1. Trang chủ
  2. » Kỹ Thuật - Công Nghệ

Handbook of Materials for Product Design Part 10 docx

80 417 1

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 80
Dung lượng 1,2 MB

Nội dung

Glass Compositions for Houseware Applications, wt% 08Seward Page 16 Wednesday, May 23, 2001 10:16 AM TABLE 8.5 8.16 Glass Compositions for Houseware Applications, wt% 08Seward Page 17 Wednesday, May 23, 2001 10:16 AM TABLE 8.5 8.17 08Seward Page 18 Wednesday, May 23, 2001 10:16 AM 8.18 Chapter TABLE 8.6 Physical Properties of Glasses for Houseware Applications At high lead concentrations, say greater than 50% PbO by weight, lead can be a network former This is especially true in binary lead silicates, borate glasses, and phosphate glasses, which will be discussed in later sections Up to 70 wt.% lead oxide is used in high refractive index optical glasses and in radiation shielding windows Typical composition ranges for lead silicate glasses and example commercial glass compositions are given in Table 8.1 Properties for those glasses are given in Table 8.2 Key characteristics The main characteristics of lead silicate glasses are high expansion, long working range, good dielectric properties, high refractive index in combination with high dispersion, and fairly good chemical durability 08Seward Page 19 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.19 Advantages Glasses containing lead are generally easy to melt Compared to soda-lime glasses of similar softening points, they have a longer working range, higher refractive index, better electrical resistivity, and lower dielectric loss, but they are more expensive than soda lime Lead provides X-ray absorption in television cathode ray tube (CRT) bulbs Disadvantages Lead oxide is considered to be a health and environmental concern It is especially so in the glass-manufacturing environment where particles of lead oxide may be airborne Once incorporated into a commercial glass product, the concerns are less unless the glass itself is ground into a fine powder such as during sawing, grinding and engraving operations Applications Major uses include I Glass tubing (e.g., neon sign tubing) I Electric filament mount structures (incandescent and fluorescent lamps) I Cathode ray tube (television bulb neck and funnel, radar screens) I Optical/ophthalmic (refractive index adjustment) I Solder glass and frits (for relatively low temperature joining or sealing of glass/metal, glass/ceramic and glass/glass; conductive, resistive, and dielectric pastes in electronic circuits; decorative enamels) I Drinkware (“hollow ware”) I Decorative and art glass I Radiation shielding (nuclear “hot” labs) Glass compositions and properties of lead-silicate glasses used for some of the above applications are given in Tables 8.3 through 8.10 Two example applications are discussed below Cathode ray tube application The construction of a color television bulb is illustrated in Fig 8.3 Several different glass components are involved: panel, funnel, neck tubing, electron gun mount, and evacuation tube Some key design requirements of the glasses are as follows: I Strength The bulbs are designed to withstand three atmospheres of external pressure I X-ray absorption Assembled tubes must have sufficiently high Xray absorption to meet federal standards Lead is generally used in neck and funnel glass for good electrical properties as well as radia- 2 0.1 TV panel glasses 55.1 3.6 Corning 9008 Panel, B&W 57.5 1.7 Corning 9039 Panel, proj 62 Corning 9061* Panel, color Corning 9068* '' 62.9 2.1 Ranges (1990) '' 59–65 1.3–2.3 Typical * '' 61.3 0.8 12 2.5 3.5 8.4 1.9 0.5 0.01 29 28 23 75.4 CeO2 ZrO2 TiO2 Other 0.5 0.5 0.1 6.8 6.4 12 1.7 02 0.4 6.3 5.8 8.7 15 0.4 0.7 † 0.7 1.7 10.3 2.4 2.2 0.2 0.7 0.5 0.2 6.99 8.9 0.85 1.79 10.31 2.37 2.31 0.16 0.38 0.52 0.21 F=0.29,† 7.0–8.5 6.6–8.9 0–1.5 0.1–2.8 8.0–10.3 2.0–8.7 0.05–2.8 0–1.0 0–0.3 0–0.6 0.3–0.5 1–2.6 0.2–0.5 F=0–0.3,† Nd2O5=0–1.00 7.6 7.6 0.05 9.2 9.2 0.5 0.3 Refs Ref Ref Ref Ref 12.2 Obsolete Typical colorant levels for contrast enhancement are Fe2O3 = 0.04, NiO = 0.012, CO3O4 = 0.002, and Cr2O3 = 0.0006 † Sb2O3 As2O3 ZnO PbO BaO SrO CaO MgO K2O 56 52.2 54 Na2O Stem tubing Neck tubing Funnel Sealing frit Li2O Lead TV glasses Corning 0120 Corning 0137 Corning 0138 Corning 7580 B2O3 Description Al2O3 Glass ID Si02 TV Glass Compositions, wt% 0.4 1.4 0.3 * Ref Ref Ref Ref Ref 08Seward Page 20 Wednesday, May 23, 2001 10:16 AM TABLE 8.7 8.20 TV Glass Properties, wt% CTE Density to 300°C g/cm 10–7/°C Glass ID Description Lead TV glasses Corning 0120 Corning 0137 Corning 0138 Corning 7580 Stem tubing Neck tubing Funnel Sealing frit 3.05 3.18 2.98 6.47 TV panel glasses Panel, B&W Corning 9008 Panel, proj Corning 9039 * Panel, color Corning 9061 * Corning 9068 '' Ranges (1990) '' 2.64 2.9 2.7 2.696 2.7–2.8 Typical * Obsolete '' 2.8 Strain point °C 89.5 97 97 98 Annealing Softening Working Young’s Resistivity point point point modulus Poisson’s Refractive 350°C Absorption °C °C °C GPa ratio index log(Ω-cm) @0.06 nm 395 436 435 293 435 478 474 311 630 661 654 374 986 978 965 890 89 406 96.4 458 99 460 98.5–99.5 455–470 '' '' 444 500 501 510–525 '' 646 680 689 687–707 '' 1004 983 1000 102 '' '' '' 59 0.22 0.23 0.24 0.25 1.56 1.55 1.565 1.65 8.3 7.6 8.2 0.24 0.24 0.23 1.506 1.553 1.518 7.4 7.7 7.5 >7 75 90 62 40 Refs Ref