• Author / Uploaded
• Jose Antonio Chambi

• Categories
• Diamante
• Piedra preciosa
• Minerales
• Cuarzo
• Materiales

Citation preview

FIGURE 36.13. (A) Combined effect of incorporation of Ti and Fe in sapphire. The distance between the dopant cations is 0.265 nm. (B) Corresponding energy band diagram for the excitation shows how a gemstone with blue coloration is obtained by absorption. FIGURE 36.12. Dichroscope for estimating the dichrosism of a stone and the optics on which it is based.

TABLE 36.5 Transition Element Dopants and Color Titanium Vanadium

Blue sapphire (with iron), blue zoisite (tanzanite) Grossular garnet (tsavorite), green vanadium beryl, synthetic corundum (alexandrite simulant), some synthetic emeralds, blue/violet sapphire Chromium Ruby, emeralda, red spinel, pyrope garnet, chrome grossular garnet, demantoid garnet, uvarovite garnetb, chrome diopside, green jadeite, pink topaz, alexandrite, hiddenite Manganese Rhodochrositeb, rhodoniteb, spessartite garnetb, rose quartz, morganite variety of beryl, andalusite Iron Sapphire, sinhaliteb, peridotb, aquamarine, blue and green tourmaline, enstatite, amethyst, almandine garnetb Cobalt Synthetic blue and green spinel, synthetic blue quartz (except for a rare blue spinel, cobalt is not found in any natural transparent gemstone); cobalt glass Nickel Chrysoprase, synthetic green and yellow sapphires Copper Diopside, malachiteb, turquoiseb, synthetic green sapphire a b

In UK and Europe only beryl colored by chromium may be described as emerald. Idiochromatic gemstones.

diamond, although such stones could be common if irradiated more often. Green. The green stones include emerald, malachite, and both uvarovite and tsavorite garnets. Red. The red stones include ruby and both pyrope and almandite garnet (almandine).

36.6 C O L O R

IN

GEMS

AND

Yellow. Yellow stones are less common but include citrine, yellow sapphire, and yellow diamond. The color of ruby red and emerald green is the result of a transition metal in a ligand field (see Section 32.5). In both cases, the coloring ion is Cr3+ substituting for an Al3+ ion. Ruby is red corundum (Al2O3) containing about 1% Cr2O3. The Cr3+ ion, which is only a little larger than Al3+, is easily accommodated into the corundum structure. The five 3d orbitals in the Cr3+ ion split in the distorted octahedral ligand field. This is a slightly more complicated case than for the simple octahedral ligand field we described in Section 32.5 and involves further splitting of the 3d levels, as shown in Figure 32.9. Nevertheless, absorption of selected values of l in the visible spectrum is due to electron transitions. Absorption is strongest in the green and violet and least in the red and blue; this gives ruby its red color with a slight purple overtone. In emerald, the color is again due to the Cr3+ ion replacing Al3+ in a distorted octahedral arrangement very similar to that in corundum. Emerald without the chromium impurity is beryl, which has the chemical formula 3BeO. Al2O3.6SiO2 or Be3Al2Si6O18. Because of the presence of the other constituents (it’s a ring silicate), the overall bonding is a little weaker and the ligand field is less strong. As a result, the splitting of the energy levels is different, with strong absorption occurring in the yellow-red and violet regions and strong transmission in the blue-green. For sapphire blue, the color results from a charge transfer mechanism. Sapphire shares the corundum structure with ruby, but the impurities are now small amounts of both iron and titanium oxides. Both Fe2+ and Ti4+ take the place of Al3+ in the corundum structure. If they are present on adjacent sites, as shown in Figure 36.13, an interaction

M I N E R A L S ....................................................................................................................................................

685

between them becomes possible. In this configuration, there is enough overlap of the dz2 orbitals of the two ions that it is possible for an electron to transfer from the Fe2+ ion to the Ti4+ ion as follows. Fe2þ þ Ti4þ ! Fe3þ þ Ti3þ

(36.3)

The energy of the combination on the right side of equation 36.3 is 2.11 eV higher than that on the left side, as shown in Figure 36.13. If light of this energy falls on blue sapphire, it is absorbed while producing the charge transfer shown in equation 36.3. You can produce the blue color by doping with Ni or Co, but the mechanism is different.

36.7 OPTICAL EFFECTS Chatoyancy. The scattering of light by aligned fibers or channels within a matrix gives the optical effect seen in the minerals cat’s eye and tiger-eye; the effect is termed chatoyancy. The crystals are cut en cabochon, with the long axis parallel to the fibers because you don’t need the internal reflection from the surfaces. Tiger-eye (or tiger’s eye) is quartz that contains oriented fibers of crocidolite. The mineral started as crocidolite (a form of asbestos) and was partly replaced by silica, which then becomes the matrix to the crocidolite fibers. This effect is shown together with others caused by interference in Figure 36.14. Precious, or oriental, cat’s eye is the rarest and most highly prized form of chrysoberyl; it is a green mineral called cymophane. The chatoyant effect is due to parallel arrays of pores. Asterism. This term refers to the star effect that can occur in sapphires, rubies, and garnets. The effect is illustrated for sapphire, where it is often the strongest (Figure 36.15). In this case, aligned precipitates in the single crystal cause the optical effect. Because sapphire has a threefold axis, the precipitates reproduce this symmetry by aligning along particular directions giving the sixfold star. In black sapphire, the needles are hematite; in essentially all other sapphires, they are rutile. Some star sapphires from Thailand contain both hematite and rutile (sometimes called silk because they are so fine) and show 12-fold stars. In star garnets and diposide, the star shows fourfold symmetry. In garnet, the crystals lie at 90 along [001] directions; in diopside, they lie at 73 to one another. Iridescence. The best known form of iridescence is opalescence. The reason opals show different colors when viewed in different directions is that the silica spheres are just the right dimensions to diffract light, as illustrated in Figure 36.16. Labradorescence and adularescence (shiller). These are two forms of iridescence caused by planar interference in a transparent mineral. Labradorite is a plagioclase

686

FIGURE 36.14. Chatoyancy: scattering of light by aligned fibers or precipitates.

feldspar named after the Labrador Peninsula in Canada; spectrolite is a special feldspar found in Finland. Labradorescence results from the presence of lamellar intergrowths inside the crystal; two phases of slightly different composition separate out as the mineral cools. The interference occurs when light is reflected from the different interfaces. The color we see (illustrated in Figure 36.17) depends on the effective thickness of the layers and thus on the viewing angle. Moonstone is transparent feldspar that shows this effect. The planar defects in these feldspars are actually lamellar twin boundaries arising from chemical twinning.

36.8 IDENTIFYING MINERALS AND GEMS Because many gemstones appear quite similar, which is why they can be simulated, it is important to be able to distinguish the real from the synthetic or from the simulant. If you have scanning electron microscopy (SEM) available, the latter task is not a problem; but usually this is not the case in the field (or the shop). The basic tests are for thermal conductivity, optical properties, and mechanical

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

FIGURE 36.16. Origin of opalescence: Bragg scattering of light from a surface of ordered spheres. Best known in opals.

FIGURE 36.17. Laboradite causes an iridescence due to scattering by twin planes in the feldspar crystal.

FIGURE 36.15. Asterism: effect of precipitates oriented along several different directions, giving rise to the star in star sapphire.

properties. Using a mechanical test to characterize a material that you don’t want to damage is clearly tricky, so the use of hardness measurements is directed more toward minerals than gemstones.

36.8.1 Hardness (Toughness) Indentation has been discussed in Section 16.3. Although these tests could be used for gemstones, they are not, except as a calibration. The Mohs scratch hardness scale

36.8 I D E N T I F Y I N G M I N E R A L S

AND

is much more popular because the principle is to test what the stone does scratch, not what scratches the stone. The Mohs hardness test (the word scratch is assumed) is almost a nondestructive test. The hardness of a gemstone is usually referred to as its Mohs hardness. As this hardness value is determined by a scratch test, it is not actually a hardness. The scale has many drawbacks, including the fact that it is not linear, it does not necessarily relate to wear resistance, and it damages the specimen—so it’s not ideal for polished stones. Table 36.6 gives Mohs hardness values for the Gem Scale and for some other materials as a comparison. (Note that in Chapter 16 we consider the extended version as defined by Ridgeway but it is not nearly so widely used in the gem industry). The table also includes a “relative” hardness scale. Remember that minerals are anisotropic, so the Mohs hardness of kyanite is ~4.5 when scratched parallel to the long axis and ~6.5 when scratched perpendicular to the long axis. Incidentally, this material is not

G E M S .............................................................................................................................................

687

TABLE 36.6 Classical Mohs Hardness Scale for Gems Mohs #

Classic Abs # mineral

1 2 3

Other materials

1 Talc 3 Gypsum 9 Calcite

Pencil lead (graphite) Your fingernail (2.2) Chalk (3), gold (2.5–3.0), dolomite (3.5–4.0), ZnS (3) 21 Fluorite Copper penny (3.2) 48 Apatite Knife blade (5.0), window glass (5.5), strontium titanate (5.5), sodalite (5.5–6.0), hematite (5.5–6.5) 72 Orthoclase Steel file, other feldspars, pumice (6), pyrite (6.5), magnetite (6), porcelin (6–7), Anatase (5.5–6.0) 100 Quartz Streak Plate (7.0), zircon (6.5–7.5), olivine (6.5), garnet (6.5–7.5), rutile (6.5) 200 Topaz Spinel (8), YAG (8), ZrO2 (8), chrysoberyl (8.5) 400 Sapphire WC (9) 1,600 Diamond Scratches everything! B4C3 (9–10), SiC (9–10)

4 5

6

7 8 9 10

FIGURE 36.18. Pocket-size instrument for measuring thermal conductivity.

are already mounted (e.g., in a ring). The instrument supplies the heat, and the sink is either the ring or another part of the device. The device shown in Figure 36.18 measures the change in temperature of its tip when the tip is placed in contact with the gemstone. When testing the poorer thermal conductors, the temperature of the tip is allowed to fall until it reaches a first set value, when the subsequent decrease to the next set value is timed—much like taking a blood pressure reading.

36.9 CHEMICAL STABILITY (DURABILITY) TABLE 36.7 Thermal Conductivity Mineral

Thermal conductivity (Wm1 C1)

Diamond Synthetic moissanite Silver Copper Gold Platinum Corundum Zircon (high) YAG GGG Rutile Quartz CZ Glass

1,000–2,600 200–500 430 390 320 70 40a 30a 15 8 8a 8a 5 1

a

Mean value between c axis and z axis directions

confused with kaolinite using this test because the latter has a Mohs hardness of 2.0–2.5. An additional problem in using mechanical tests for minerals is that many tend to cleave. Wear (or abrasion) might be a better test, but it’s more difficult for the gemologist.

36.8.2 Thermal Conductivity The thermal conductivity of gemstones provides a useful way of identifying a mineral in the field. Table 36.7 lists values used for gemstones. The common assumption is that gemstones are poor conductors of heat. Diamond is actual a much better thermal conductor than Cu. The measurement of thermal conduction of gemstones is particularly attractive as a test because it can be applied to stones that

688

Gemstones are usually thought of as being chemically stable. Some are; some are certainly not. Opal contains a significant amount of water; if this water is removed, the opal fractures and degrades. Emeralds are slightly different in that they usually already contain many fractures that have been filled with oil or polymer. If this filler is removed—for example, by cleaning with a solvent—the fracture may extend. Even though the diamond in an engagement ring is durable, the 18 k gold setting holding it in place may not be, especially if it is exposed to chlorinecontaining liquids, such as in a swimming pool or hot tub.

