TATE 1197 1 25, Caleulate the ferml energy, of silver A pohly of the metal bs 10-5 g em, metal, given that the atumnle mass of silver bs 1TH pS gem? = 105 710 bem? M,UT-H6H 10? Kean | p 10-5710 bem > yo? Fi? 1632109 mmol! soca one stan ofA contains tne valence electron, therefore, the a electron, therefure, the number of valenae elects per mate of the = |= (o-006 oP [seen 0 mat PO ” W HCVMGT 7 AO bg) | a LB 2 Spo = sn P 28 am, and ‘Theory of Solids, ‘The band theory of solids aims at explaining vast differences in the sprit of metals which the wave mechanteal free electron theory failed to do. This theory was arymuated by Flach (of NMI ane, best known for the Bloch ‘equations in NMR), begins dee ccinadinger wave equation and incorporates the periodic potential enerey Vix) of a lattice of sere pe equation thus obtained, called the fBlach equation, is writen as woidy HAY yin = EW (46) am, det a Y ose the posentlal energy |s pericxlie, Viz) = Vortina) where @ Us tht Tepes distance along the nd. is-an integer. Bloch gave the solution of By. 46 as direction in the Fattic yoo = yx) exp (ihe) AAT) sete fe) is an eigenfunction with the periodicity a of the lati ‘A one-electron wave function of see ty ie called a Bloch function and can be decomposed into 4 butt of travelling waves. dintuactions can be assembled into localized wave packets 0 represen electrons that propagate frely through the periodic potential field of the jon cores, The dependence of the energy on & is duainatie U2, = 1242/2m,), as in the free-electron theory. However, for k = + n/a, discontinuities Tppeat in energy that fead to a band structure (ig. 31). E ie Zz | BZ tw) () ro) Fig. 31. Electron energies in solids (a) Free-clectron theory showing the quadratic dependence of F on kkh is. quad aetton energies in tron), (b) Energy bands showing gaps OE at values of F for which the Bragg :juation holds, The ponitions A an B correspond to waves. yf (cos (0), ‘and yp (sin (rx/a)), respectively + the Yalues of k Cinn/a) define the boundaries of the 1¢-dimensional Brillouin zones, fc) Structure of quasi- Continuous energy hands separated by gaps AE) (i = te 2. +) rec am which ty measured is wenerally called K-space. It is the reciprocal space of X-ray Hogtaphhy ut whereas crystallography is concerned with the (weighted) reciprocal Jattice points. Each energy state weird under investigation. As expected, & has the dimensions of reciprocal length. can be revatded ay a poi Wby kor by 1,1, and n., can accommodate two electrons with spins £1/2 and in k-space. [For a cubic erystal, k = ilk, + ky + ky = Walaa, + my + nyae PRINCIPLES OF PHYSICAL CHEMISTRY ‘The surface of constant energy Ep in k-space is called the Fermi surface ; it separates filled orbitals from unfilled orbitals. For most values of &, the electrons behave very much like free electrons. However, at values of equal to n/a, the condition for Bragg reflection of electron waves in one dimension is realized : k = + nna is equivalent to the Bragg equation 2a sin 0 = 1 2 where k = 2n/d and sin O=1. The first order reflection at k = 4zin/a arises because waves reflected from adjacent atoms interfere constructively, the phase difference being 2x. The region lying between + n/a is called the first Brillouin zone (Fig, 32), The energy is quasi-continuous within a zone because the energy levels are very closely spaced in a solid but discontinuous at the zone boundaries. As k increases towards nx/a, the eigenfunctions (6) contain increasing amounts of Bragg-reflected wave. At k = n/a, for example, the wave exp(inx/a) reflects as exp(cinx/a) and the resulting combinations are standing waves yr and y of the forms cos(rx/a) and sin (n/a), respectively. Brillouin zones are not normally encountered in X-ray crystal structure analysis but they are essential to an analysis of electron energy levels in solids. Second zone Second zone 1 —— First zone — ek “nla -2nla wa 0 wa nla 3xla Fig. 32. Boundaries for the first two Brillouin zones in a one-dimensional lattice of periodicity a. The probability densities of the two standing waves are |y;2| and | y2|?, whereas that for the travelling wave is exp (i2kx). Fig, 33. Illustrates a one-dimensional periodic potential field and the wave probability functions described above, The travelling wave distributes charge uniformly along the x-axis ; yy has its peaks at na andy» has its peaks at (1-+1/2)a. The potential energies of the two distributions follow the order | y4|? Fig, 34, Formation of energy bands in a solid, lic solid where the 1 be