Ref Ref Ref '' 20 Ref 28 Ref 35 Ref Ref 29.1 Ref '' ~29 08Seward Page 21 Wednesday, May 23, 2001 10:16 AM TABLE 8.8 8.21 08Seward Page 22 Wednesday, May 23, 2001 10:16 AM Sealing and Solder Glass Compositions, wt% TABLE 8.9 8.22 Physical Properties of Sealing and Solder Glasses 08Seward Page 23 Wednesday, May 23, 2001 10:16 AM TABLE 8.10 8.23 08Seward Page 24 Wednesday, May 23, 2001 10:16 AM 8.24 Chapter Figure 8.3 Schematic showing cross section of components of a conven- tional color television tube The glass envelope consists of a funnel section, a faceplate section, and a neck section The tubulation is used to evacuate the tube and is removed after vacuum processing (From Ref 7, Fig 1, p 1039, reproduced with permission of publisher.) tion absorption (see Sec 8.2.3.5); it is not used as much in the panel glass (screen), because it “browns” under irradiation I Radiation damage The panel glass must not change color or “brown” under X-ray or electron irradiation Since the presence of lead promotes electron browning, only small amounts of lead can be used in a panel glass Cerium oxide is often added to the composition to help minimize the browning Cerium ions act as electron traps that not absorb light in the visible portion of the spectrum I Electrical properties Properties include high bulk electrical resistivity (excellent insulator) and high dielectric strength to resist elec- 08Seward Page 25 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.25 trical puncture or discharge through the glass under high potential gradients (anode voltage can exceed 25 kV for color television) I Optical properties The luminous transmittance and chromaticity of the screen is specified and varies among TV set designers and manufacturers (absorption of light is needed to improve viewing contrast); doping elements used for tinting include Ni, Co, Cr, and Fe I Viscosity sizes I Thermal expansion All parts (panel, funnel, neck, solder glass, and shadow mask metal alignment pins) must match fairly closely, 90 to 100 × 10–7/°C It has a working range that allows forming into large Compositions and properties of lead-containing glasses useful for television bulb manufacture are given in Tables 8.7 and 8.8 Drinkware (“hollow ware”) and art glass applications of lead glass Historically, the term crystal has been used for any clear glass of high transparency, especially when cut so that it has many light-reflecting surfaces, as in crystal chandeliers We should note here that this use of the term crystal indicates nothing about the atomic level structure of the material As is true for all the glasses in this section, the structure is amorphous, not crystalline ASTM defines crystal as “(1) colorless, highly transparent glass, which is frequently used for art or tableware (2) colorless, highly transparent glass historically containing lead oxide.” In Europe, lead crystal must contain at least 24% (wt.) PbO and full lead crystal at least 30% SteubenTM glass has traditionally contained about 30% PbO Attributes Attributes of lead crystal ware include brilliance, clarity, sonority (acoustic resonance), density, meltability, and workability Brilliance is related to the refractive index, transmittance, and the degree that surface polish is maintained Sonority is related to the high elastic modulus and low internal friction (acoustic damping) of the glass, a function of the mixed alkalis present Lead is not an absolute requirement for any of these attributes A negative: high-lead glasses are more easily scratched or abraded Most lead crystal manufacturers are developing non-lead compositions for at least some of their products Forming processes Forming processes include pressing, blowing, tube drawing, casting, centrifugal casting, and fritting but are not limited to these 08Seward Page 81 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.81 ron, phosphorous, and germanium), fluorine (which can substitute for oxygen in the network), and possibly titania and alumina It is advantageous from a strength point of view to use dopants that give the core a slightly higher thermal expansion than the cladding Figure 8.8 shows the effects various concentrations of these components have upon refractive index of silica Synthetic silicas can be doped during manufacture by incorporating appropriate precursors along with the silica precursor in the vapor stream fed to the deposition burners Methods for doing so are described in Sec 8.7.4 Example precursors include the chlorides GeCl4, BCl3, and POCl3, the fluoride SiF4, and the solid AlCl3, which can be delivered by subliming at high temperatures Fiber amplifiers and all-optical systems The future directions of optical communication are toward all-optical systems That is, the trend is to utilize optical signal processing (coupling/splitting, switching, amplifying) rather than converting light signals to electronic signals, processing electronically, then converting back to light signals Such signal processing can be done with fiber devices (constructed from optical fiber) or planar devices (constructed by building optical paths on flat substrates using patterned layers of different refractive index glass, generally doped silica, often in combination with vapor-deposited conducting and semiconducting films) Light amplifiers are essential to all-optical