauro [2]: "Glass is a solid having a noncrystalline structure which continuously converts to a liquid on heating"; glass displays continuity of volume change as it smoothly transitions to a liquid state on heating, displaying a glass transition range behavior, discussed in Chapter 13. Substances such as Si and Ge have been brought to the disordered solid state only by means of thin film techniques (thermal evaporation, ion implantation, etc.) not involving melt-quenching. Apparently, this disordered state for both the materials is thought to have a volume discontinuity from the liquid state. When a-Si is heated, it rapidly crystallizes above 550°C. As a result, a-Si and a-Ge are often called amorphous solids, as opposed to being called glasses. While the exact volume-temperature relationships for a-Si and a-Ge are yet to be established, it is possible that, on heating, the amorphous solids make a discontinuous or continuous volume transition to a crystal or a vapor, in contrast to glasses making continuous volume transition to a liquid. The student should also note that the term amorphous solids excludes substances such as amorphous powders which may simply be microcrystals and which display more or less sharp peaks in XRD analysis. In this book, our discussions will include noncarbon based, melt-quenched substances for the most part; however, some amorphous solids will also be included. To get a clearer picture of the fundamentals of glass, the balance of this chapter is devoted to a consideration of the volume-temperature relationship (the V-T diagram) of the glass with respect to a liquid and a crystal, and pair correlation/radial distribution functions (RDFs) to describe atomic disorder in terms of the spatial relationships between the locations of various atoms in a substance and, finally, revisiting the definition of glass and its distinction from a liquid and an amorphous solid.

2.2 The V-T diagram

Consider a small volume of material at a high temperature in liquid form. Its state is given by the point a on the V-T diagram (Fig. 2.1). On cooling, the volume gradually decreases along the path abc. Point b corresponds to Tm, the melting point of the corresponding crystal, which may be defined as the temperature at which the solid and the liquid have the same vapor pressure or have the same Gibbs free energy. At this temperature, an infinitely small amount of crystals is in thermodynamic equilibrium with the liquid. However, for a perceptible level of crystallization, some finite amount of undercooling of the liquid to a point c below Tm is required. Crystallization occurs if, and only if: (i) there are sufficiently large number of nuclei present in the mass, and (ii) a large enough crystal growth rate follows. The location of the point c below Tm varies depending on when the thermodynamic driving force created by the undercooling causes a particular group of atoms to transform from the liquid state to the crystal state, and on the velocity at which the atoms from the liquid can be transported to the crystal-liquid interface. For these reasons, we have shown a wide, shaded region with varying probability representing the crystallization path. (These thermodynamic and kinetic concepts are explained further in Section 3.3.) A volume shrinkage generally accompanies the crystallization. On further cooling, the crystals so formed shrink along the crystal line to the point e.

Fig. 2.1 The volume-temperature diagram for a glass-forming liquid.

If crystallization does not occur below Tm (usually because the cooling rate is sufficiently high), the liquid mass moves into supercooled liquid state along the line bcf which is an extrapolation of the line abc. No discontinuities in the V-T curve are observed. The volume, however, shrinks continuously, that is, the structure of the liquid rearranges itself into a lower volume along the line bcf required by the lower energy corresponding to a lowered temperature. As cooling continues, the molecules become less and less mobile, that is, the viscosity of the system rapidly increases. At sufficiently low temperatures, the molecular groups cannot rearrange themselves fast enough to reach the volume characteristic of that temperature. The state line then starts a smooth departure from bcf, and soon becomes a near-straight line (often roughly parallel to de), ending at point g, when cooled fast, or at h, when cooled slowly. The material in the near-straight, low-temperature part of the curve behaves essentially as a solid. This is the glassy state.

The smooth curve between the onset of departure from the supercooled liquid line and the completion to a seemingly rigid condition is termed the glass transition region, or the glass transformation range, or abbreviated simply as the Tg range. It must be emphasized that the transition to the glassy state does not occur at a single, sharp value of the temperature. Nonetheless, a glass transition temperature Tg is customarily used (not shown in Fig. 2.1) to mark the transition. Some experimental data suggest [3] that Tg ≈ 2/3Tm for many systems. In the upper regions of the transition, glass has a viscosity of ~ 10⁸ Pa s (= 10⁹ P) or less, whereas in the glassy state the viscosity exceeds ~ 10¹⁵ Pa s or more to qualify for appearance as a solid. The intersection of the extrapolated glass line and the supercooled liquid line is termed the fictive temperature (Tf). One may imagine that Tf is the temperature at which the structure of the supercooled liquid is instantly frozen into the glass, or, the liquid is thrown into a state of suspended animation at Tf. At first glance, Tg and Tf are seemingly synonymous. But it shall be discussed in Chapter 13 that this concept of equating structural changes in the liquid within the curved glass transition region to a single intersection temperature Tf and the synonymous usage of Tg and Tf are only approximate.

The departure from the supercooled liquid line is dependent on the rate of cooling. Slower cooling allows the structure to rearrange itself to stay on bcf somewhat longer, and hence, the slower-cooled glass at h would be expected to have a lower volume (higher density) and a lower fictive temperature than a faster-cooled glass at point g. For now, it suffices to suggest that the volume of the corresponding crystal is expected to be less than that of the slowest cooled glass. (There are some interesting anomalies and questions, some of which are discussed further. Some others shall be discussed in greater detail in Chapter