grouped into red, as is the case in a me form a series which | compared with tor and a 1s within a band is very ns of a conductor, an insul Conduction on Tomfuction Band te le alence: Band Vatence Valence Band Band ( « Comtuctor (Meta Insola Semiconductor ig. 38. Biergy band diagrams of (a) conductor, (insulator () semiconductor} 1200 PRINCIPLES OF PHYSICAL Giihistiy Conductors. In a conductor (such as a metal), the valence band (V.B.) is full of ele the conduction band (C.B.) is only partly filled. Only a small amount of energy sutfices 4 ie ons to move within the conduction band, some rising to a higher energy level ang. 42% returning to a lower energy level. This movement of electrons within the C.B. constitutes ef others veMluction, In some metallic conductors, the V.B. and the C.B. actually overlap resulting in a wast! filed top band. The diference in energy between the V.B. and the C.B. is called the energy gap en” Insulators. In an insulator, the V.B. is full and the E, is very large. Thus, it will take a pp deal of eneray to make an electron jump the energy gap and to cause the insulator to break deat The break-down does not occur even at very high temperatures or under very large electric fen Rubber is an example of an insulator. elds, Semiconductors. We shall deal with the so-called intrinsic semiconductor which is a semiconduc in tg own right, -e., no impurity has been added to it. The best known intrinsic semiconductors ac silicon and germanium. In an intrinsic semiconductor, the V.B. is full and the C.B. is empty at yery ow temperatures. The Ey between the two bands is, however, so small that electrons can jump across it by the addition of a small amount of thermal ° energy (kg/T) alone, i.e., only heating the material results in electrical conduction. The electrical conductivity increases with increase in temperature since more and more electrons are liberated with increase in temperature. The smaller the value of Ey, the better the semiconductor. Thus, germanium (E,=0-67 eV) is a better conductor than silicon (E,=1'14 eV). As an electron jumps from the V.B. to the C.B., it leaves behind a hole in the V.B. The hole is positively charged and since an electron can jump into the hole from another part of V.B., it is as if the hole was moving ! Conduction can occur either by the negative electrons moving within the C.B. or by positive holes moving within the V.B. A semiconductor at room temperature generally has : emperature-dependence of eesti] much lower conductivity than a metallic conductor because | conductivity of a metal, a semiconductor ata only very few electrons and holes can act as charge superconductor. carriers. The temperature-dependence of electrical conductivity, o, for a metallic conductor, a semiconductor and a superconductor is shown in Fig. 36. ‘A super conductor is one which has zero or very little electrical resistance. As can be seen, a metallic conductor is a substance with electrical conductivity that decreas with increasing temperature, a semiconductor is a substance with an electrical conductivity that increases with increasing temperature and a superconductor is a substance that has zero electrical resistance below a certain temperature called critical temperature. Bd Electrical Conductivities and Enere? Gaps for the Elements of Group T cItOns while e z & 5 2 3 z $ 8 The conductivity of a semiconductor follows the Arrhenius type temperature-dependence with activation energy, Eq, approximately equal to half of the energy gap, E,. Thus, C (diamond) Go exp (- EgkT) = op exp (- Eg/2kT) Si The E, and o values for the elements of Group 1V | Ge are given in Table 8, Sn (gray) ogun sine 1201 ne é semiconductors ; ae ; rine ® ible to increase the charge carriers if atoms with more electrons than the parent element iris Psd by the process known as doping. Extremely low concentrations of the dopant (about pein rons of the Host material) are needed. Pure silicon (Si) and germanium (Ge) ean b ching in a controlled manner by adding dopant impurities whic act as charge carrcr., seater made extremely pute by zone refining. Then some arsenic (As) atoms with five jecirons are added 10 the Ge crystal. An extremely small number of Ge atoms are randomly wesc Ce noms. Only four of the five outer electrons on each As atom are required to form bonds ty Me ih eletron on As atom isnot bonded. AL low temperatures, the fies i he i on the As atom. However, at normal temperatures, some of these fifth electrons on As are sve ie conduction band where they ac a5 charge