systems (no electronic repeaters, no light-to-electronic conversion), since they compensate for optical losses within the fibers and the associated connectors, couplers, and switches They are achieved by doping the fiber core with Figure 8.8 Refractive indexes (nD) as a function of dopant level in bulk SiO2 F levels are in atomic percent.25 08Seward Page 82 Wednesday, May 23, 2001 10:16 AM 8.82 Chapter fluorescent lasing ions such as erbium or neodymium, and “pumping” using laser diodes The operation of the erbium optical fiber amplifier (OFA) is based on the lasing action at 1.55 µm of the Er3+ (erbium) ion The fiber is optically pumped at either 980 or 1480 nm by a suitable solid state laser diode Light emission by electronic transitions between the 4I13/2 and 4I15/2 levels is stimulated by the light signal at 1.55 µm and transfers power to that signal To construct the erbium OFA, the fiber core is co-doped with Er3+ at levels up to about 300 ppm, in addition to the index adjusting elements (for example, germania) A major challenge is to obtain uniform gain (amplification) across a wide band of wavelengths, such as is required for WDM (wavelength division multiplexing) at 1.5 and 1.6 µm Some methods for achieving this use separate gain-leveling devices Nd (neodymium) doped fibers have been investigated for operation within the 1.31 µm communications band but have not yet been commercially successful For amplifier operation at those wavelengths, non-silica-based fibers may be of advantage, leading to hybrid (mixed) glass fiber systems 8.3 8.3.1 Glass Making I—Glass Melting Introduction and General Nature The term glass melting as generally used applies collectively to all the steps used to convert raw materials into a molten mass that can be subsequently formed as an object The process of preparing a quality melt from which glass products are to be made consists of several stages, as follows: I Batch preparation (raw material selection, weighing and mixing) I Batch melting (conversion of the batch raw materials into a viscous liquid that is essentially free of any crystalline material) I Fining (the removal of bubbles) I Homogenizing (the removal of chemical and thermal variations within the melt, often occurring simultaneously with fining, the two processes together sometimes being called refining) I Conditioning (bringing the melt to the uniform temperature required for whatever forming process will be used to make the product) Each of these stages is performed more or less in sequence, either at one location, as in a crucible, or at a progression of locations, as in a continuous glass melting tank consisting of several or many zones 08Seward Page 83 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.83 The various types of melters are described in Sec 8.3.3 Here, we will describe glass-melting steps in more detail 8.3.2 Steps in Glass Melting Glass batch considerations Rarely does a commercial process start with an all-oxide batch This is in part because the expense of the materials would be too great, but mostly because oxides melt at much higher temperatures than the salts of the corresponding metallic elements, thus requiring longer times at higher temperatures to complete the chemical reactions Many different combinations of raw materials can yield the same final glass composition Final choice of raw materials is based on factors such as chemical composition, the level of impurities tolerated, particle size, particle size distribution, the precision to which these characteristics are controlled (maintained) by the vendor, and price 8.3.2.1 Batch materials Because of space restrictions, we cannot discuss all of the batch materials used in glass manufacturing, but we will discuss several of the most relevant to give a feeling for some of the key considerations Silica (SiO2) I Silica consists of quartz sand, crushed quartzite, or beneficiated sandstone I For sandstone, a combination of quartz sand and clay, beneficiation consists of crushing, washing, and froth flotation to remove refractory heavy metal oxides I In the U.S.A., it is shipped and handled dry; in Europe, wet I Generally, it is > 99% SiO2; often > 99.5% with Fe2O3 < 0.03% (by weight) I Generally, it is 40 to 140 mesh (coarser is harder to melt; finer leads to “dusting” effects) Also, finer particles may melt too fast and raise the melt viscosity before air trapped in the batch can escape I Cost trade-offs include iron content and particle size distribution I Chemical analysis of raw materials, either in-house or by supplier, is important Limestone/calcite (CaCO3) I This is either high calcium (~CaCO3) or dolomitic (< 46% MgCO3) I Dolomite is CaMg(CO3)2 08Seward Page 84 Wednesday, May 23, 2001 10:16 AM 8.84 Chapter I Sometimes the oxide (CaO) is used (lime, quicklime, burned lime, or dolomitic quicklime) I Sometimes aragonite, a different mineral form of CaCO3 is used I Limestone generally > 96% CaCO3, with SiO2 and MgCO3 as major impurities Soda ash (Na2CO3) I Sources for the U.S.A and most of world are trona deposits in Wyoming (Trona is a hydrated sodium carbonate sodium bicarbonate ore; Na2CO3 · NaHCO3 · 2H2O.) I The ore is calcined and treated to remove insoluble impurities, generally achieving > 99.5% Na2CO3 I Na2CO3 is also made by the Solvay process (react salt with limestone to get