36.10 DIAMONDS, SAPPHIRES, RUBIES, AND EMERALDS Why diamond? Yes, it’s hard, but it is its optical qualities (and great advertising) that have made it so popular. The refractive index of diamond is 2.42, whereas rutile (once proposed as a diamond simulant) and moissanite (now being used as such) have refractive indices of 2.6/2.9 and 2.65/2/69, respectively, so they do not have the best optical properties (for internal reflection). Diamonds can now be synthesized to weigh more than 0.6 g (3 ct). This is an art in regard to the color and clarity. The production of gem-quality synthetic diamonds (see Section 29.14) gives a deep yellow color. Colored diamonds tend to be more valuable than colorless ones simply because they are more rare. The Dresden green diamond weighs 40.70 ct and is the largest known green diamond in the world. Natural red, pink, and yellow diamonds can also command prices of near $1 million per carat. The danger is that diamonds can also be artificially

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

colored by irradiating them; diamond was perhaps the first TABLE 36.8 Famous Diamonds irradiated gem. Blue B-doped diamond is a semiconductor, Name Weight Notes whereas blue irradiated diamond is still an insulator. 6 A concentration of 10 B in diamond gives it a deep Hope 45.52 ct Blue due to B doping L ¼ 25.60 mm, W ¼ 21.78 mm, blue color as seen in the Hope diamond. A B-doped diaD ¼ 12.00 mm mond fluoresces under ultraviolet (UV) light and continues In the Smithsonian to glow for minutes afterward. 105.60 ct Originally 186 carats Koh-I-Noor Diamonds are actually mined in large quantities in (Mountain of Tower of London South Africa and Russia, with newer sources in Sierra Light) Leone and northeast of Yellowknife in Canada’s NorthCullinan I (Great 530.20 ct Original rough Cullinan Diamonds west Territories. Natural diamonds are created 150 km Star of Africa) 3,106 ct Tower of London (10
6
5 cm) beneath the earth’s surface and are transported to the Cullinan II (Lesser 317.40 ct In the British Imperial Crown surface by volcanic activity. The Kimberley Pipe is the Star of Africa) Tower of London remains of such volcanic activity. A pipe is a carrot-shaped The Regent 140.50 ct In the Louvre volcanic neck; there is a cluster of 11 pipes at Kimberley— The Centenary 273.85 ct Found in 1986 (de Beers Centenary). they are about 1.2 billion years old. The largest kimberlite Rough wt. 599 ct. In the Tower of pipe is now the Premier Mine at Cullinan, which produces London 1.6 million ct annually. The original location of kimberlite The Tiffany Yellow 128.51 ct Found in De Beers mine, Kimberly is now known as Kimberley’s “Big Hole.” 1887. Rough weight 287.42 ct Still at Tiffany’s in New York Nitrogen causes both synthetic and natural (e.g., the Tiffany Yellow) diamonds to be yellow. A Florida company has up to 200 diamond-making growth chambers, to give the current 45.52 ct. We say “presumably” because each weighing about 4,000 lbs (cost ~$50,000 each). Each the diamond disappeared for 20 years following looting after chamber can produce eight 3-ct rough stones per month. A the French Revolution in 1792 (the current Hope could little history: a team of five Russian scientists in indeed fit inside the old Novosibirsk, Siberia, manHope). The Presidente aged to create gem-quality Vargas was found in the DIAVIK DIAMOND MINE diamonds at the relatively Rio Santo Antonio in Production began in 2003 in Lac de Gras 300 km north low pressure of Minas Gerais and weighed east of Yellowknife; it may be the richest diamond lode 60,000 atm. in 1989, 726.6 metric carats in the world. It has estimated reserves of 30 million tons which avoids the high pres(56.2
51.0
24.4 mm) and yields over 3 ct per ton of ore (more than three sure of the so-called GE but was cut into 29 separate times the world average). In 2009, 5.6 million ct were process used to make stones. extracted valued at over US$ 1.3 billion. industrial diamond. Carter Figure 36.20 shows the Clarke, an American entredifferent facets found on preneur, paid $57,000 in natural diamonds; all these shapes just involve {001} and 1996 for a “diamond-growth chamber” during a business {111} surfaces. Modern diamond faceting does not need to trip to Russia; the chamber was the size of a washing take into account the crystallography because, as a facemachine. He then founded Gemesis. centered cubic (fcc) crystal, diamond is structurally quite For some time, mixtures of H2 and natural gas have isotropic. been used for chemical vapor deposition (CVD) growth of diamond-like carbon (DLC) films. It is now possible to use this technique to grow diamond seed crystals to produce clear, perfect, colorless diamonds. Diamonds grown by the high-pressure methods are invariably doped and thus colored. One company, Apollo, has used the technique to grow 1-ct diamonds. Examples of famous diamonds are listed in Table 36.8. One feature that is obvious when you think about it: these gemstones were all cut from much larger rough stones. When first extracted from the Premier Mine near Pretoria, the Cullinan diamond weighed 0.621 kg. It was later cut into 9 major gemstones (the largest 2 being in the list and shown in Figure 36.19) and 96 other stones. The Hope diamond, thought to be from the Kollur Mine in Golconda, India, was 3 first cut into a triangular 112 16 -carat stone. In 1673, it was FIGURE 36.19. Two examples of blue diamonds. Both were cut from 1 recut to give a 67 8 -carat stone and then presumably cut again the same Cullinan rough diamond.

36.10 D I A M O N D S , S A P P H I R E S , R U B I E S ,

AND

E M E R A L D S ..................................................................................................................

689

FIGURE 36.20. Different diamond shapes produced by diamonds faceting on three different crystallographic planes while keeping the same symmetry.

TABLE 36.9 Beryl (Be3Al2Si6O18) Name

Color

Dopant

Beryl Emerald Aquamarine Morganite (rose beryl) Heliodor Goshenite

Blue Green Blue/light green Pink Gold-yellow None

Cr Cr or V Fe Mn Fe

Also in red beryl

36.10.1 Sapphire and Ruby Sapphire and ruby are both mainly Al2O3. If the stone is red, it is called ruby; if it’s any other color, it is called sapphire or fancy sapphire. There are many ways to produce the color, as summarized in Section 36.6. Ruby is Crdoped, whereas the blue gemstones contain the Fe-Ti complex (see Section 36.7). Sapphires can be colored by diffusing in dopants. Co gives blue, and Be gives yellow. Doping with Be is quite new; the Be diffuses much more quickly that Co and can penetrate the whole crystal, so there are no telltale effects at the facets. The process involves surrounding the sapphire with chrysoberyl (BeAl2O3) powder and heating to ~1,300 C.

36.10.2 Emerald and Beryls The mineral is generally referred to as beryl and is found is several forms, as summarized in Table 36.9. Commercially, the mineral beryl is the principal source of Be. Beryl occurs in three forms: emerald, aquamarine, and precious beryl (everything else); this is somewhat analogous to the naming of sapphires. Some precious beryls do have their own names in the gem trade. In the field, gemologists use the Chelsea filter specifically to identify emeralds. The Chelsea filter is a dyed gelatin film designed to transmit the red but absorb the green; through it, emeralds appear red. The filter became less useful when synthetic emeralds became common. The formula is Be3Al2Si6O18 (with up to 1 H2O); it has n ¼ 1.595. The crystal structure is hexagonal and is composed of sixfold rings of SiO4 tetrahedra, which make up

690

FIGURE 36.21. Emerald cut in the classical style that minimizes the likelihood of fracture.

the Si6O18 unit. It cleaves along both {10
10} and (0001) faces; fractures on these planes do not need to break Si-O bonds. Actually, cleavage is so easy that essentially all natural emeralds contain fractures. The so-called emerald cut, shown in Figure 36.21, has the corners removed; it was developed to minimize the likelihood of fracture (while maximizing the size of the stone). Synthetic emerald can be grown by the flux method or hydrothermally (like synthetic quartz and natural emeralds), as shown in Figure 29.15, and is much more perfect. The largest natural single crystal of gem-quality aquamarine was found in Minas Gerais and weighed 110.5 kg. In one form of aquamarine known as Maxixe, the blue is enhanced by irradiation, but the color is not permanent. The coloring of morganite can often be improved by heating the gem to >400 C. The same treatment can change green beryl into blue aquamarine. The likely effect is that the process reduces Fe3+ to Fe2+. Not included in Table 36.9 are Co and Ni doping, which produces pink/violet and pale green, respectively.

36.11 OPAL Natural opal was deposited in fissures in rocks or fossilized (silicified) wood from water-based solutions at relatively low temperatures, as illustrated in Figure 36.22. Precious opal consists of perfect arrays of identical spheres of SiO2, as shown in the SEM image in Figure 36.23. The spheres

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

FIGURE 36.22. Natural opal in a vein. FIGURE 36.24. Polished precious opal.

FIGURE 36.25. Amethyst crystals from a section of a cathedral. The crystals were initially inside the geode.

FIGURE 36.23. Scanning electron microscopy image of a Gilson opal.

have a radius of ~300 nm, so arrays of the spheres look like crystals (three-dimensional diffraction gratings) to incident visible light. Thus, the light is diffracted (in the actual “reflection” Bragg geometry), which is why we see different colors when viewing the opal from different directions. The spheres can be amorphous or partially crystalline. Not all opal shows these colors because the term opal refers to any material made up of such SiO2 spheres; other examples are common opal or fire opal. A key component of opal is the included water, which is typically present at 3–9 wt.% but may be as much as 20 wt.%. It is a little tricky to polish opal (a fine example is shown in Figure 36.24), partly because you mustn’t remove the water and partly because it is quite soft. Opal thus has much in common with chalcedony, another form of fine-grain quartz.

Opals are synthesized commercially, as we saw in Chapter 28. Because the structure is quite open, it is easy to diffuse a dye into the matrix to change the overall color. We can produce inverse opals in a similar way using latex spheres instead of SiO2 and then various sols such as TiO2 instead of the dye. When the impregnated polymer is burned out, the inverse opal has potential as a photonic band gap material. The last point reminds us that opal was the first photonic material; we are now exploring how we can exploit this natural phenomenon using synthetic materials.

36.12 OTHER GEMS Quartz crystals are known by several different names. The colorless form is known as rock crystal. Amethyst is single-crystal quartz. The crystals grow naturally by a hydrothermal process and are found as cathedrals and wheels, as shown in Figure 36.25. The purple color can be caused by Fe that is in the excited 4+ state due to natural (or artificial) irradiation or Mn. The Fe concentration is probably in the range 20–40 ppm. Smoky quartz is gray

36.12 O T H E R G E M S .....................................................................................................................................................................................

691

also due to irradiation, but the dopant is probably Al3+. When heat-treated, the stones become yellow-orangebrown and are known as citrine. Rhinestones were originally quartz pebbles collected from the River Rhine. The lesson is that each of these forms of quartz grew by natural hydrothermal processes and were then “treated” by nature (or humans). Topaz is a silicate with the general formula Al2SiO4 (F2OH)3. It is the hardest silicate, testing at 8 on the Mohs scale. The crystal structure is unusual. It is orthorhombic, consisting of chains of AO6 octahedra linked together with SiO4 tetrahedra. Cleavage is parallel to the basal plane as there are no Si-O bonds crossing this plane. It is found in a variety of shades of blue and the Imperial Yellow. Figure 36.26 shows a natural single crystal. Color is

produced either by high-energy neutron irradiation in a nuclear reactor or in a gamma cell (in Brazil they use 60 Co to produce g-rays) or with a linear accelerator (linac), producing a high-energy electron beam. The stones may be radioactive for some time after processing (a few weeks for the sky blue produced by the linac; several months for the London blue produced with neutrons). One story concerns a gem dealer who was surprised to find that the gems in his pocket were still hot. Gamma radiation can produce both yellow and blue centers, giving a brown color; the yellow can be annealed out at ~450 C, leaving the blue, which lasts a lifetime (but may not be an heirloom). Tourmaline is another mineral that can show many colors because it can contain different cations, some of which are listed in Table 36.10. This table is quite new and still being developed. For example, the Y ion in dravite can be replaced by Mn(II) or V(II); Mn doping can be as high as 9 wt.% and makes tourmaline pink. Some cations tend to share the Y site, compensating charge. Color variation is especially known for the variant watermelon tourmaline, which is naturally pink in the interior and green on the outside, as illustrated in Figure 36.27. As always, we have to be careful because electron irradiation can change TABLE 36.10 Tourmalines

FIGURE 36.26. Topaz. Examples of irradiated rough, natural imperial topaz and a huge natural single crystal.

692

Species

X

Y3

Z6

T6O18

V3

W

Elbaite Dravite Chromdravite Schorl Olenite Buergerite Uvite Rossmanite Foitite

Na Na Na Na Na Na Ca Ca vac

Li1.5Al1.5 Mg3 Mg3 Fe(II)3 Al3 Fe(III)3 Mg3 LiAl2 Fe(II)2Al

Al6 Al6 Al6 Al6 Al6 Al6 MgAl5 Al6 Al6

Si6O18 Si6O18 Si6O18 Si6O18 Si6O18 Si6O18 Si6O18 Si6O18 Si6O18

(OH)3 (OH)3 (OH)3 (OH)3 O3 O3 O3 (OH)3 (OH)3

(OH) (OH) (OH) (OH) Black F F F (OH) (OH)

Color

FIGURE 36.27. Watermelon tourmaline.