carriers, This produces what is osm as excel m9 guction. Since the current is carried by excess electrons, tis is kanam as n-type esis, The magoitude of semiconductivity in an extrinsic semiconductor is far greater tan that icon : set rinsie pure semiconductor. in rive donor As atoms are far apart from one another, their electron are localised and the donor sand wil be marow (Fig. 374). The filled dopant band fies near Gust below) the empty conduction and of the lattice. nt one cont ‘Alternatively, a crystal of pure Si or Ge is doped with atoms of an element which has three valence elec per ce at as ssl Condon fim (Ga) or indium (In). Each gallium atom an : a ree outer electrons to form three bonds ae inthe lattice, these electrons cannot form four fonds in the lattice to complete the covalent ( atructure. One bond is incomplete and the site { which remains unoccupied because of the missing Donor Band electron is called 2 positive hole. More for- | mally, the dopant atoms form a very narrow | % emply acceptor band that lies above the full | & silicon valence band (Fig. 37b). At T=0 K, Acceptor Band the acceptor band is empty but at T > 0K, it [omental can accept thermally excited electrons from J the Si valence band. This forms new positive pas syalenee: holes in the Si valence band. The holes appear to move in a direction opposite to the motion (a) @ of electrons. The positive holes ‘hop’ through Fig. 37. The band structure of (a) an n-ype the band. Since the charge carriers are now semiconductor () a p-type semiconductor. the positive holes, this type of conduction is known as p-type conduction. In principle, any of the Group V elements (such as P, As, Sb. Bi) can be used to make n-type semiconductors but, because of its low melting point, As is most commonly sed. Similarly, any of the Group III elements (such as B, Al, Ga and In) can be used to make p-type semiconductors, though indium (In) is most commonly used because of its low melting point. A minute amount of donor impurity can produce a dramatic change in the conductivity of a semicondcutor. For example, 1 part of a donor impurity per 10° parts of germanium increases its conductivity by a factor of 103, : ‘The combination of n and p-type semi-conductors (known as_n=p junction), finds interesting sonlicaions in the manufacture of transistors. This device can conduct electric current more easily one particular direction than in the reverse direction and, therefore, can be used as a rectifier foris an acronym for ‘light amplification by stimutated emis are interested in increasing the Probability of mulated ey ! ed Ct ¢ iy ated emi sy, pan): The Way t0 do it When Byyy is small is to somehow in Boatiion of the excited state over that of the ground stare ion of radiation’, Thus, i ion which is given by © Nin ies, increase the ft In other wi re opulati Ae evry os ev Words, We must bring rr own 38 popolation inversion. The population inversion can be achieved through optical winping by electrical discharges or by a chemical reaction ine Produces molecules in the ever p z One the population inversion has been achieved, the inpingtng phowan ye te sMeqountering aN excited stale molecule rather than thee tie! resect age! casequenee, photon emission rather than photon absorption is simulate Thus, the emitted photon team as greater intensity than the incident intensity. "This is called the lover ost The laser action can be pes understood by considering Fig. 10. Here the ground state inocu absorbs a photon and is nised (0 the excited state {Fig. 10). In Fig. 10(6), the excited state molecule un: dergoes spontaneous emission af a photon, reverting to the ground state. In Fig. 10(c), the impinging photon encounters a molecule that is already in the excited state: it stimulates the molecule to revert to the ground state by emitting a second photon. Here laser action has been achieved. ABSORPTION SPONTANEOUS STIMULATED EMISSION, EMISSION (a) o © Fig. 10. Elucidation of the laser action, Three types of energies can be delivered to an absorbing system to bring about population inversion, These are radiation energy, collision energy and chemical energy. Population inversion can occur in systems which have three or four energy levels (Fig. 1a). Ina three level system, irradiation of the system at the Correct frequency populates energy level 3, the process being called sptical pumping, as already mentioned. If the system then rapidiy decays to energy level 2, a (a) (o) Population inversion is said to occur (a) Three-tevel laser (6) Four-level