CaCl2 as a by-product) Solvay is still used as a source in Europe and other parts of the world I Hydration of the raw materials at temperatures < 110°C is a problem, so materials are often supplied as the monohydrate (Na2CO3 · H2O) Alumina (Al2O3) I Alumina is a very refractory material and slow to dissolve, so it is generally batched as a mixed-oxide mineral I The source is generally feldspar or nepheline syenite (minerals containing 60 to 70% silica, 18 to 23% alumina, calcia, magnesia, soda, potassia, and iron oxide (0.1%) I Since alumina is generally a minor batch component, the accompanying oxides can usually be easily assimilated in the composition I Accurate chemical analysis is important I It is relatively expensive but, when refined from the ore bauxite, it is even more expensive—often too expensive Borates I Borates are obtained mostly from California or Turkey I Borax (Na2O · 2B2O3 · 5H2O), sometimes called mol borax I Boric acid (H3BO3 or B2O3 · 3H2O) I Colemanite (Ca2B6O11 · 5H2O or 2CaO · 3B2O3 · 5H2O), from Turkey I The mixed sodium-calcium borate minerals, ulexite and probertite, are used in the fiberglass industry 08Seward Page 85 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.85 I Anhydrous borax (Na2O · 2B2O3), also known as anhydrous sodium tetraborate (Na2B4O7), is generally not used because it absorbs water The fully hydrated form of borax (Na2O · 2B2O3 · 10H2O), on the other hand, is generally not used because it loses water, decomposing to mol borax at temperatures above 60°C Such changes in water content make batch calculations and weighing somewhat uncertain I Anhydrous B2O3 is too hygroscopic and too expensive to be a good commercial batch material Others I Litharge (PbO) and red lead (Pb3O4) are being replaced as batch materials by various lead silicates, which are safer to handle I Salt cake (Na2SO4) is used as a melting accelerator (forms eutectic with Na2CO3) and fining agent I Gypsum (CaSO4 · 2H2O), melting accelerator, fining agent I Sodium nitrate (Na2NO3), oxidizing agent, stabilizes color Cullet Waste or broken glass softens readily in glass melters to bring batch particles together, thereby increasing batch reaction rates, often immensely This batch material is referred to as cullet There are many economic and environmental factors favoring use of cullet in batch melting operations Sources include the following: I In-house cullet, generally of known composition This is usually the same as the glass being produced, so few special batch adjustments are required (exceptions being laminated/clad glass) I Foreign cullet, from sources other than the manufacturing plant (other plants, customers, other manufacturers, post-consumer) I Post-consumer cullet, which is increasing in importance as recycling expands It must be carefully processed to remove metal, ceramic, and other detrimental contaminants Transparent glass-ceramics mixed in with the glass cullet are a concern for container manufactures, especially in Europe Cullet can be used in any proportion, but 30 to 60% is generally considered most effective Sizes coarser than the raw materials seem to work best; sometimes they are added as chunks Cullet can be added in the mixer, at the fill (doghouse), under, over, or layered with the batch It must be added in a continuous enough manner to maintain melting stability 08Seward Page 86 Wednesday, May 23, 2001 10:16 AM 8.86 Chapter Batch formulation A combination of batch materials is selected based on the desired glass composition and oxidation state in combination with manufacturing process requirements (including those of mixing, melting, fining, etc.) and product cost The quantities of each component are calculated to give the required weight of oxide after the volatile components and reaction products, such as water, carbon dioxide, sulfur dioxide, and nitrogen oxides, are lost to the atmosphere Raw materials handling Transportation and storage Raw materials that are used in large quantities are often delivered by railroad car and stored in large vertical silos or moderate-sized storage bins A conveyor or elevating mechanism is used for loading the silos Lesser batch components are delivered and stored in bags or other containers Collecting and weighing The trend in large manufacturing operations is to automate these steps as much as possible Batch components can be weighed in hoppers at individual silos or storage bins and conveyed to the mixing location More often, the batch components are fed into a single scale for weighing The scales themselves must be accurate and regularly calibrated The weight sensors are often electronic, with digital readouts and modern computer interfacing and control Such electronic weighing systems can be sensitive to part in 4,000 (250 ppm) Mixing A wide variety of mixers are used Most are of rotary (like a cement mixer) or pan-type construction A variety of paddle and mixer blade designs are used The trend is to mix the batch at a location very close to the melting furnace to minimize opportunity for batch segregation (see below) to occur Filling or batch feeding A variety of filling devices or machines are used They often involve a hopper and a mechanism such as an auger or mechanical vibrator to feed the batch into the furnace via a chute or tube Multiple feeders are generally used to