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

yellow-brown tourmaline into pink tourmaline. The terms red tourmaline and blue tourmaline have now replaced the names rubellite and indocolite because they are not distinct minerals. Natural crystals of schorl can be >15 cm inches long. Tourmaline is an unusual crystal in that it has true threefold symmetry; the space group is R3m with a ¼ 1.594 nm and c ¼ 0.7138 nm, but the actual values of a and c depend on which cations are present. Commercially, this mineral is a principal source of boron. We mentioned the piezoelectric property earlier; it is strongest along the c-axis but quartz is now used more often except in special pressure sensor applications. Spinels are less common gemstones but sometimes quite famous even if known by another name. The Black Prince’s Ruby (5 cm long) and the Timur Ruby (polished but not faceted, 361 ct) in the English imperial state crown are both spinels. The largest known natural spinel crystal weighed 104 g (520 ct). Although the spinel structure can accommodate wide variations in chemistry, most spinel gemstones are actually naturally doped MgAl2O4. One variation is gahnite, the zinc aluminate spinel, ZnAl2O4. The best known spinel, magnetite (FeFe2O4), is not very attractive as a gemstone. Synthetic spinels can be doped in many colors and have the advantage over synthetic sapphire that they are more nearly isotropic. Garnet crystals are some of the older popular gemstones and a special class of mineral (much like spinel). The crystal structure is able to accommodate many different cations, which then produce different colors. All those shown in the box are natural silicates, although several synthetic garnets are now available; synthetic garnets produced for gemstones are usually doped yttrium aluminum garnet (YAG). All natural garnets have the general formula R3M2(SiO4)3, where R is Ca, Mg, Fe(II), or Mn; M is Al, Fe(III), or Cr. They form two groups.

FIGURE 36.28. As grown crystal of rare-earth-doped cubic zirconia.

FIGURE 36.29. Single-crystal of rare earth (RE)-doped synthetic cubic zirconia that has been cut and polished.

similar to almandite. Almandite is Dana’s name for almandine, which is also known as precious garnet and was probably called alabandine (after the ancient Turkish city of Alabanda) by Pliny, so the two names are used interchangeably. Cubic zirconia is a special case in that it is ubiquitous Pyralspites as a synthetic gemstone; an example of an as-grown stone Ugrandites is shown in Figure 36.28 and a polished and faceted stone in Figure 36.29. Naturally occurring zirconia is rare and Natural garnets are rarely pure. Natural almandines usuthen only as the monoclinic baddeleyite. ally contain Ca, Mg, and Fe3+ so they denote the end-member Peridot is better garnets. We then use whichknown as the mineral olivever name most closely ine. The gemstones usumatches the composition SILICATE GARNETS ally have a composition we have. There is then a Pyralspites: named according to the dominant R cation close to (Mg0.9,Fe0.1)2 whole extra array of names. present: SiO4 and have a unique Rhodolite lies between 2+ Mg : Pyrope [Mg3Al2(SiO4)3] green color. Peridot was pyrope and almandite; Fe2+: Almandine [Fe3Al2(SiO4)3] mined for 3,500 years on tsavorite is a green variation 2+ Mn : Spessartine [Mn3Al2(SiO4)3] the island of Zabargad of grossular; demantoid (the The dominant M cation in these garnets is Al; usually (St. John’s Island) in the most valuable garnet) and 3+ with Fe Red Sea near Aswan, the black melanite are Ugrandites (calcic garnets with R ¼ Ca): named after Egypt. The largest cut perandradite garnets. Pyrope is the dominant M cation. idot (319 ct) was found on also known as Bohemian 3+ Cr : Uvarovite [Ca3Cr2(SiO4)3] this island. garnet and was the favorite 3+ Al : Grossular [Ca3Al2(SiO4)3] Alexandrite is a variety red stone in the 1700s and 3+ Fe : Andradite [Ca3Fe2(SiO4)3] of chrysoberyl. It is special 1800s but can look very

▪ ▪

36.12 O T H E R G E M S .....................................................................................................................................................................................

693

because absorption is so different in two directions. When viewed along the different crystallographic axes, its color changes from red to orange-yellow to emerald-green. In daylight the color is green; in artificial light it is red. Tanzanite is a purple variety of zoisite. Most stones are green when mined and become purple when heat-treated. The stone is particularly interesting (valuable) because it is only found in one place. It is quite hard (Mohs 6.5–7.0), can be essentially perfect, and is pleochroic.

36.13 MINERALS WITH INCLUSIONS Inclusions in gemstones are quite common and are often used as an indication that the stone is natural. The bestknown gems with inclusions are the star stones, where the stone can be sapphire, garnet, etc. Perfect star sapphires can now be synthesized. Quartz crystals containing a hematite seed and the rutile needles are particularly interesting. You can also see hematite needles in quartz. When considering the origin of such

FIGURE 36.30. Rutile growing in hematite. This combination often causes patterns in quartz crystals.

structures, you should know that the same arrangement of seed and needles occurs without the quartz matrix, as shown in Figure 36.30. The name cat’s eye originally referred to chrysoberyl, which contains inclusions, but the effect is seen in many other gemstones as well.

36.14 TREATMENT OF GEMS Two methods that are widely used to enhance the color of gemstones are irradiation and heating, usually in that sequence. Table 36.11 summarizes some of the heat and irradiation treatments that have been used. The reason for treating (processing) gemstones is invariably to enhance their appearance and thus increase their value. We review the general features of the different treatments and the science behind them here but refer you to the sections on particular materials for discussions of the details. All irradiated samples are heated. Heating. Sapphires. Over-heating can cause good stones to become so dark that they are not transparent. Most citrine is produced by heating amethyst. This process can also occur naturally. Irradiation. Topaz is sold as Swiss blue, London blue, etc. In all cases, the material has been irradiated and heated. Composite stones can be constructed by joining different stones, as illustrated in Figure 36.31. One of these methods is essentially equivalent to making a bicrystal with an amorphous (polymer) intergranular layer. Filling cracks with oil, polymer, or glass is commonplace. The different procedures produce a similar result, the difference being mainly how long it takes until the result degrades. The extreme example of this is the transformation of turquoise chalk to something that looks like the real thing, as shown in Figure 36.32. In 2004, some highly priced rubies were found to have had internal fractures filled with Pb-containing glass (chosen to match the n).

TABLE 36.11 “Treatment” of Gems: Its Effectiveness and “Publicity” Treatment

Stable?

Detectable?

Disclosure?

Aquamarine turned from green to blue by heat Zircon heated to turn colorless or blue Sapphire or ruby heated to remove silk Sapphire heated to modify or develop a blue color Topaz turned blue by irradiation Topaz or sapphire irradiated to a yellow or brown color Beryl irradiated: Maxixe-type blue color Turquoise or opal impregnated with a colorless stabilizer Emerald or ruby impregnated with a colorless substance Beryl or emerald impregnated or coated with color Sapphire impurity diffused to produce a surface color or surface asterism Fracture-filled diamond

Yes Virtually all Yes Yes Yes No No Usually Variable No Yes/No

No No, but these colors are very rare in nature Usually Usually No No, only fact of fading Yes Usually Yes Yes Yes

No No Yes Yes Explained Explained Yes Yes Yes Yes Yes

Yes/No

Yes

Yes

694

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

FIGURE 36.31. Different methods of simulating (faking) gemstones by forming composites (e.g., bicrystals).

Lasers are used to drill channels into diamonds to remove internal blemishes, afterward filling the hole with glass having a similar n. This is a variation on the technique is used to created three-dimensional images inside glass blocks for use as decorations. In either process,

36.14 T R E A T M E N T

OF

a pulsed laser is focused on a small area at the required location in the crystal and then pulsed to create an internal fracture (in the glass model) or ablating material (in the gemstone). This process is illustrated in Figure 36.33.

G E M S .....................................................................................................................................................................

695

FIGURE 36.34. Diffusion to modify the color of a gemstone.

Hence, after diffusion, the stone should be equilibrated for a long period to remove the concentration gradient. This equilibration would take too long for the gem trade, so the dopant has a concentration profile peaking at the surface. You diffuse dopant into the cut stone because otherwise the surface region has a different color when you facet it. In the case of Co, which produces a blue color in sapphire, diffusion is slow so that the facets have a color different from that of the bulk. The color difference is particularly obvious at the facet junctions, as shown in Figure 36.34.

36.15 THE MINERAL AND GEM TRADE FIGURE 36.32. Examples of turquoise (A) before and (B) after processing. Both samples are real turquoise. An old method of infiltration to modify the properties of a material is common for turquoise.

FIGURE 36.33. Laser etching to remove inclusions in a gemstone.

How effective diffusion is depends on the rate of diffusion. If the intention is to improve the color of a gemstone, then you want the color to be uniform throughout the stone.

696

There are many interesting international aspects to the history of gems. Economies of countries have been based on gem production. Though not usually as bad as preciousmetal mines, the effect of gem mines on the environment can be very negative. The Kimberley Mine is always associated with “Big Hole,” which by 1914 when work on it stopped was the largest man-created excavation in the world, having a depth of 215 m, a surface area of ~17 ha, and a perimeter of ~1.6 km; 22.5 million tons of earth were excavated to produce 2,722 kg of diamonds. In the 1990s and later, the diamond trade in Sierra Leone became associated with raising funds to fund ongoing wars. The stones were referred to as “blood diamonds,” and international organizations tried to minimize the trade in these stones. Now these diamonds account for 0.2% of diamonds sold, down from a peak of 4%. It is often very difficult to reach the mines in Myanmar (Burma), but historically Magok is the center of the rubyproducing region. Many gemstones are exported through Thailand, which is a worldwide center for processing (cutting and polishing) gemstones. In Brazil, the most important production area for gems and minerals in Minas Gerais. It is a more remote region of Brazil but with a growing industrial presence.

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

BLOOD DIAMONDS IN POP CULTURE Diamonds from Sierra Leone by Kanye West won a 2006 Grammy for Best Rap Song. Blood Diamond starring Leonardo DiCaprio and Jennifer Connelly was released by Warner Brothers in 2006. FarCry2 a 2008 video game set in a Central African country in the midst of civil war uses diamonds as currency. The book Diamonds in the Shadow by Caroline B. Cooney (Delacorte Books for Young Readers, 2007) involves a refugee family smuggling blood diamonds into the United States from Africa. In 2010 war-crimes trial of Liberian leader Charles Taylor, supermodel Naomi Campbell admits he gave her “dirty rocks.” In India, the major source of talc is located in the Dagota District. The talc is prepared by crushing boulders that have been produced in the soapstone mines.

Mining is threatening the existence of the Indian tiger (see Section 37.6). The history of De Beers and the diamond trade has been covered in several books, so we do not examine it here except to say that it makes for a fascinating story. It surprises some to hear that diamonds are not rare. One aspect of the diamond business is that the stones are very portable and easily can be made unrecognizable. The company De Beers is now based in Amsterdam and Antwerp. A new development is to use a laser to inscribe each diamond with a code number after it has been faceted (another “modification” using lasers). The diamonds from Canada’s Northwest Territories have a laser-inscribed polar bear and are known as “Polar Bear Diamonds”. There are, of course, many other uses of gemstones, including jewel bearings for watches and other precision machinery; but here we have concentrated on their use in decoration and the science behind the preparation of the gemstone from the rough. There is also a worldwide trade in mineral specimens such as you may see in museums that can be priced in excess of $100,000.

CHAPTER SUMMARY The topic of this chapter is unusual in a ceramics textbook. It is an example of real-world ceramics, where mineralogy, chemistry, physics, materials science, art, and commerce meet.