laser benveen energy levels 2 and 1 (Fig. 11a), Population inversion Occurs better in systems having ‘our energy levels (Fig. 116). If such a system is raised to energy level 3 by pumping Say to energy level 2. a population inversion occurs between energy levels 2 and 1. For a four-level-laser to give continuous wave (cw) operation, the 1 > 0 transition should be very fst so that there can be a steady state pumping from energy level O to energy’ level 3. Fast decay Fast decay Optical pumping Fast decay Fig. 11, Energy level systems for laser action, nd there is fasta a 42 PRINCIPLES OF PHYSICAL CHENSTE Very high intensity of radiation can be obtained by placing mirrors at each end of a cayj Containing the medium of interest, The mirrors are placed in. such a maner that they (eeu Fesonant cavity for the frequency y= (Ey-E;)ihy tha is. the cavity is exactly an integra me half wave lengths in length, The electromagnetic radiation is reflected back and forth ewweeetot | mirrors and is amplified on each passage through the medium by stimulated emission of radian from energy level 2 to energy level 1. The energy required for the population inversion may Supplied by an electric discharge or by a chemical reaction as already mentioned. The laser bene highly directional since it propagates along the cavity direction, Because of its directional propery” it is extremely bright, i.e., the power per unit area per unit solid angle y characteristic feature of the laser radiation is that it is monochromatic, very high." Anothy ‘The laser radiation is coherent in the sense that the electromagnetic waves are all in sep, Spatial coherence, the waves are in step across the cross-section of the bi + tn ss n emerging from the cavity. In temporal coherence, the waves remain in step along the beam. The Iatcr is expresses | | { | | | | terms of a coherence length, ic, the distance over which the waves remain coherent and'is rela) to the range of wave lengths A present in the medium as { | i } Ie = 2212M0) For a perfectly monochromatic beam, with strictly one wave length, AX = Would remain in step for an infinite distance. However, when many waves aed out of step in a short distance and the coherence length is small. As an example, a typical light buth gives out light with coherence length of only about 400 nm; a He—Ne laser with AA = 2 pm hay Coherence length of about 10 cm. We may mention here that the laser beam can also be polarized, plane (or in some other state of polarization), by including a (4) 0 and the waves are present, the wave with its electric vector in a particular polarizing filter into the cavity. We next consider two important features of lasers, Q-switching and the mode locking. A laser can generate radiation for as long as the population inversion is maintained; it can operate continuously when heat is easily dissipated, since then the population of the upper level can be replenished by pumping. If, however, overheating occurs, the laser can be operated only in pulses (of microsecond or millisecond duration) so that the medium gets a chance to cool or the lower state discards its population. Cavity hon-resonant vis, in fact, desirable to have pulses of radiation rather than continuous output, with considerable power concentrated into a brief pulse. Such pulses can be achieved by Q-switching, the a modification of the characteristics of the resonant cavity. The name comes from the ‘Q-factor’ used as a measure of the quality | Fig 12. The principle of Qswichin of a resonance cavity in microwave engineering. The aim of Q- | The excited ae switching is to achieve a heathly population inversion in the | ths cavity is non-resonant. Thi te absence of the resonant cavity, then to plunge the population | feounes cianctensics ar se inverted medium into a cavity and hence to obtain a sudden pulse | merges ara gant pulse, q of radiation. This is illustrated in Fig. 12. Cavity resonant To achieve Q-switching, we use a Pockels cell (named after a lady scientist, Anne Pockels). which is an electro-optical devise based on the ability of some crystals, such as those of potassium dihydrogen phosphate (KH,PO,), to convert plane polarized light to a circularly polarized light when an electrical potential difference is applied. If a Pockels cell is made part of a laser cavity, it brings about a change in polarization so that the light polarized in one plane after reflection gets polarized in the perpendicular plane. The result is that the reflected light does not stimulate more emission. I