supply batch to large melters Reciprocating feeders, whereby a pile of batch is deposited on a shelf extending the full width of the melter entrance and then pushed into the melter by a bar, are used for large container and float-glass furnaces This reciprocating action results in a sequence of batch “logs” being fed into the tank The resulting undulating batch surface improves the melting process by allowing rapidly melted batch to run downhill, exposing fresh batch to the heat 08Seward Page 87 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.87 Many variations are possible for all the operations described above In smaller factories and hand shops, the automation tends to be less, but care and accuracy remain important Batch segregation The term batch segregation refers to some batch components becoming spatially separated from the rest on some relatively large scale (It can also refer to a separation of different particle sizes of the same batch component.) It is a type of demixing, which is problematic, because different regions of segregated batch may then melt to produce glasses of different compositions Some regions may not melt at all The resulting inhomogeneities must be removed later in the process Segregation can occur inside the mixing machine, as the mixed batch is being transported to the melter, or as it is being fed into the melter Forces causing demixing operate whenever moving streams of particles are present Often, they rapidly create steadystate patterns of segregation under constant flow conditions These patterns are reproducible Factors affecting batch segregation include the following: I Size difference is the main contributing factor (A size difference ratio of as little as 5% can lead to measurable demixing.) I Nominal grain size and density differences are lesser factors I Other much lesser contributing factors include particle shape, lubricity, and surface charge Methods for controlling batch segregation include the following: I Use multiple discharge silos I Use raw materials having a narrow size distribution I Match their particle size and size distribution I Do not mix too long I Minimize the movement of mixed batch (locate the mixer at the furnace, not in the batch house) I Wet the batch Wetting of the batch helps reduce batch segregation by suppressing the free-flowing characteristics of batch (It provides a thin liquid coating on each particle.) Wetting also helps decrease dusting and batch carry-over into the exhaust system Generally, water or a 50% NaOH solution is used Of course, wetting complicates the batch handling system and adds expense The technique must be used with care when hydratable batch components are present, since they can absorb water, which may lead to “setting up” of the batch 08Seward Page 88 Wednesday, May 23, 2001 10:16 AM 8.88 Chapter Batch preheating Several manufacturing groups are currently investigating the advantages of preheating of batch using heat from the furnace exhaust gases The goals are at least twofold: improve the efficiency of the batch melting step, and recover waste heat This will, of course, complicate the process and add some expense 8.3.2.2 Processes occurring within the melter Batch melting involves heat transfer from a source of heat energy; for example, fuel combustion in a burner or electric resistance (Joule) heating, and complex chemical reactions among the batch components Since a formulated batch often contains water (moisture), and many batch components such as boric acid contain chemically combined water, the first step of melting consists of drying and dehydrating the powdered raw materials Many chemical reactions among the dry batch materials occur, some simultaneously Solid state reactions between batch components occur, sometimes leading to molten phases and other times to new crystalline phases This is followed or accompanied by melting the salts and other low-melting-temperature ingredients (fluxes); decomposing some of the fluxes, producing gases; reacting silica sand and other refractory materials with the melted fluxes (for example, reaction of sodium carbonate with silica sand); and dissolving the remaining refractory batch components (e.g., silica, zirconia) into the melt Fining involves the removal of the bubbles generated during batch melting Such bubbles consist of air that was trapped between grains of batch and gases such as CO2, SO2, and H2O generated by the reactions of the batch materials; e.g., decomposition of sodium carbonate releases CO2 gas Gases can also be generated by reaction of the melt with materials comprising the melter, i.e., the container and electrodes The bubbles tend to rise to the surface of the melt, driven by buoyant forces according to Stokes’ law This is often referred to as Stokes fining If the bubbles are small, say 0.2 mm diameter or less, this can be a very slow process To aid the fining, the temperature of the melt is often increased so as to decrease the viscosity of the melt and accordingly to decrease the viscous drag on the rising bubbles Occasionally, pressure changes above the melt may also be utilized to aid in fining During fining, gases produced by further chemical reactions in the melt enlarge the bubbles Generally, the glass composition is designed to contain chemical compounds, known as fining agents, which