PEOPLE AND HISTORY Cullinan, Sir Thomas owned the mine where, in 1905, the world’s largest diamond was found. De Beer, Johannes Nicholas and Diederik Arnoldus are brothers who owned the farm that became the Kimberley “Big Hole.” Mohs, Fredrich (1773–1839) introduced the term “scratch hardness” in 1826. He was born in Gernrode/ Harz Germany, studied at the University of Halle and at Freiberg. He later worked in Austria. Moisson, Ferdinand Frederic Henri (1852–1907) discovered naturally occurring SiC in 1905 in a meteorite from the Diablo Canyon in Arizona (USA). He developed the electric furnace, which he used to make carbides and prepare pure metals. He received the Nobel Prize in 1906 for successfully isolating fluorine (1886). Winston, Harry (1896–1978) A key figure in the diamond trade, he opened his business in New York City in 1932. In 1958, he donated the Hope Diamond to the Smithsonian.

EXERCISES 36.1 What are the lines in Figure 36.9? 36.2 In Figure 36.11 the stones are immersed in a liquid. Why is this liquid chosen? Show that the observations are what you would expect. 36.3 What can you deduce regarding the size, shape, and alignment of the particles causing the stars in Figure 36.15? 36.4 Explain the phenomenon of labradorescence seen in Figure 36.17. 36.5 What is the common flaw found in natural emeralds? Explain your answer from a crystallographic point of view. 36.6 If you had a good means for measuring thermal conductivity, would you prefer such a test to Mohs’ scratch test? How sensitive would your apparatus need to be? We can use a handheld tester to distinguish diamond and moissanite. How is this fact connected to the electronics industry? 36.7 Why must you be particularly careful when polishing opal? How is opal related to today’s electronics industry? 36.8 What do the stabilization of turquoise, the treatment of emerald, and ZnO varistors have in common? 36.9 Diamond has a high n and is also a very hard material. (a) Are these two features linked? (b) If so, explain why SrTiO3 has a higher n but is not as hard. C H A P T E R S U M M A R Y .................................................................................................................................................................................

697

36.10 Explain, using your knowledge of ceramic processing, how you might take turquoise powder and turn it into a gemstone. 36.11 We answered the question: is this turquoise natural? Clarify the answer. 36.12 When asked how to wash an opal, we answered: “No.” Why? 36.13 Diamonds have now been found in Zambia. Where else are they found worldwide? 36.14 Diamond is known for its sparkle. Which gemstones sparkle more? Why are they less valuable than diamond? 36.15 Watermelon tourmaline ranges in color from green to pink. Did the color affect its use as a piezoelectric? Explore and discuss. 36.16 You can make green tanzanite purple, white turquoise “blue,” and colorless topaz blue. What are you doing in each case, and how does each one relate to more technologically relevant ceramics? 36.17 Beryllium can be diffused into sapphire to produce a yellow color that is quite rare in nature. How would you, as a ceramist, show that it had been doped? 36.18 Talc has been used for many years. Discuss how its crystal structure influences its use. 36.19 Natural garnet occurs in many different colors. So does synthetic cubic zirconia. Discuss what features these two materials have in common to facilitate this coloration. 36.20 Why can we say that opal was the first photonic material?

GENERAL REFERENCES Hughes RW (1997) Ruby and sapphire. RWH Publishing, Boulder, A great book by the sapphire guru; beautiful illustrations Hurlbut CS, Kammerling RC (1991) Gemology, 2nd edn. Wiley, New York, The 1991 edition has trigons on the cover Johnsen O (2002) Photographic guide to the minerals of the world. Oxford University Press, Oxford, Another excellent pocket guide Nassau K (1994) Gemstone enhancement, 2nd edn. Butterworth-Heinemann, Oxford, The book on the topic. Easy reading and fascinating details (see also his books on crystal growth) Nassau K (ed) (1998) Color for science, art and technology. Elsevier, Amsterdam, A collection of articles describing the origins of color in gemstones Nassau K (2001) The physics and chemistry of color, 2nd edn. Wiley-Interscience, New York Read PG (2008) Gemmology, 3rd edn. Robert Hale Publishers. 2 mm’s in UK English. This is a classic manageable text at a similar level to this one on ceramics though aimed at the practicing gemologist Schumann W (2009) Gemstones of the world, 4th edn. Sterling Publishing, New York, This is the pocket book Smith GFH (1972) Gemstones, 14th edn. Chapman & Hall, London, Worth a trip to the library Ward F (2003) Rubies & sapphires, 4th edn. Gem Book Publishers, Malibu (also Emeralds, Opals, Pearls, Jade, Diamond)

SPECIFIC REFERENCES Guinel MJ-F, Norton MG (2006) The origin of asterism in almandine-pyrope garnets from Idaho. J Mater Sci 41:719, Microscopy study showing origin of the “star” in star garnets and why both 4- and 6-ray stars are possible. Published in the 40th Anniversary issue of Journal of Materials Science Muller H (2009) Whitby JetShire Themelis T (1992) The heat-treatment of ruby and sapphire. Gemlab, Clearwater Yavuz F, G€ultekin AH, Karakaya MC ¸ (2002) CLASTOUR: a computer program for the classification of the minerals of the tourmaline group. Comput Geosci 28:1017

WWW http://www.t-matrix.de/. Sites for light scattering www.kruess.com/. Makers of gemologists equipment www.diavik.ca/. The Diavik diamond mine in Canada www.debeersgroup.com/. Website for De Beers; includes a history of the company www.debeers.com/. Where to buy their diamonds!

WHERE TO SEE GEMSTONES The Smithsonian Institution, Washington, DC. www.minerals.si.edu The Tower of London http://www.hrp.org.uk/toweroflondon/stories/crownjewels.aspx Muse´e du Louvre, Paris www.louvre.fr

698

....................................................................................................................................................................... 36

MINERALS

AND

GEMS

37 Energy Production and Storage CHAPTER PREVIEW One of the grand challenges for the twenty-first century is to produce sufficient clean energy for the world’s growing population and to meet the increasing demand among rapidly developing countries. Annual world energy consumption now exceeds 500
1018 J (500 exajoules) and, according to the U.S. Energy Information Administration (EIA), is projected to exceed 650
1018 J by 2025. China and India lead the growth in energy demand. Although our current energy needs are met largely through the consumption of fossil fuels, future clean energy technologies rely heavily upon developments in materials science. In this chapter, we look at areas where ceramics are playing an important and, in some cases, defining role in energy generation and the related issue of energy storage. As an example, the yearly supply of energy from the sun to the earth is 3
1024 J; that is 104 times the energy we currently consume. However, we don’t see much energy being generated from solar radiation (it is responsible for generating less than 0.1% of the electrical energy used in the United States). The challenge is making solar cells more efficient and cheaper. Ceramics are critical components of a number of different solar cell technologies. In dye-sensitized solar cells, TiO2, a ceramic with a wide band gap, is used as the porous support structure for the light-absorbing material, which could be ceramic quantum dots. The term “quantum dot” should make you realize that “nano” is important in energy applications for ceramics. Nanomaterials are used for both the anode and cathode in lithium-ion batteries. In the case of the cathode, the material is nanopowder LiFePO4 This is a phospho-olivine, which takes us back to Chapter 7, where we described the olivine crystal structure.

37.1 SOME REMINDERS Ceramic materials are an essential component of devices for production and storage of energy. Some of the topics covered in this chapter are summarized in Table 37.1. In many cases, a more efficient and cleaner process can be designed through the use of catalysts, or better catalysts. The problem is that the catalyst may change during use, and it is very difficult to see what is actually happening on the catalyst surface during the reaction. As you search through the literature, you can find papers written in many different units, with the SI units of the Joule now being the standard. The conversion between these units is summarized in Table 37.2.

37.2 NUCLEAR FUEL AND WASTE DISPOSAL Uranium dioxide (UO2) has been important as a nuclear fuel since the mid-1950s and is obtained from its major ore, uraninite. After leaching the ore, an impure uranium

concentrate called “yellowcake” is precipitated out of solution. Yellowcake, which is actually brown or black rather than yellow, is typically made up of 70–90% U3O8 with some UO2 and UO3 also being present. The price of uranium is currently about $100/kg (the peak was in 2007 when it reached over $260/kg). The United States uses about 25 kt of uranium oxide each year, most of which is imported from Canada and Australia. Before use in a nuclear reactor, U3O8 is converted into UF6 gas. Then, using either diffusion or centrifugal processing, 235UF6 is separated from 238UF6. The 235Uenriched gas is reacted to form UO2 powder, which is pressed into pellets. These pellets are loaded into zirconium alloy tubes, making the fuel rods. A 1,000-MW reactor “burns” 25 t of UO2 per year, and a kilogram of fuel costs about $900. Thus, a 1,000-MW reactor consumes about $22.5 million of UO2 annually. In the United States there are presently about 100 operable nuclear reactors (over 400 worldwide), producing 21% of the countries electrical power. The number of operable reactors has decreased since it reached its peak in 1990, but there is some interest in reviving the nuclear industry

C.B. Carter and M.G. Norton, Ceramic Materials: Science and Engineering, DOI 10.1007/978-1-4614-3523-5_37, # Springer Science+Business Media New York 2013

TABLE 37.1 Energy Production Using Ceramics Source

How?

Problem

What’s special?

Fossil fuels

Combustion

Limited resources CO2 emitters

Ceramics for catalysts and catalyst supports (e.g., cracking and reforming) See Chapter 38

Bio-based fuels

Combustion

Polluting

Need catalysts Ceramics for catalysts and catalyst Various supports compositions Contain O

Nuclear

Radioactive decay

Storage

Solar

Electronic Low efficiency transitions High cost

UO2 fuel Synroc and glass for storage Support structures for dye-sensitized cells Ceramic quantum-dots Thin films Mixed oxidation states

TABLE 37.2 Energy Conversion 1 eV 1J 1 foot-pound force 1 British thermal unit (Btu) 1 horsepower-hour 1 calorie 1 Wh 1 erg 1 erg/cm2

1.60217653
1019 J 1 Newton-metre (N.m) 1.3558 J ~1,055 J 2.6845 MJ 4.184 J 3.6 kJ 107 J 1 mJ/m2

Note that the Newton meter is also the unit of torque: use the Joule instead. The Btu depends on the temperature! The mJ/m2 is the energy unit from grain boundaries. You encounter Wh even on your electricity bill.

The main method of solidifying HLW not already contained in spent fuel rods is to “vitrify” it into a borosilicate glass and cast it into stainless steel cans for ultimate burial. Vitrification of civil HLW first took place on an industrial scale in France in 1978. Environment Mechanical Low power Piezoelectrics A year’s worth of HLW from a 1,000-MW reactor can vibrations High cost One-dimensional (1D) nanomaterials be stored in about 26 m3. A second-generation immobilization material, “synroc,” is in development. This synthetic rock, because it is a zero “greenURANIUM OXIDE based on mixed titanate house emission” technolOne uranium oxide fuel pellet (about the size of a dime) phases (e.g., zirconolite, ogy and because it does not contains as much energy as 17,000 cubic feet of natural hollandite, or perovskite), consume limited resources gas, 1,780 lb of coal or 149 gal of oil. incorporates the HLW of fossil fuels. elements into its crystal One of the major structure, yielding excelproblems with increasing lent chemical stability. Synroc features leach rates more nuclear reactor capacity is what to do with the spent fuel. than an order of magnitude lower than borosilicate glass. There are two current approaches. An initially unforeseen issue in putting radioactive material, particular heavy a-emitters, into crystalline Reprocessing (adopted by United Kingdom, France, materials is that the radiation can disrupt the atomic Germany, Japan, China, India) arrangement within the crystal structure enough to Storage (adopted by United States, Canada, Sweden) cause it to become amorphous. The technique used to analyze the extent of the lattice disruption was In Europe, spent fuel is frequently reprocessed, which nuclear magnetic resonance (NMR) (see Section involves dissolving the fuel elements in nitric acid. 10.10). Figure 37.1 shows examples of NMR spectra Because plutonium is created in the fission process, from a number of samples. In each case, the sharper reprocessed fuel contains both radioactive U and Pu and peak in the spectrum arises from Si atoms in crystalline is referred to as mixed-oxide (MOX) fuel. regions (or domains), and the broader peak at lower The remaining liquid after Pu and U are removed is frequency (more negative parts per million values) is high-level waste (HLW), containing about 3% of the spent from Si atoms in amorphous regions. The results fuel. It is highly radioactive and continues to generate a lot predicted that the crystalline structure of zircon of heat. This waste must be immobilized; and because of (ZrSiO4) incorporating 239Pu would become amorphous the presence of radioisotopes with long half-lives, it must after only 1,400 years. Although this seems like a very be immobilized for tens of thousands of years. Ceramics long time, remember that these radioactive materials are key materials in this process. have to be immobilized for about 250,000 years! The major requirements for waste immobilization are: In terms of phase transformations, the a-radiation causes a crystalline-to-amorphous transformation. So The radioactive elements must become immobilized in the thermodynamically stable phase is being replaced the crystal or glass structure. by a metastable phase. In earlier chapters, such as in The leaching rate of radioactive elements must be low. Section 26.13, we were interested in crystallizing glasses The cost must be acceptable.