however, the cell is suddenly turned off, the polarization effect is extinguished and all the ener2Y stored in the cavity emerges as an intense pulse of stimulated emission. wwoct” of? eehnique of mode locking results in the prody rhe ef osecond duration and less. A Taser Png cM pferent Frequencies, depending upon the wre fat he resonance chiaeteristics of the eavity eas MF on te number of half wave lengths af radiaicn be tapped within the mirrors (the cavity modes), 1 modes differ in frequency by multiples of 13 me vity. Normally, these modes ste dont plses relative (0 exch other, However: Wig Ne fa io lock thei phases together. ‘Then the interferencs posible wie giving a series of sharp peaks and the energy of the ns “plained in short bursts (Fig. 13) het ‘he sharpess of the peaks depends on the range of modey | 7 SS perimposed: the wider the range, the narrower the pulses. | Fig. 13. The output of a mode-locked laser us. for instance, in a taser with a cavity of length 30 cm, } Consists of a stream of very narrow pulses aks are separated by 2 ns; if 1000 modes contribute, the | *¢Patated by an interval equal to the time it the i ‘vdih of the pulses is 4 ps. takes for light to make a round trip inside the cavity Application of Lasers in Chemistry 1. Multiphoton Spectroscopy. In an incident laser beam, the photon density is so high that more than one photon can be absorbed by a single molecule giving rise to multiphoton process, with the result that states inaccessible by conventional one-photon spectroscopy become observable because the overall transition occurs with no change in parity. Thus, for example, in one-photon spectroscopy. only g <> W transitions are observable. In two-photon spectroscopy, however, the otherwise forbidden transitions g > g and w > u become observable. 2, Raman Spectroscopy. The advent of lasers has revolutionized Raman spectroscopy. With an intense laser source, the intensity of the scattered radiation (Stokes and anti-Stokes lines constituting the Raman spectrum) is increased and hence the sensitivity of Raman spectroscopy is also dramatically increased. The laser radiation, being monochromatic, also makes possible the observance of scattered light that differs by only fractions of wave numbers from the incident radiation, Such a high degree of resolution is particularly important for observing the rotational Raman lines since the rotational transitions are of the order of a few wave numbers. Also, the techniques of Fourier-transform Raman spectroscopy and resonance Raman spectroscopy can be investigated. Lasers can also be used in stimulated Raman spectroscopy. In this form of spectroscopy, the Stokes and anti-Stokes radiations in the forward direction are powerful enough to undergo more scattering thereby giving up or acquiring more quanta of energy from the molecules in the sample. This multiple scattering results in the emission of lines of frequency v; + 2%, ¥it¥m. and so on, where vj is the frequency of the ion oF incident radiation and vy, the frequency of molecular excitation. ns. The monochromatic laser radiation allows us to excite specific states with very high precision. A consequence of y for photochemistry is that the illumination of a sample may be photochemically precise and hence efficient in stimulating & Teaction because its frequency can be tuned exactly to an absorption, A related phenonrenon is the study of state-to-state reaction dynamics in which a specific state of a reactant molecule is excited and we monitor not only the rate at which it forms products but also the states in which they are Produced. 4. Time-resolved Spectroscopy. In this branch of spectroscopy. the absorption, emission or Raman spectrum of rea intermediates states of reactants, It is also possible to study ¢ molecular rota conversion from one mode of motion to another. 3. Precision Specified Trans ser pulses are used to obtain products and even transition ns, vibrations andoe PRINCIPLES OF PHIYSICA’ Gy T1144 As 5. Single-molecule Spectroscopy. In the 1990s, a new (cs nla ne ‘leveloned hich inv new experimental probes of very stall specimens, His becoming possible, wsing the specie single molecules, to understand biochemical processes such as enzymatic catalysis, folding gel of insertion of DNA into the nucleus of a cell. Also, the techniques that can probe the guint dynamics and reactivity of single molecules will be required to explore the frontier areas ol, ‘on nanometer-sized molecules. Though still a relatively new tech single-molecule spect rad