release these additional gases during fining Examples of such fining agents are As2O5, Sb2O5, and Na2SO4 The elements As, Sb, and S can exist in the glass in different oxidation states (see the discussion of oxidation state below), depending on the temperature and oxygen activity of the 08Seward Page 89 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.89 glass If the glass temperature is raised above that at which the glass was melted, these elements become more reduced, giving off oxygen (O2) or sulfur dioxide (SO2) gas according to the following equations: As2O5 ⇒ As2O3 + O2 (gas) Sb2O5 ⇒ Sb2O3 + O2 (gas) 2Na2SO4 ⇒ 2Na2O + 2SO2 (gas) + O2 (gas) The gases generated by the above equations remain dissolved within the glass in their molecular state until supersaturation becomes sufficient to nucleate bubbles of gas However, if bubbles are already present, the dissolved gas molecules diffuse to those bubbles, thereby enlarging them The enlarged bubbles then rise to the surface of the melt and break Bubbles not reaching the surface dissolve when the melt is allowed to cool from the fining temperature This dissolution process is aided by the reverse action of the fining agents Sometimes, during batch melting and fining, the gas bubbles not burst when they reach the melt surface but rather create a foam layer Such foaming adversely affects heat transfer into the melt and can lead to glass defects, so it should be avoided by any effective means Homogenization Sources of inhomogeneity in the melt include the physical segregation of chemical components during batch mixing or batch melting, preferential volatilization of certain chemical components from the melt surfaces, and reaction of the melt with the furnace wall refractories Mechanisms operating to improve homogeneity are composition-gradient-driven mass diffusion in combination with stirring (forced or thermally driven convection) Stirring serves to reduce the diffusion distances required for homogenization Thermally driven convective motion in melters serves several functions It increases the melting rate by moving hot molten glass under the floating batch layer (called the batch blanket), supplying it with a new source of heat Upward convection part way along the tank can serve to help separate the batch melting and fining sections of the melter Simple patterns of convective motion in a large tank-type melter are illustrated in Fig 8.9 Sometimes a row of bubblers is built into the tank bottom to release large air bubbles that forcefully rise to the top aiding the convective motion And, as described above, convection helps to mix and homogenize the melt Conditioning and delivery For a molten glass to be processed into a product by any of the many commercial forming processes, it first must be adjusted to the temperature required for that process In gen- 08Seward Page 90 Wednesday, May 23, 2001 10:16 AM 8.90 Chapter Figure 8.9 Major convection flows in a fuel-fired tank furnace (longitudinal vertical section on tank centerline) (From F E Woolley, Ref 7, Fig 3, p 388, reproduced with permission of publisher) eral, all the glass being delivered from the tank must be at that same temperature This is referred to as conditioning (or thermal conditioning) the glass Conditioning is usually done either in a final zone of the glass-melting tank or in a separate unit, called a forehearth, attached to the exit of the melter Heat input and heat losses are adjusted to yield glass of uniform temperature To achieve the needed balance, the forehearth is often heated electrically or by fossil-fuel burners Mechanical stirrers are often placed in the forehearth to improve the chemical and thermal homogeneity The forehearth often has more than one exit from which molten glass is delivered; thereby several different forming machines can be fed from one melter Forehearth exit configurations vary depending on the forming process that follows They can be simple single or multiple circular orifices of refractory material, an orifice containing a flow controlling needle, a tube or pipe, or a horizontal “lip” over which the glass flows When the forming process requires delivery of discrete gobs of molten glass, automated metallic shears are generally used to cut the stream of glass into sections as it exits the melter Often, the motion of a needle or plunger within the delivery orifice is synchronized with the shearing action to better control the gob weight and shape One common type of gobbing feeder is illustrated in Fig 8.10 Oxidation state (redox) When a glass is melted, a chemical equilibrium is approached between the glass composition and that of the surrounding atmosphere If there are polyvalent ions present within the glass, their average oxidation state is influenced by the partial pressure of 08Seward Page 91 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.91 Figure 8.10 A typical gob feeder process (From H J Stevens, Ref 7, p 396) oxygen in the atmosphere above the melt Thus, a glass melted in a fuel-fired furnace, where the atmosphere has been depleted of oxygen by the combustion process, tends to be more reduced than one that has been electrically melted in a standard atmosphere containing about 20% oxygen The degree to which a glass is oxidized is often referred to as its oxidation state The fining equations