▪ ▪

▪ ▪ ▪

700

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

FIGURE 37.2. Solid oxide fuel cell (SOFC) auxiliary power unit (APU) designed for the tail cone of the Boeing 787. The fuel cell would operate on Jet-A fuel or bio-based aviation fuels. FIGURE 37.1. Increasing cumulative a-radiation doses cause substantive changes in the local structure around silicon atoms as amorphization of the crystals progresses. (A) Undamaged zircon. (B) UG40, a zircon with 1.2
1018 a/g. (C) Cam26, a zircon with 2.9
1018 a/g. (D) Ni12, a zircon with 7.1
1018 a/g.

37.3 SOLID OXIDE FUEL CELLS In Section 30.16 we introduced the topic of solid oxide fuel cells (SOFCs), which use an oxygen-ion-conducting ceramic membrane. Why do we want to look at this technology again in Chapter 37? There are many advantages of SOFCs; and despite the fact that they are not in widespread use, there are several potentially important applications that could be realized in the next several years. SOFCs have the potential to be a key power-generation technology in the future because:

to form glass-ceramics (i.e., amorphous-to-crystalline transformations). Whether the final HLW is vitrified material from reprocessing or entire spent fuel assemblies, it needs eventually to be disposed of safely. This means that it should not require any ongoing management after disposal. Although final disposal of HLW will not take They have high electrical efficiency (~60%). place for some years, preparations are being made for They can operate over all output ranges from kilowatts sites for long-term disposal. One of these is Yucca Mounto megawatts. tain in Nevada. Another is the Waste Isolation Pilot Plant They can run on many fuels: they are not limited to near Carlsbad in New Mexico, which has been the source hydrogen, like protonfor rock salt—hence the exhange membrane mines. The United States (PEM) fuel cells. may already have ~100 t FUEL EFFICIENCY of Pu to store. Calculations by Boeing have shown that a SOFC APU One exciting area that is The depository in would use 40% less Jet-A fuel than a conventional APU being investigated for Morsleben, Germany— (in-flight) and 75% less fuel when the plane is on the SOFCs is to replace the another rock salt mine— ground. gas-turbine auxilliary was closed for nuclearpower unit (APU) in the waste storage in 1998: the tail cone of commercial airplanes, such as the Boeing 737 domes of the salt mines have been reinforced with saltand the new Boeing 787 “Dreamliner” (Figure 37.2). To 3 concrete (480,000 m since 2003) with another offset the increase in weight, the fuel cells would have to 3 4,000,000 m needed to temporarily stabilize the lower run directly on Jet-A fuel, which is the kerosene-based fuel levels of the mines. The cost to date is 2.2 billion euros. for commercial airplanes, and eventually on next-generation The follow-up question is: where is the rest stored? The bio-based aviation fuels. The benefits would be that the fuel example of Morsleben shows that creating structures that cells would have much greater efficiency than the gas turbine can survive 30 years (the Morsleben site began storing engines, enabling a “more electric” airplane, and NOx nuclear waste in 1971) is difficult enough, but imagine emissions and other pollutants would be reduced making the challenges creating structures that survive intact for the airport environment cleaner and healthier. thousands of years!

▪ ▪ ▪

37.3 S O L I D O X I D E F U E L C E L L S ..............................................................................................................................................................

701

A key requirement is to develop a catalyst that can be used on the fuel side (the anode) of the SOFC, which is both resistant to coke formation (carbon build up that blocks the active sites of the catalyst) and is tolerant to high sulfur concentrations (Jet-A can contain 500–1,000 ppm of S). Conventional Ni-based catalysts quickly deactivate under Jet-A fuel reforming environments. A small number of ceramics have been shown to be active catalysts for reforming hydrocarbons and resistant to coke formation and sulfur poisoning. One example is molybdenum dioxide (MoO2). The catalytic activity of MoO2 can be explained in terms of the Mars-van Krevelen mechanism, which involves the consumption of oxygen ions provided by the oxygen sublattice with the purpose of sustaining the redox cycles taking place on the catalyst surface. The Mars-van Krevelen mechanism is illustrated in Figure 37.3 and described by the reaction equations 37.1 and 37.2 (these equations represent the partial oxidation of n-dodecane, which is a component of Jet-A fuel). Partial oxidation of fuel using lattice oxygen

which can be produced in large quantities in fluidized bed reactors. More efficient, but more expensive, singlecrystalline material can be used for roof installations, for farms, traffic, and literally in the field. SolarWorld uses 8-in.-diameter Czochralski-grown Si, lines up 16 ingots at a time, and cuts near-6-in. squares. Sixty of these squares are used for one panel. Using single-crystal Si means less electron–hole recombination at defects [grain boundaries (GBs)] and thus higher efficiency. The highest efficiency ever achieved with a solar cell to date is 42.3%. A value of 41.0% was obtained using a Boeing-Spectrolab multijunction GaAs cell consisting of multiple layers, as shown in Figure 37.4. The layers are deposited using molecular beam epitaxy (MBE) or metalorganic chemical vapor deposition (MOCVD). These cells are much too expensive for large-area applications and are usually placed at the focal point of mirrors or lenses that

C12 H26 þ 12O2 þ 6Mo4þ ! 12CO þ 13H2 þ 6Moð4dÞþ þ 6ð4dÞe

(37.1) FIGURE 37.3. Mars-van Krevelen mechanism.

Partial oxidation of catalyst using atmospheric oxygen 6O2 þ 6Moð4dÞþ þ 6ð4dÞe ! 12O2 þ 6Mo4þ (37.2) Molybdenum dioxide is an unusual transition metal oxide because of its high metallic-like electronic conductivity (we showed the temperature dependence of electronic conductivity of a number of ceramics in Figure 30.8). In bulk samples, s has been measured to be 1.1
104 S/cm at 300 K. The high electronic conductivity means that MoO2 could be used as an anode in direct Jet-A SOFCs (i.e., where the sole fuel feed is Jet-A; there is no separate fuel reformer). MoO2 has also been studied as a possible anode material for lithium-ion batteries. In this application, good electronic conductivity is also important.

37.4 PHOTOVOLTAIC SOLAR CELLS The basic idea of the solar cell is to use light to create excitons (electron–hole) pairs and then separate these charge carriers to produce an electric current. In Chapter 28, we studied how to deposit the thin films that are critical for the high-efficiency cells; in Chapter 29, we discussed growing the thin single crystals. Chapters 30–32 concern charge transport and generating light. These five chapters form the basis of making thin-film solar cells. Inexpensive solar cells use polycrystalline Si,

702

FIGURE 37.4. Spectrolab multijunction solar cell. Sizes up to 32 cm2 are possible with this structure.

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

concentrate the solar light by up to a factor of 1,000 (they’re called concentrator cells), or they are used for space applications (e.g., the Mars Exploration Rover), where cost is secondary.

37.5 DYE-SENSITIZED SOLAR CELLS The dye-sensitized solar cell (DSSC) is a type of photoelectrochemical (PEC) solar cell that uses a hybrid structure consisting of inorganic semiconductors and organic molecules. There are several different geometries. The one shown in Figure 37.5 uses a film of sintered TiO2 nanoparticles (10–30 nm) on a conducting glass substrate. Dye molecules that absorb sunlight are coated onto the TiO2 particles. The TiO2 itself doesn’t absorb a significant amount of sunlight; it is a wide band gap semiconductor (Eg ¼ 3.0 eV). The dye molecules absorb by electrons moving into excited states. These excited electrons are injected into the TiO2, which creates positively charged dye molecules. The incident solar energy has created electron–hole pairs. If these pairs are separated, we have a photovoltaic (or solar) cell. The circuit is completed by a liquid electrolyte and transparent electrode. The commercialization of DSSCs has been restricted because of the low efficiencies (at present), short lifetime of the photoelectrode, and the high cost of the device. Table 37.3 compares efficiencies for several different types of solar cell. There is lots of research around the world to improve the efficiency and lower the cost of DSSCs. An important component is the ceramic electrode. The electrode must be porous with a high surface area and provide a threedimensional scaffold to hold the dye molecules. Nanomaterials, such as the TiO2 nanotubes shown in Figure 29.19B, are an alternative to using nanoparticles. Semiconductor quantum dots (QDs) can replace dyes as the light-harvesting component in DSSCs. The absorption of light produces excitons (electron–hole pairs) in the QD. The electron is injected into the semiconductor oxide support. A hole conductor or electrolyte in the pores of the nanocrystalline oxide film captures the hole. Figure 37.6 shows an example of a QD-sensitized solar cell. The QDs are CdSe, which are attached to ZnO nanowires.

FIGURE 37.5. Dye-sensitized nanocrystalline solar cell.

TABLE 37.3 Solar Cell Efficiencies Under AM1.5 Simulated Solar Illuminationa Type of Cell Crystalline Si cells Single crystal Polycrystalline Thin film Cu(In,Ga)Se2 CdTe Amorphous Si Poly-Si Emerging PV Dye cells (DSSC) Organic cells

Efficiency (%)

24 20 20 16 13 16 11 5

a Solar cells are usually characterized according to the International Electrotechnical Commission (IEC) Norm (ICE-904-3, 1989) under standard test conditions, which correspond to 1 kW/m2 (100 mW/cm2) direct perpendicular irradiance under a global AM1.5 spectrum at 25 C cell temperature.

▪ Enabling electric and hybrid-electric vehicles ▪ Load-leveling (storing energy during periods of low usage and releasing it during peak demand) ▪ Backup supply to avoid interuptions in delivery from the grid ▪ Commercialization of renewable resources, such as wind and solar, which are intermittent

In transportation applications, in particular automobiles, and in widepsread implementation of renewable energy, low cost and long life are essential for commercial success. Developing techniques to store electrical energy efficiently Neither can be obtained is critical for a number of with the current battery applications. technology. In this section, ECONOMIC IMPACT we look specifically at Increasing the lifetime Power interruptions cost US industry about $80 billion lithium-ion batteries (Li-B) of mobile devices such per year. The problem could be solved with better because at the present time as laptops and cell batteries. this is the most widely phones

37.6 CERAMICS IN BATTERIES

▪

37.6 C E R A M I C S

IN

B A T T E R I E S .................................................................................................................................................................

703

Quantum dots (e.g., CdSe) attached to nanowires (e.g., ZnO) Liquid electrolyte or hole conductor

Pt/F:SnO2/glass

Load

FIGURE 37.7. Ragone plot comparing the power and energy densities for batteries, capacitors, and fuel cells. (Energy is the capacity to do work; power is the rate at which work is done).

F:SnO2/glass Light FIGURE 37.6. Quantum dot (QD)-sensitized solar cell. An array of ZnO nanowires—grown vertically from an F-doped SnO2/glass substrate and decorated with CdSe quantum dots—serve as the photoanode. A second F-doped SnO2/glass substrate, coated with a 10-nm layer of Pt, is the photocathode. The space between the two electrodes is filled with a liquid electrolyte, and the cell is illuminated from the bottom.