has already been used to address important problems in chemistry and biology. Two such yeh PY whose details we shall not consider here but where lasers are finding wide use are thy pls optical microscopy and the far-field confocal microscopy. These techniques are based on fyore, Ret microscopy with laser excitation. ted 6. Isotope Separation. The laser isotope possible because of highly precise selectivity of lasers. Isotope separation le eeause two isorapomers (species that difey in their isotopic composition) have slightly different energy levels and have slighty gigi!’ absorption frequencies. One uses the process of photoionization, rent the ejection of an electron by the absorption of electromagnetic radiation. Direct photoionization by the absorption of a single photon does not distinguish between isotopomers because the upper level belongs to a continuum. To distinguish isotopomers we must deal with discrete states. At least two absorption processes are required. In the first step, a photon excites an atom to a higher state; in the second step another photon brings about the ionization of the excited atom (Fig. ’ a 'sotopomer 15). The energy separation between the two states involved bell in the first step depends on the nuclear mass. Hence, if the A” sotopomer’ laser radiation is tuned to the appropriate frequency, only one of the isotopomers will undergo excitation and hence will be available for photoionization in the second step. We shall consider the photoionization of uranium vapour, in which the incident laser is tuned to excite 25U but not 238U. The ?55U atoms in the atomic beam are ionized in the two- step process; they are then attracted to a negatively charged electrode and may be collected. This procedure is being | Fig. 15. 19 one method of otope sarin used in the latest uranium separation plants. ‘one photon excites an isotopomer to an excl Practical Lasers. It is customary to classify lasers as | Sots ane, en second photon ahi, gas lasers, solid state lasers, chemical lasers, dye lasers | Gepends on the nuclear mass and semiconductor lasers, etc. In the case of gas lasers, pumping is caused by passing an electric current. The ions and electrons produced are accelerated. The electrons, colliding with the gas molecules, excite the molecules of the gas. The first gas laser Produced was the helium-neon laser. It is generally used at 0-633 ym, though it can oscillate at any of the following wave lengths : 2,=3-39 pm, 22=0-633 jum and Ag=1-15 pm, While the pumping process occurs between the helium energy levels, the laser action occurs between the neon energy levels. A commonly used solid state laser is made of Y3Al,0js (i.e., yttrium aluminium g2met abbreviated as YAG) in which some of the Y* ions are replaced by Nd?* ions. It is pumped with xenon-filled flash lamp. Lasers can also be produced from organic dyes. The dye lasers have advantage that they are nmable over a range of about 30 nm. Thus, with a series of dye lasers th entire visible and near ultraviolet ranges may be covered continuously. Energy —> The so-called exeimer lasers have also been prepared, This is a peculiar type of laser in whist the lower state does not exist. An excimer is a combination of two atoms that survives ont i excited state and which dissociates as soon as the excitation energy has been discarded. An exalt of an excimer is a mixture of xenon, chlorine and neon (which acts as a buffer gas). The pass? i ; ce atts an electric discharge through the mixture produces excited Cl atoms which attach to the Xe 3"8.2.2 ProserioRs IN FLUORESCENT Licuts psorb energy and re-cmit it as light. As in the lasers we usually an impurity ion in a host lattice. However, for the uses to which phosphors are put it is not necessary lo produc intense, coherent beams of light, and the emitting process 1S spontaneous instead of induced. Phosphors have many applications, for example, the colours of your television picture are that are bombarded with clectrons from a beam (cathode rays) or from a transistor (flat screen LCD displays). In terms of tonnage produced, ‘one of the most important applications is the fluorescent light tube. Fluorescent lights produce radiation in the ultraviolet (254 nm) by passing an electric discharge through a low pressure of mercury vapour. The tube is coated inside with a white powder which absorbs the ultraviolet light and emits visible radiation. For a good fluorescent light, the efficiency of the conversion should be hieh and the emitted light should be such that the appearance of everyday objects viewed by it should resemble as closely as possible their appearance in daylight. Most phosphors for fluorescent lights have been based on alkaline earth halophos- phates such as 3Ca,(PO,),.CaF,. As in lasers, the usual dopants are transition metal or lanthanide ions, but more than one impurity ion is needed to approximate the whole visible spectrum. Not all the impurity ions need to be capable of absorbing the exciting radiation, however, as the host lattice can act to transmit the energy from one site to another. For example, in a phosphor doped with Mn?