listed above describe reversible oxidation-reduction reactions, called redox reactions Thus, the oxidation state of a glass is sometimes referred to as its redox state When the temperature of a glass melt is decreased below its melting temperature during glass manufacture, it generally has insufficient time to equilibrate with the atmospheres it encounters at the various steps Consequently, it tends to retain the overall oxidation state that it established during melting However, as the glass cools, different multivalent elements (for example, iron and manganese) compete with each other for the available oxygen The relative oxidation states (percent oxidized) of those elements can change with respect to each other during cooling of the glass Thus, properties such as glass color 08Seward Page 92 Wednesday, May 23, 2001 10:16 AM 8.92 Chapter can change as a glass melt cools In photosensitive or photochromic glasses, the photosensitivity is affected by the oxidation state of the melt, and in some cases of the solid glass The oxidation state of the molten glass also strongly affects the rates of melting and fining Redox control Sulfates (salt cake and gypsum) and nitrates are sometimes used as oxidizing agents, and carbon or calumite (a reduced calcium-aluminum silicate slag from steel manufacture containing carbon and sulfides) as reducing agents Of course, the use of cullet containing polyvalent coloring ions and impurities can also alter the oxidation state of a melt Oxygen sensors, usually in the form of galvanic cell probes inserted into the melt, are being used with increasing success as devices for monitoring and controlling redox state Melter-created glass defects Despite the best efforts to generate a perfectly uniform glass melt, within any melter there are naturally occurring processes that oppose those efforts These include refractory corrosion (dissolution), electrode corrosion, and preferential volatilization of some species from the melt surface These produce localized and sometimes more global deviations from the desired glass composition Localized composition deviations lead to inhomogeneities in the product, called cord and striae Other glass defects originating in the melter are solid inclusions called stones These may be particles of nonmelted batch or pieces of eroded furnace refractory Devitrification (crystallization of the melt resulting from it having been cooled to temperatures below its thermodynamic liquidus temperature and held there for extended periods of time) is another source of solid inclusions, generally called devit Bubbles and blisters may result from incomplete fining or from reboil (the exsolution of bubbles in an otherwise bubble-free melt, generally the result of thermally driven chemical or electrochemical reactions) Such reboil occurs either homogeneously within the melt or heterogeneously at a melt-refractory interface The selection rate (percentage of “good” ware) and the profitability and competitiveness of a glass manufacturing operation often depend on how successfully these melting-related defects are avoided The design of the melter and its method of operation are key factors for success 8.3.3 Types of Melters General Glass can be melted on greatly different scales, ranging from a few grams in a laboratory experiment to hundreds of tons per day in a large-scale manufacturing operation A great deal of 8.3.3.1 08Seward Page 93 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.93 glass is melted on a scale somewhere in between It is important to distinguish between continuous and discontinuous melting of glass In a discontinuous process, the batch is placed in a container, such as a crucible, pot, or a tub-like vessel, and heated according to a prescribed time-temperature cycle to carry the contents through the melting stages described above, namely batch melting, fining, and conditioning Homogenization may be enhanced by mechanical stirring The vessels may be constructed of metal, refractory ceramic blocks, or cast fireclay; the stirrers generally of metal, fireclay, and high-density ceramic refractories, occasionally with platinum cladding In a continuous process, batch materials are continually fed into a melter at one location, and molten glass is continually withdrawn from the melter at another location All the steps required for melting take place approximately sequentially as the molten mass progresses through the melter, either vertically or horizontally Large glass melters, or tanks, are generally constructed so that the melt passes through them horizontally, with melting, fining, homogenizing, and conditioning occurring progressively as the melt passes through In some tank designs, attempts are made to confine the process steps to zones separated from each other by physical walls or partial walls, or by thermal convective patterns within the melt itself Examples of simple discontinuous melters include crucibles of various sizes and configurations and pot melters Crucibles are generally made of refractory ceramic materials such as alumina, or refractory metals such as platinum alloys If small, the crucibles can be inserted and removed from a heated furnace and the glass delivered to a suitable forming operation by the simple means of tipping the crucible and pouring the glass Larger crucibles are generally