TABLE 37.4 Energy Output of Electrochemical Devices and Petroleum Energy Source

Energy Output (kWh/kg)

Ultracapacitors Lithium-ion battery Hydrogen fuel cell Gasoline

0.01 0.8 1.1 6.0a

a

Assuming 30% combustion efficiency.

researched battery technology and relies on ceramic materials. Table 37.4 compares the energy-storage capabilities of several different technologies with that of gasoline. Another way to compare the performance of energy-storage technologies is to use a Ragone plot, shown in Figure 37.7: values of energy density (Wh/kg) are plotted versus power density (W/kg). (Energy is in Wh and power in W). The Ragone plot was first used in 1968 to compare batteries but is now used to compare any energy-storage device. Note that you also see power density as W/m3 and energy density as Wh/m3. Franklin’s battery was simply a “battery of capacitors” using glass as the dielectric. Conventional capacitors store low amounts of energy, but that energy can be delivered very quickly; that is, they have a high power density. Fuel cells can have very high-energy densities, particularly if they are operated on liquid fuels such as methanol, but their power density is poor. In between these extremes

704

TABLE 37.5 Examples of Electrochemical Systems Used in Batteries Negative electrode

Electrolyte

Positive electrode

Comment

Zn Li Li Cd Zn Pb Zn

NH4Cl/ZnCl Organic Organic Alkali Alkali H2SO4 Alkali

MnO2 MnO2 CuO NiO MgO PbO2 Ag2O

Zinc-carbon battery Lithium battery Lithium battery Nickel-cadmium battery (NiCd) Mercury battery Lead-acid battery Silver-oxide battery

you can see there are batteries and ultracapacitors. We introduced capacitors in Section 31.7; in Section 37.8 we look at ultracapacitors. Batteries are generally electrochemical devices; many rely on ceramic oxides, particularly as the positive electrode. The oxides of Mn, Ni, Hg, Ag, and Cu are all currently used. Note that we say “the oxides of” to avoid being specific about the charge state of the cation (usually MnO2, NiO, and Ag2O but also NiOOH). Incidentally, the reason batteries are interchangeable (so all AA batteries are the same size and voltage) is due in part to the International Electrotechnical Commission (IEC) and the American National Standards Institute (ANSI). Table 37.5 summaries a few of the electrochemical systems used in batteries today.

37.7 LITHIUM-ION BATTERIES In this section, the focus is lithium-ion batteries, which is the main power source for most rechargable electronic devices, such as laptops, iPads, and cell phones. An Li-B is so-called because lithium ions (Li+) move between the anode and cathode.

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

FIGURE 37.8. LiC2 cells. (A) Unit cell. (B) 2
2 Supercell.

In most current Li-Bs, the anode is graphitic carbon with a very high surface area. Carbon reacts with lithium at room temperature to form a compound with the stoichiometry LiC6. This is an example of an intercalation compound. The lithium ions occupy regions between the layers of carbon atoms, as shown in Figure 37.8. The hexagonal unit cell (a ¼ 0.4316 nm; c ¼ 0.3703 nm) of LiC6 has alternating layers of Li and C atoms, as shown in Figure 37.8A. We can also represent the structure in terms of the 2
2 supercell (Figure 37.8B). Although carbon is widely used, it suffers from the following limitations.

▪ If the intercalation rate of Li into C is too low, the ▪

lithium ions can plate out as metallic lithium, which could cause a fire. The energy density of carbon-lithium anodes is low: 372 mA.h.g1 compared to >3,000 mA.h.g1 for silicon.

FIGURE 37.9. Transmission electron microscopy (TEM) images of mesoporous MoO2 at different magnifications and a typical selected area electron diffraction (SAED) pattern.

▪ The material must contain a readily reducible/oxidizable ion (e.g., a transition metal). ▪ The material must react with Li reversibly and without major structural change. ▪ The material must react with Li with a high DG for high voltage. The ▪ material should react with Li very rapidly on both insertion and removal to give high power. ▪ The material should ideally be a good electronic conductor to alleviate the need for a conductive additive. ▪ The material must be economical to produce, environ-

Two interesting nanostructured ceramics that have been studied as anodes are mesoporous MoO2 (shown in Figure 37.9) and SnO2 nanowires. The mesoporous MoO2 was prepared using a silica template, which was subsequently removed using a dilute hydrofluoric acid (HF) solution. Initial results have shown a reversible lithium storage mentally benign, and abundant. capacity as high as 750 mA.h.g1. Another ceramic material that has been proposed as an alternative to carbon-based All commercial cathode materials belong to one of two anodes is SnO2. Tin oxide has a lithium storage capacity of classes. 781 mA/hg but has been limited by the large volume change (up to 300%) that accompanies the insertion and removal of Cubic oxides, such as LiCoO2 (layered rock salt) and lithium. By forming the film as nanostructured columnar LiMn2O4 (spinel) grains, the local volume changes; and associated mechaniPhospho-olivines, such as LiFePO4 (olivine structure cal stresses can be reduced and the cyclability of the matewas shown in Figure 7.6) rial improved. Thus, we use carbon because it works and is cheap, but we need to develop better materials and be able The phospho-olivines are particularly interesting to make those materials in an economical way. because of the stability of the phosphate group (PO4). On the cathode side of an Li-B, there is much to interest Only under extreme conditions is the olivine structure the ceramist and an opportunity to develop a new material destroyed, and even then that can increase both the the phosphate group is storage capacity and battery maintained as lithium OLIVINE lifetime. The requirements phosphate. Because this Forsterite is Mg2SiO4. Replace the Mg with Li and Fe for the cathode of an Li-B material is an electrical and replace Si with P and you have LiFePO4 an imporhave been summarized by insulator, it has to be tant cathode for lithium-ion batteries. Whittingham (2008). mixed with a conductor,

▪ ▪

37.7 L I T H I U M - I O N B A T T E R I E S ................................................................................................................................................................

705

FIGURE 37.10. Basic structure of an ultracapacitor.

An individual ultracapacitor usually carbon, when it is ELECTRICAL DOUBLE LAYER (EDL) cell is shown in Figure formed into the cathode. The first electrical layer is due to the charging of the 37.10. When voltage is To improve the energyelectrode surface by the applied potential. The second applied across the plates, storage capacity, several electrical layer is due to ions in the electrolyte being the positive electrode groups are looking at catattracted to the surface charge on the electrode. We attracts the negative ions in ion substitutions (e.g., Mn therefore have a “double” layer on each electrode. the electrolye and vice for Fe to form LiMnPO4). versa. A dielectric separator There are still lots of (paper works well) that is placed between the electrodes opportunities to improve the performance of Li-B through prevents the ions from moving between the two electrodes. the use of different materials systems. Also, even with the The charge is stored on the surface, which is why high surface existing materials, there is a need to be able to develop lowareas are important. The cell voltages are limited to ~3 V to cost, high-volume synthesis methods. This need is particuprevent decompositon of the liquid. As the amount of energy larly true for nanomaterials, such as the nanophosphates that can be stored is very directly related to surface area, the being used by A123 Systems, a company that came out of goal is to find lightweight materials with high surface area. the research done at MIT in the United States. Table 37.6 lists surface areas of a number of commercial and experimental carbons, and you can see the close correlation with specific capacitance (up to a point!). Several groups 37.8 ULTRACAPACITORS have proposed replacing activated carbon—which is From the Ragone plot in ULTRACAPACITORS ON PLANES currently the material of Figure 37.7, you can see Each Airbus A380 uses 16 ultracapacitors in the emerchoice—with nanomaterials that ultracapacitors (also gency doors. such as carbon nanotubes called supercapacitors and and nanoporous carbons. electrical double-layer The biggest problem capacitors, EDLCs) have associated with this at the moment is the high cost of making very high power densities and energy-storage capacities— the nanomaterials and the need to attach them to the alumiexceeding those of conventional capacitors. Whereas Li-B num current collector with a suitable binder, which inherently manufacturers (and university researchers) are putting a lot reduces the accessible surface area. of effort into increasing the power density of Li-Bs, ultracapacitor manufacturers are looking at increasing energy density. Also, because capacitors store charge on the surface, rather than through a chemical reaction, they can be cycled hundreds of thousands of times (i.e., they 37.9 PRODUCING AND STORING don’t have to be thrown away as frequently as batteries!) HYDROGEN In Section 31.7, we described the structure and properties of conventional capacitors that used ceramics 37.9.1 Thermochemical Processing as the dielectric layer. In ultracapacitors, the structure is quite different: high-surface-area carbon is used as the We know it is completely feasible to produce hydrogen electrodes (aluminum foil is used as a current carrier), using solar energy from the point of view of thermodyand the electrolye is a liquid, typically acetonitrile. namics. The idea is to use thermochemical processing. The

706

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

TABLE 37.6 Specific Capacitance and Specific Surface Areas for Different Carbons Carbon

SDFT SBET Cg (F/g) (m2/g)a (m2/g)b Comments

Vulcan 12.6 240 CX72 MM192 66.5 1,875

131.6

Made by Cabot Corp.

910.2

Black 70.5 1,405 Pearl 2000 Picatif 81 2,048

831.8

Experimental carbon black from Timcal Made by Cabot Corp.

W30

41.5 790.4

490.8

W37 W54.5 K1500

53.1 913 72.4 950 93.8 1,429

660.5 879.1 1,219

K2200

93.6 2,286

1,613

K2600

87.8 2,592

1,744

a b

1,172

Vegetable-based activated carbon from PICA SAS (France) W ¼ wood-based activated carbons; 30% weight loss during activation 37% weight loss during activation 54.5% weight loss during activation Activated novoloid fibers from American Kynol Inc. The numbers refer to the nominal BET specific surface area Activated novoloid fibers from American Kynol Inc. Activated novoloid fibers from American Kynol Inc.

Surface area measured by Brunauer, Emmett, and Teller (BET method. Surface area determined using the differential functional theory (DFT) model.

goal is to convert water (in the form of steam) into its components, hydrogen and oxygen, and then collect the hydrogen for use as a fuel. Simple! The process is particularly exciting because efficiencies of up to 76% should be achievable, which makes it both commercially and environmentally very attractive. The principle behind the process is to use solar energy to drive a highly endothermic reaction and then produce hydrogen by a subsequent exothermic reaction. This is known as a two-step water-splitting cycle. Solar energy in the form of heat is used to reduce the ceramic (a metal oxide). This reduced metal oxide is then reoxidized by removing the oxygen from water, thus producing hydrogen gas. The reduced (lower valence) metal oxide in step I (thermal reduction) could alternatively be a metal carbide, a metal nitride, or even a metal. The second step (oxidation by water) is known as hydrolysis. The limit for step I would be: Step I: metal oxide ! reduced metal oxide + O2 Step II: reduced metal oxide + H2O ! metal oxide + 2H2 Mx Oy ! xM þ (y=2ÞO2 If we use FeO in step I, then the temperature must be >1,600 K. However, step II is then spontaneous at a temperature of ~1,200 K.

37.9 P R O D U C I N G

AND

FIGURE 37.11. TEM image of a silica nanospring for hydrogen storage (on a lacy C film).

Several systems are being explored for this purpose. They include the reduction of ZnO (in some cases enhanced by the presence of C), the reduction of ceria, and the reduction of oxides containing Fe3+ (both iron oxide and a range of ferrites). All that we need to do is to optimize the ceramics and, perhaps the more challenging step, optimize the geometry of the sample to allow repeated cycling (1) without degradation and (2) while allowing easy extraction of the hydrogen. Hence, it’s the usual process: improve the design (tailor) of both the material (e.g., powder versus multilayer films or foams) and the reactor itself. This whole process should remind you of our discussion of phase boundaries (PBs) in Chapter 15 and reactions in Chapter 25. The solar part of this has many similarities to processes that are now being explored for producing lime (the endothermic calcinations of CaCO3) and other ceramics. Currently, we use fossil fuels for producing cement and lime, which accounts for 5% and 1%, respectively, of the global human-made CO2 emissions—up to 40% of this is from burning fossil fuels.

37.9.2 Hydrogen Storage Of all the limitations preventing the achievement of a hydrogen economy, the most significant is hydrogen storage. For transportation applications, storage requirements are particularly stringent and none of the current approaches comes close to meeting targets. Indeed, some of the approaches are actually dangerous. Storing hydrogen on the surface of nanomaterials is an exciting possibility. The idea is again to use the very large surface areas available at the nanoscale. The hydrogen attaches nondissociatively (i.e., as H2) through weak molecular-surface interactions, such as van der Waals forces. Studies have shown that hydrogen attaches to the surface of carbon nanotubes, but the temperatures don’t seem to be ideal for transportation needs. Recently, using glasses for hydrogen storage has been proposed; and, experimentally, hydrogen has been shown to attach to the surface of one-dimensional silica glass nanostructures (e.g., wires, ropes, tubes, springs) at room temperature. Figure 37.11

S T O R I N G H Y D R O G E N .........................................................................................................................................