* and Sb ions, the ultraviolet radiation from the mercury lamp is only absorbed by the anti- mony (Sb) ions. The excited antimony ion drops down to a lower excited state via a non-radiative transition. Emission from this lower state produces a broad band in the blue region of the visible spectrum. Some of the energy emitted by the antimony travels through the host crystal and is absorbed by the manganese ions. The excited Mn** ions emit yellow light and return to the ground state. The two emission bands together produce something close to daylight. Phosphors have been introduced which is, for example ob a cl ot approximation to daylight. A good a ane ee y a combination of blue from barium magnesium aluminate P ivalent europium (Eu), green from an aluminate doped with cerium Phosphors ate solids which have just described, the emitter is produced by phospho299 cal Properties of S opt x) and terbium (Tb™) ions, and red from yttrium oxide doped with trivalent ce") 4 ium (Eu). ' 7. = aseen lights also emit broadband radiation in the near inf ared. Finding juminescent materials that will convert this to visible light has generated much jnterest. In this case, the incident light of lower energy than the emitted light and the process is known as upconversion. Such proc so exploited in upcon yersion lasers where a phosphor is used to produce shorter wavelength light from a red laser. Obviously, a ground state atom or ion cannot absorb a photon of one frequency radiation and then emit a photon of higher frequency radiation from the excited state reached by the absorption process. So how does upconversion work’? In upconversion systems absorption takes place in two stages. An ion absorbs a photon of the incident radiation and goes to an excited state. It then transfers most of the energy either to another state of that ion or to the excited state of another ion. If this second excited state is metastable, it has time to absorb another photon before it spontaneously emits radiation and returns to a lower state, Either the absorbing ion or one to which the energy has been transferred is now in a higher level, whose energy above the ground state is greater than the cnergy of the absorbed photon. Emission from this higher level produces shorter wavelength radiation than that absorbed. The ions doped into upconvertors are frequently lanthanide ions, or a transition metal ion and a lanthanide concentrations of Ho™ in YO, absorb red laser light Ho* in the ground state, ‘Ig, absorbs ion, For example, low and emit yellow-green light. a photon and is excited to an excited state, *F,, from where it drops to a succession of lower energy states ending in the 51, State, Figure 8.6. While in the ‘I, state the ion absorbs another Photon of red light and goes to the 5F, state from where it emits yellow-green light and returns to the ground state (Figure 8.6). In phosphors and in the ruby laser, light was absorbed localised on an impurity site, but in oth the radiation. In the next section, ther emission of radiation in soli ductors, and emitted by electrons er optical devices, delocalised electrons emit efore, we shall consider the absorption and ids with delocalised electrons, pa ticularly in semicon-5.8 PLANAR DEFECTS The introduction to this chapter mentions that crystals often contain extended defects as well as point defects. The simplest /inear defect is a dislocation where there is a fault in the arrangement of the atoms in a line through the crystal lattice There are many different types of planar defects, most of which we are not able to discuss here either for reasons of space or of complexity, such as grain boundaries, which are of more relevance to materials scientists, and chemical twinning, which can contain unit cells mirrored about the twin plane through the crystal. However, oT © Vacancy FIGURE 5.30 Successive Ti layers in the structure of TiO, 55.nd Non-Stoichiometry oie 223 shall look briefly at two forms