fixed in position and molten glass delivered, after it has been suitably homogenized, by the means of an orifice, or tap, at the bottom of the crucible Large crucible and tub-like containers operated discontinuously are sometimes referred to as day tanks Pot melters generally consist of single or multiple refractory ceramic pots located in a single furnace The pots are shaped something like tall bee hives with an opening at the top front through which batch or cullet (broken glass) can be inserted and molten glass withdrawn using a ladle or blow pipe Such pots have been used for hand glassmaking operations, for both artistic and utilitarian products, for centuries and are still used for such purposes today Fuel-fired continuous melters, often called tank melters, or simply glass tanks, generally consist of large refractory swimming-pool shaped tanks, heated from above the melt by gas or oil flames from multiple burners The top of the melter is covered by a refractory superstructure consisting of walls and a self-supporting roof called a 08Seward Page 94 Wednesday, May 23, 2001 10:16 AM 8.94 Chapter crown In the simplest version, the so-called direct-fired furnace, the gas flames pass across the top of the melt, and the melt is heated by radiation from the flames and from the hot crown of the furnace The combustion gases are then exhausted from the furnace and dispersed, generally via a tall stack and often after passing through electrostatic bag precipitators to reduce particulate emissions and chemical scrubbers to remove acidic vapors Such fuel-fired continuous melters can be very large For example, a soda-lime-silica float glass melter may contain as much as 2,000 tons of molten glass and deliver it at a rate of 800T/day The top surface area of the glass in the melter could be more than 5,000 ft2 Large-container glass melters can deliver 100 to 300 tons of glass per day with melt surface areas between 600 and 2,000 ft2 Because of the rather slow forward motion of the glass, in combination with the convectiondriven mixing effects, it is not unusual to expect an average of to days residence time for the glass in such large melters Heat recovery—regenerative and recuperative furnaces Historically, the heat energy required for melting glass was produced by wood or coal fires Today, the heat is generally provided by burning natural gas or oil, or by immersed electrodes utilizing electric resistance (Joule) heating Obviously, not all of the heat generated by a flame is transferred to the glass Some of it is used to heat the nitrogen gas contained in the combustion air fed to the burners (which is about 80% nitrogen by volume), and much is carried away by the nitrogen and by the gaseous products of combustion (H2O and CO2) as they go up the stack For economic reasons, especially when operating large melting tanks, it is important to recover some of the exhausted heat This is generally done by using the exhaust gases to preheat the incoming combustion air, either regeneratively or recuperatively, in rather massive refractory heat exchangers In a cross-fired regenerative glass furnace, of the type first developed by Siemens toward the end of the nineteenth century (see Fig 8.11), the furnace is fired for a period of time by a row of burners along one sidewall, while the hot gasses are exhausted from the opposite wall and passed through large chambers called regenerators The regenerators consist of stacks of firebrick in open three-dimensional checkerboard patterns, called checkers, that absorb heat When the checkers have become suitably heated, the burners are shut down A set of burners on the opposite side of the furnace is then fired and fed with combustion air drawn in through the hot regenerators The combustion gases are now exhausted from the opposite side of the furnace 8.3.3.2 08Seward Page 95 Wednesday, May 23, 2001 10:16 AM Inorganic Glasses 8.95 Figure 8.11 Cross-fired regenerative-type furnace construction typical of glass melt- ers.26 through another set of regenerators This process is reversed about every 15 to 20 In a recuperative furnace, the burners are fired continuously, and heat is exchanged continuously in the exhaust system where the exhaust gases flow outward through a central tube or tubes with the incoming air channeled along the outside of the tube(s) Construction materials/refractories The refractory materials serve three purposes: they contain the glass melt, provide thermal insulation and heat transfer as needed, and act as key structural elements of the tank The refractories are generally held in place by an open steel framework, but much of the mechanical load is born by the 8.3.3.3 ... 42 to 100 × 10? ??7 /°C Thermal expansion of sealing glasses is often modified by use of low-expansion fillers (fine particles of low-expansion crystals or glasses) Compositions and properties of some... required to melt the raw materials (1723°C is the melting point of cristobalite, the high-temperature crystalline form of silica), silica cannot be melted to form good quality products by conventional,... (105 to 108 °C/s) from the molten state While they are of commercial value for their significantly enhanced electrical, magnetic, and structural strength aspects, a full discussion of these materials

Ngày đăng: 10/08/2014, 19:20

TỪ KHÓA LIÊN QUAN