707

shows a transmission electron microscopy (TEM) image of a silica nanospring. The only difference, apart from size, between these nanostructures and “bulk” glass fibers appears to be that the surface is more ionic, which may be important for hydrogen attachment. The nanosprings grow by the vapor–liquid–solid (VLS) mechanism described in Chapter 29. The Au catalyst is the dark particle. The deposition process occurs at temperatures as low as 300 C, allowing them to be formed on polymer substrates.

37.10 ENERGY HARVESTING

for laptops and the personal entertainment systems used on long-haul flights. Of course, lightweight materials would have to be used in this application. An advantage of piezeoelectric generators is that they are scalable: to generate more power, individual piezoelectric elements can be combined; to create small localized sources of power, the device dimensions can be scaled down even to the nanoscale so long as the aspect ratio is maintained. ZnO nanostructures such as nanowires and nanobelts (shown in Figure 37.13) are creating lots of interest in nano-piezoelectrics in terms of scavenging energy from the environment, such as wind, friction, and body movement. However, this is a technology that is still in the very early stages of development, and significant engineering challenges need to be overcome before there are commercial products based on these exciting nanomaterials.

Kinetic energy (the energy due to motion) can be harvested using piezoelectric-based generators. Piezoelectric ceramics such as lead zirconate titanate (PZT) have been used to make simple cantilever generators, such as the one illustrated in Figure 37.12. A cantilever structure has a PZT film on the top and bottom surfaces of a metal plate. 37.11 CATALYSTS In Figure 37.12, the metal AND CATALYST is a strip of brass. The canSUPPORTS ENERGY HARVESTING tilever deflects due to Mechanical energy is converted into electrical energy. external vibrations (e.g., The final section in this shaking), which strains the chapter looks at ceramics PZT films, generating a as catalysts and catalyst supports for emerging energy charge. The mass at the end of the beam lowers the resoapplications. We briefly introduced this possibility in nant frequency. Section 13.17. Here, we show some specific examples. There is a lot of interest in using human-powered To maximize reaction rates, it is necessary to have very energy-harvesting devices for personal generation of high surface areas (so we are dealing with nanomaterials). energy. We might generate our own electrical energy to Surface areas are typically hundreds of square meters per power our cell phones and laptops, for example, by simply gram (a few grams would have the equivalent surface area walking or swinging our arms. Studies have shown that an of several football fields). Table 37.7 lists the dominant average-gait-walking human weighing 68 kg produces catalytic processes for energy conversion that have been 67 W of energy at the heel of the shoe. Although it would not be possible to “harvest” all this energy, it should be possible to generate as much as 1.3 W from walking. A group at MIT in the United States put a PZT piezoelectric dimorph into the heel of a Navy work boot that generated energy each time the heel struck the ground. Each event generated an average of 8.4 mW. The maximum power generated from a PZT cantilever generator is about 0.4 mW. Boeing is developing seats for its commercial airplanes that would harvest the motion of passengers moving in their seats. This energy could be used to power outlets

FIGURE 37.12. Cantilever piezoelectric generator.

708

FIGURE 37.13. TEM image of ZnO nanobelts.

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

TABLE 37.7 Catalysts, Applications, and Reactor Types Reaction

Reactants

Major Products

Cracking

Petroleum fractions (e.g., straight-chain and cyclic alkanes)

Naphtha reforming

Petroleum-derived naphtha (e. g., straight-chain and cyclic alkanes) Sulfur-containing compounds in fossil feedstocks (e.g., thiophene, dibenzothiphene)

Lighter hydrocarbons (alkanes and Acidic zeolite in matrix of silicaEntrained alkenes) alumina with components to bed oxidize CO (Pt) and sequester sulfur oxides Aromatic compounds, isomerized Pt nanoclusters with another metal Fixed bed or (branched) alkanes (Re, Sn, or Ir) supported on Al2O3 moving bed H2S and hydrocarbons (e.g., MoS2 nanoclusters promoted with Fixed bed butenes, biphenyl) Co and supported on Al2O3 (typically trickle bed) or slurry NH3 and hydrocarbons MoS2 nanoclusters promoted with Fixed bed Ni and supported on Al2O3 (typically trickle bed) or slurry H2O and hydrocarbons (e.g., MoS2 nanoclusters promoted with Fixed bed or benzene) Co or Ni and supported on Al2O3 slurry

Hydrodesulfurization (HDS)

Hydrodenitrogenation (HDN)

Nitrogen-containing compounds in fossil feedstocks (e.g., quinoline, triglycerides)

Hydrodeoxygenation (HDO)

Oxygen-containing compounds in coal or biomass-derived feedstocks (e.g., phenol, sugars) CO2 + H2 CO + H2O

Water-gas shift Fischer-Tropsch synthesis

CO + H2

Methanol synthesis

CO + H2

Straight-chain alkanes, alkenes, oxygen-containing compounds (e.g., alcohols, aldehydes, ketones) Methanol (with CO2 and H2O)

developed for fossil fuels, such as petroleum, natural gas, and coal. Many catalysts are composite materials consisting of a carrier or support—which may constitute as much as 99% of the total mass but is catalytically inactive—combined with catalytically active components and possibly promotors. Typical ceramic supports include aluminas (e.g., g-Al2O3, Z-Al2O3), silica gel, zirconia, magnesia, zeolites, and carbon. Aluminas are widely used because they are inexpensive, thermally stable, and durable. As an example, commercial ultra-pure g-Al2O3 (99.99% pure) with a surface area of 70–100 m2/g is available in bulk quantities for 1,000 times in its first 4 years. Generated lots of ideas and some controversy Whittingham MS (2008) Materials challenges facing electrical energy storage. MRS Bull 33:411, A review of current battery technology by one of the pioneers in the field

C H A P T E R S U M M A R Y .................................................................................................................................................................................

711

SPECIFIC REFERENCES

Farnan I, Cho H, Weber WJ (2007) Quantification of actinide a-radiation damage in minerals and ceramics. Nature 445:190, Used NMR to show that a-particles could amorphize zircon Fletcher EA (2001) Solarthermal processing: a review. J Solar Energy Eng 123:63 Fletcher EA, Moen RL (1977) Hydrogen and oxygen from water. Science 197:1050 Gr€atzel M (2005) Solar energy conversion by dye-sensitized photovoltaic cells. Inorg Chem 44:6841, This is not the original paper describing DSSC, which was published in Nature in 1991. This paper is a more recent review Leschkies KS, Divaker R, Basu J, Enache-Pommer E, Boercker JE, Carter CB, Kortshagen UR, Norris DJ, Aydil ES (2007) Photosensitization of ZnO nanowires with CdSe quantum dots for photovoltaic devices. Nano Lett 7:1793, Uses quantum dots instead of a dye to absorb solar energy Marin-Flores O, Turba T, Ellefson C, Wang K, Breit J, Ahn J, Norton MG, Ha S (2010) Nanoparticle molybdenum dioxide: a highly active catalyst for partial oxidation of aviation fuels. Appl Catal B-Environ 98:186, Demonstrates the use of MoO2 for the partial oxidation of aviation fuels Qin Y, Wang X, Wang ZL (2008) Microfiber-nanowire hybrid structure for energy scavenging. Nature 451:809, Growth of ZnO nanowires on Kevlar fibers. Low cost methods of synthesizing nanomaterials are critical to their use in industry. Integrating ceramic nanostructures with polymer fibers is essential for many large-area applications and wearable devices Shi Y, Guo B, Corr SA, Shi Q, Hu YS, Heier KR, Chen L, Seshadri R, Stucky GD (2009) Ordered mesoporous metallic MoO2 materials with highly reversible lithium storage capacity. Nano Lett 9:4215 Smith BH, Gross MD (2011) A highly conductive oxide anode for solid oxide fuel cells. Electrochem SolidState Lett 14:B1, SrMoO3, a conducting ceramic, was investigated as an anode material for SOFC. For fuel reforming a catalyst would need to be added, the authors used Pd Yang L, Wang S, Blinn K, Liu M, Liu Z, Cheng Z, Liu M (2009) Enhanced sulfur and coking tolerance of a mixed ion conductor for SOFCs: BaZr0.1Ce0.7Y0.2xO3d. Science 326:126, The ceramic is a mixed ion conductor that was used for fuel reforming

WWW http://dodfuelcell.cecer.army.mil/. The ERDC-CERL fuel cell website www.dyesol.com/. Dyesol is located in Queanbeyan, New South Wales. It was formed in 2004 with the goal of commercializing dye-sensitized solar cells (DSSCs) www.solaronix.com/. Solaronix is located in Aubonne, Switzerland and supplies components for DSSCs www.spectrolab.com/. Spectrolab manufactures very high efficiency solar cells. These are for space applications, not for terrestrial power generation www.a123systems.com/. A123 Systems manufactures Li-ion batteries. The company was founded in 2001 using nanophosphate technology developed at Massachusetts Institute of Technology. The name “A123” comes from the Hamaker constant used to calculate the attractive and repulsive forces between particles at nanodimensions. We introduced the Hamaker constant in Section 4.7. In October 2012, A123 filed for bankruptcy

712

............................................................................................................................................ 37

ENERGY PRODUCTION

AND

STORAGE

38 Industry and the Environment CHAPTER PREVIEW Throughout this book we have highlighted the engineering applications for ceramics. In the final analysis, the importance of any material is based on what it can be used for. For example, at the present time, high-temperature superconductors (HTSCs) are of research interest but are not commercially important. Because of the unparalleled range of properties shown by ceramics, they find application in a vast number of areas. This last chapter looks at the field from an industrial perspective. Because it is impossible in one chapter to cover every aspect of the multibillion-dollar ceramic industry, we have chosen to focus on a few topics, mainly through examining case studies. One of the exciting prospects for the industry over the next decade is in nanotechnology. Ceramic nanopowders already represent the biggest segment of the nanopowders market and are used for polishing, sunscreens, etc. With the demonstration of the successful growth of ceramic nanowires, nanoribbons, nanosprings, and nanotubes, etc., there exists the potential for even more applications in critical areas such as solar cells, batteries, and hydrogen storage. As we have often done, we begin with some history.

38.1 BEGINNING OF THE MODERN CERAMICS INDUSTRY

was (and still is) clay. The proximity of raw materials provided an economic advantage over other rural potteries that were still using the diminishing supply of timber. Staffordshire is a long way from the major metropolitan In Chapter 2, we described some of the early history of areas of London, Bristol, and Norwich. Early pictures ceramics and their production. The transition to a largeof Tunstall, one of the six towns that formed the Potteries scale manufacturing industry occurred in Western Europe and in 1910 became during the eighteenth cenabsorbed into the city of tury as part of the period Stoke-on-Trent, show a that became known as the SE`VRES town surrounded by hilly Industrial Revolution. The Royal Commission: The factory at Se`vres was countryside. great porcelain factories commissioned to make an 800-piece dinner service for By the mid-eighteenth established and subsidized Catherine II of Russia. It took 3 years to complete. century there were many by royal patronage at separate potteries employing Miessen in Germany and a large number of workers. A petition presented before the Se`vres in France began to give way to purely commercial British Parliament in 1763 read: products being made in Staffordshire in the north of England. Later, the factories at Miessen and Se`vres began to imitate In Burslem [another of the six towns that made up the English designs. They were certainly helped in this area by Potteries] and its neighborhoods [sic] are nearly 150 separate immigrant workers. Emigration was a concern for the potteries for making various kinds of stone and earthenware, which, together, find constant employment and support for ceramics industry more than many others, such as iron pronearly 7,000 people. duction, because it relied on secret processes, such as specific body and glaze compositions. Once these became known, a In the early days of the pottery industry in England, worker would become valuable to a competitor. transport of raw materials in, and products out, was ineffiThe development of the Staffordshire area as the promcient. The costs of transportation had to be included in the inent pottery center in selling price of every artiEngland was in large part cle produced. Clearly, SIX TOWNS: THE POTTERIES due to the use of coal as a quantity production could Tunstall, Burslem, Hanley, Stoke-upon-Trent, Fenton, fuel for the kilns. Coal was not be achieved without Longton abundant in this area, as better transportation. C.B. Carter and M.G. Norton, Ceramic Materials: Science and Engineering, DOI 10.1007/978-1-4614-3523-5_38, # Springer Science+Business Media New York 2013

Master potter and entrepreneur Josiah Wedgwood was instrumental in organizing a potters’ association to push for the development of improved roads and a canal system. Wedgwood realized that cheaper and more regular transport meant an even flow of production, less breakage, lower prices, wider markets, and greater sales. Staffordshire potters lobbied successfully for the development of a canal that would link the rivers Trent and Mersey, which was authorized by an Act of Parliament in May 1766. The project was completed in 1772 at a total cost of £300,000. The completion of the Trent-Mersey Canal ensured that Staffordshire would remain the center of English pottery production. A complex web of railway routes followed, and these developments transformed an isolated rural area into a major industrial center. Wedgwood made contributions in several areas that helped transform the production of pottery into a major industry. He changed the manufacturing process and adopted mechanization that would enable him to increase production while lowering prices; and the increased productivity would help to maintain a stable wage for his employees. He had many ideas about sales and marketing of his products and was the first manufacturer to introduce the “satisfaction-guaranteed-or-your-money-back” policy, which is now an extensively used tool for selling. Wedgwood was an advocate of free trade, and a commercial treaty with France was welcomed by many of the ceramic manufacturers as a means of stimulating imports. Industries that had not adapted to new technology (e.g., the use of steam) feared the competition of imports. Wedgwood wrote on this issue of the treaty with France: An exchange of the produce of one nation for the manufactures of another are happy circumstances, and bid fair to make the intercourse lasting; but sensible as I am to the interests of trade, manufacturers and commerce, they all give place to a consideration much superior in my mind to them all. I mean the probability that a friendly intercourse with so near and valuable a neighbour [sic], may keep us in peace with her—may help to do away with prejudices as foolish as they are deeply rooted, and may totally eradicate that most sottish and wicked idea of our being natural enemies . . ..