of planar defect which are relevant to the kind ag or ral discussions in this book: of «crystallographic shear planes where the oxygen vacancies effectively collect together in a plane which runs through the crystal tergrowth structures where two different but related structures alter- nate throughout the crystal gi.1_ CRYSTALLOGRAPHIC SHEAR PLanes Non-stoichiometric compounds are found for the higher oxides of tungsten, molyb- enum, and titanium — WO,,,, MoOs.,, and TiO,.,, respectively. The reaction of these systems to the presence of point defects is entirely different from what has been discussed previously. In fact, the point defects are eliminated by a process known as crystallographic shear (CS), In these systems, a series of closely related compounds with very similar for- mule and structure exists. The formulae of these compounds all obey a general formula, which for the molybdenum and tungsten oxides can be Mo,Os,.. Mo,Osy3. W,0,,.., and W,O,,, and for titanium dioxide is Ti,O,,.,; n can vary taking values of 4and above. The resulting series of oxides is known as an homologous series (lke the alkanes in organic chemistry). The first seven members of the molybdenum trioxide series is: MoO, ,, Mo,0\s, Mo,0j7, Mo;O,9, Mo,03,, Mo;O,5, Moy03o, and Mo,,0. In these compounds, we find regions of comer-linked octahedra separated from each other by thin regions of a different structure known as the crystallographic shear (CS) planes. The different members of a homologous series are determined by the fixed spacing between the CS planes. The structure of a shear plane is quite difficult to understand, and these structures are usually depicted by the linking of octahedra as described in Chapter 1. WO, has several polymorphs, but above 900°C the WO, structure is that of ReO,, which is illustrated in Figure 5.31 (ignore the bold squares for the time being). (The structures of the other polymorphs are distortions of the ReO, structure.) ReO ismade up of [ReO,] octahedra that are linked together via their comers; each corner ofan octahedron is shared with another. Figure 5.31(a) shows part of one layer of linked octahedra in the structure. Notice that within the layer, any octahedron is linked to four others; itis also linked via its upper and lower corners, to octahedra inthe layers above and below. Part of the Res structure is drawn in Figure 5.32, wl You can see that every oxygen atom is shared between two metal atoms. As six Y€ens surround each Re, the overall formula is ReOs. 5 Inetg wSoichiometry in WOs., is achieved by some of the cone a we Octahene changing from comer-sharing to Steet poet caer ae ing the sg tFKed in bold in Figure 5.31(a). The edge-sharing correspon WT i : © Structure so that the chains of bold octahedra are ee le ae ure 53 4 : i es and is inepesey (Dis shearing oecurs at regular inlets yeah) Sed with slabs of the ‘ReO,’ structure (comer linked eras “S groups of four octahedra which share edges. The direction of max aity of the edge-sharing groups is called the crystallographic shear plane zd i ed by an arrow in (b). hear alters the stoichiometry of WO,, we need» find the stoichiometry of one of the groups of four octahedra that are linked together by sharing edges; part of the structure depicting one of these groups is presented is Figure 5.33(a). The four octahedra consist of 4 W atoms and 18 O atoms (Figs 5.33(b)); 14 of the oxygen atoms are linked out to other octahedra (these bonds 2 indicated), so are each shared by two W atoms, while the remaining 4 oxygens only involved in the edge-sharing within the group. ‘The overall stoichiomety * given by [4 W + (14 x '/,)0+ 4 O], giving W,O,,. Clearly, if groups of four octahedra with stoichiometry W,O,; are imterspe! throughout a perfect WO, structure, then the amount of oxygen in the struct ® reduced and we can write the formula as WO,.,. The effect of introducing the £07 of four in an ordered way can be quantified. If the structure sheared in such 4% that the entire structure was composed of these groups W,0,,. If one [WO,] octahedron is formula becomes [W,0,, + WO, |= the formula would beco™ , ert! used for cach group of four, then the W,0,,. Clearly, we can extend this Pre any number of [WO,] octahedra regularly interspersed between the £FOUPS! Wouraed wee W,0,, + 3WO, = W,0 W,0,, + 4WO, = W,O,,; WO), + WO, = W,Ous WO, + 6WO, = Wy0,5; WO ¢ WO! = W One : The basic formula of the group of four, W,O,, can be writen a5 W,Or! Ne, n= 4. This formula also holds for all the other formulae that are listed. TH |