The production of ceramics became an important and growing export industry. Vast quantities of ceramic ware produced in the potteries were exported from the major seaports of London, Bristol, Liverpool, and Hull to America, the West Indies, and all over Europe. Today, many of the most famous names associated with the Staffordshire Potteries, such as Royal Doulton and Spode, can still be seen. A visit to any department store demonstrates that these names are still associated with some of the highest-quality ceramic tableware. However, most of the Royal Doulton pieces are now made in China, not in the small towns in the Potteries. As with other industries, the ceramic industry has seen much consolidation and acquisition in recent years. Wedgwood merged in 1986 with Waterford Crystal, forming Waterford Wedgwood plc to become the world’s largest tableware

714

company, with sales in excess of $1 billion. Unfortunately, the company bearing the great name of Wedgwood went into receivership in 2009, and production at its headquarters in Waterford, Ireland shut down with the loss of several hundred jobs. Also in 2009, the Portmeirion Group acquired Spode; and production still takes place in Stoke-on-Trent.

38.2 GROWTH AND GLOBALIZATION Although the United Kingdom was a traditional leader in the development of ceramics, there were major changes during the latter half of the twentieth century, when Japan became the major producer of ceramics. Rapid transportation routes meant that manufacturing sites no longer needed to be near mineral resources. For example, Japan has no significant domestic energy supplies but is a major industrialized manufacturing nation. One of the significant changes that led to the growth and dominance of Japan was a shift in its business from traditional low value-added basic ceramics to one that has a large component of high value-added. Table 38.1 shows the market for hightechnology ceramics as it was in 1980. Japanese companies satisfied about half of the $4.25 billion demand. In some areas they were dominant, producing over 60% of the worldwide market for integrated circuit (IC) packages and almost 80% of the ferrites. The market for IC packages, which is based on alumina, was established largely by U.S. companies, but there are few remaining that sell on the open market. The rapid growth in the Japanese production of ferrites in the 1970s and 1980s coincided with a decline in this area in the United States and in Europe. The only serious constraint on the expanded production of ferrites in Japan during this period was a shortage in raw material (secondary iron oxide) caused by weak steel production. In 2011, Japan remains the largest market for advanced ceramics. TABLE 38.1 1980 Market for High-Technology Ceramics (Excluding Fibers, Nuclear Fuels, Spark Plugs) Product

Japan

World

Ceramic powders Electronic IC packages/substrates Capacitors Piezoelectrics Thermistor/varistors Ferrites Gas/humidity sensors Translucent ceramics Cutting tools: carbide, cermet, coated noncarbide Structural ceramics (heat and wear resistant) Totals

$130 540 325 295 125 380 5 20 120 120 $2,065

$250 880 750 325 200 480 45 45 1,000 250 $4,250

IC integrated circuit. Markets in millions of dollars.

............................................................................................................................................... 38

INDUSTRY

AND THE

ENVIRONMENT

TABLE 38.2 Challenges Facing the Ceramic Industry According to Percent of Survey Respondants Environmental standards Changing markets Cost of labor Imports Health and safety standards Cost of materials Quality of labor Capital for expansion Quality control Cost of fuel

TABLE 38.3 Types of Ceramic Industry 39% 33% 32% 27% 26% 25% 20% 20% 19% 19%

Activity

Examples

Ceramic powders

SiC for abrasives Nanosized TiO2 for sunscreen Bioactive glasses for bone reconstruction Bayer process Al2O3 for the production of Al using Hall-He´roult cells

Forming powders into Slip casting of toilet bowls bulk forms Cz growth of Nd:YAG single crystals AlN sheets by tape casting Glass melting Fabricating ceramic components

Ceramic chip capacitors Packages for integrated circuits SiC pressure sensors

Table 38.2 shows some of the challenges that face ceramics companies worldwide. This information was gathered from a survey of over 250 ceramics companies. The major challenges are meeting environmental 38.4 CASE STUDIES standards, adapting to changing markets, and labor costs. The ceramics industry, like many others, can establish As we described in Chapter 1, the ceramics industry covers production facilities where labor costs are lower. For a wide range of materials and products. We can generally example, KEMET Corpodivide the activities of this ration, based in Greenville, industry into three distinct SC, a manufacturer of tancategories, as listed with CERAMIC IC PACKAGES talum electrolytic and mulexamples in Table 38.3. In Together with substrates, ceramic packages compete tilayer ceramic chip this section, we describe in with polymers but are superior in terms of thermal capacitors, has relocated more detail one example of conductivity and hermeticity and are used in highall manufacturing to each activity and some curreliability applications. lower-cost facilities such rent industrial trends. as in Mexico and China.

38.4.1 Silicon Nitride Powder

38.3 TYPES OF MARKET As we described in Chapter 1, the ceramics industry is generally divided into six distinct markets.

▪ ▪ ▪ ▪ ▪ ▪

Glass Advanced ceramics Whiteware Porcelain enamel Refractories Structural clay

It is in advanced ceramics that many of the exciting developments are occurring. The 2007, U.S. advanced ceramics market was $12 billion, and by 2015 it is projected to reach $16 billion. The global market for advanced ceramics is expected to exceed $56 billion by 2015. The largest growth segments are electronic ceramics, which includes capacitors, piezoelectrics, and ferrites. In chemical processing and environmental-related applications, ceramics are routinely used for automotive catalyst supports and filters that reduce pollutants in response to regulations on both automobile and industrial emissions.

Silicon nitride, Si3N4, is not a naturally occurring mineral. All the Si3N4 that we use must be synthesized, usually by one of the following methods (more details in Chapter 19).

▪ Direct nitridation of Si ▪ Carbothermal reduction of silica in N ▪ Vapor phase reaction of SiCl or silane (SiH ) with 2

4

4

ammonia

The characteristics of the resulting powder that are important to end-users are:

▪ Particle

▪ ▪

size and distribution. Powder compacts containing a few coarse particles produce components with significantly reduced strength and toughness (two of the properties we are often trying to maximize). Milling can be used to reduce particle size but often leads to significantly increased costs and the introduction of unwanted contamination. Surface area. Affects how easily the powder can be densified during sintering and the final grain size in the sintered component. Purity. Depends on the processing route and wide variations are possible. Oxygen on the surface of the

38.4 C A S E S T U D I E S ....................................................................................................................................................................................

715

FIGURE 38.1. Schematic showing processing steps in forming Si3N4 by LPS. The metal oxide additive (m) would be something like Y2O3 and the liquid an oxynitride/silicate.

▪

powders can affect densification; however, we need enough to form the liquid phase during sintering. Structure. A high a-Si3N4 content is desirable because it favors the conversion to rodlike interlocking b-Si3N4 during subsequent processing into bulk components, as illustrated in Figure 38.1.

The cost of Si3N4 powders can vary from $30/kg up to $150/kg depending on particle size and purity. The high costs of raw material and the subsequent shaping and forming processes have restricted the use of Si3N4. Table 38.4 shows a summary of a cost analysis performed for direct-nitrided Si3N4 powder. Most of the cost of the powder is due to the raw materials and the process materials (i.e., the milling media). Si3N4 milling media is very expensive; it costs about $150/kg, compared with alumina or steel media at $16/kg and $4/kg, respectively. Some of the present applications for Si3N4 parts include cutting tool inserts, bearings and rollers, refractory parts, cam followers in engines blades, vanes in heat engines, and turbocharger rotors. The advantage of using Si3N4 for cutting tool inserts should be clear from Figure 38.2. You may see the units given as surface feet per minute (SFM), which is a measure of the distance covered by a rotating tool (traditionally a saw or lathe now used in wear); the surface foot is a linear foot (3.28 SF ¼ 1 m). There are several powder manufacturers—primarily in China, Germany, and Japan—producing hundreds of tons of Si3N4. There are currently no U.S. suppliers of Si3N4

716

TABLE 38.4 Summary of Costs for Direct-Nitrided Si3N4 Powder Cost distribution By cost element Silicon powder Silicon nitride seed powder Capital equipment Direct labor Energy Process materials Total By process step Silicon powder Silicon nitride seed powder Direct nitriding Crushing Fine grinding Total

$/kg

% of total

7.49 1.71 0.49 1.18 1.88 16.74 29.48

25.4 5.8 1.7 4.0 6.4 56.8 100.0

7.49 1.71 3.30 6.62 8.36 29.48

25.4 5.8 11.2 29.3 28.4 100.0

powder. GTE, Dow Chemical, and Ford Motor Company developed high-quality Si3N4 powders between about 1973 and 1995, but none of these companies is a supplier today.

38.4.2 Ceramic Chip Capacitors We described the structure of a multilayer chip capacitor (MLCC) in Chapter 31. The MLCCs are used in a large number of products, in particular personal computers and

............................................................................................................................................... 38

INDUSTRY

AND THE

ENVIRONMENT

FIGURE 38.2. Improvements in the rate of metal cutting for various cutting-tool inserts.

FIGURE 38.4. Trends in number (N) of dielectric layers and thickness (D) of the dielectric layer.

1980s, most MLCCs produced were either 1206 or 0805 (the two largest sizes). By 2000, the 1206 type accounted for less than 10% of the market, and 30% of the market was the 0402 type: a component with a fraction of the area and using much less material. Since 2000, the very small 0201 captured an increasingly larger market share. The use of lower operating voltages in handheld devices and microprocessors has allowed dielectric layer thickness (D) to be reduced; consequently, higher layer counts (N) are possible within the same overall device dimensions, as shown in Figure 38.4. The trends in N and D predicted by the arrows at the end of the lines in FIGURE 38.3. Multilayer ceramic capacitor-size trends. Figure 38.4 are consistent with where the industry cell phones. A typical cell was in 2011: N > 1,800 EIA CAPACITOR CODE phone may contain 400 and D < 1 mm. The size of MLCCs is defined as “llww”: ll is the length MLCCs. The goal is to You may recall from of the capacitor and ww is the width, both in make smaller components Chapter 31 that capacithousandths of an inch (a case where Imperial and with larger capacitances at tance, C, is given by: U.S. units are still widely used in industry!) a lower cost. Example: 0805 means a capacitor of length 0.080 in Capacitors are extremely A C ¼ e0 k (31.13) (~2 mm) and width 0.050 in (~1.25 mm). price-competitive because d of their relatively simple structure (see Figure 31.19). By reducing d and The following costs are involved. increasing the number of layers (effectively increasing A), it has been possible to expand the capacitance of Ceramic dielectric. The ceramic capacitor industry MLCCs into the tantalum and aluminum electrolytic uses more than 10,000 t of BaTiO3-based dielectrics capacitor range. (about 90% of the total produced). The ability to cast thin layers (