Unit 4 Quantum Mechanic It is established that classical mechanics was incapable of explaining the optical spectra emitted by atoms, of even the existence of atoms. Bohr developed a model for the atom of hydrogen by assuming, the quantization of the angular momentum, which was an introduction to wave ‘or quantum mechanies. Quantum mechanics isa more precise approach to describe nearly all physical phenomena which reduces to classical mechanics in the limit where the masses if energies of Une particles ie large or macroscupic, The success of quantum mechanics can be established with historically important examples of blackbody radiation and wave- particle duality. ‘The basic idea behind quantum physies is that energy is not necessarily continuot instead, energy comes in suiall packets called quanta. Quantum physics explains the wave me- Ghanica} phenomenon of microscopic and submicroscopic particles such as electrons, protons, neutrons, ete. Alvo it explains the particle nature of light, in quantum physics, we consider the wave nature of particles and particle nature of waves light}. Perhaps it is not surprising that quantum physies plays an important role in many of today’s electronic devices, 4.1 Blackbody Radiation ‘All bodies, no matter how hot or cold, continuously radiate electromagnetic waves. For instance, we see the glow of very hot objects because they emit electromagnetic waves in the visible region of the spectrum, However, at relatively low temperatures, cooler objects emit visible light waves only weakly and, as a result, do not appear to be glowing. At a given temperature, the intensities of the electromagnetic waves emitted by an object vary from wavelength to wavelength throughout the visible, infrared, and other regions of the spectrum. An extremely important experimental discovery which played a central role in the development of quantun mechanics was the problem of the spectral density of a blackbody radiation. If we take a body with a surface that absorbs any radiation (a blackbody) we find that it emits radiation at different wavelengths. By definition a ‘blackbody’ is an ideal system that absorbs all the radiation incident on it. Since a blackbody is the perfect absorber, it must also be the ideal emutter. As in Fig.4.1, a cavity with a tiny opening absorbs any radiation entering it, so it acts like a blackbody. The radiation emitted by the opening when the walls of the cavity are hot is characteristic of a blackbody. The intensity (power per unit area) of the radiation emitted between wavelengths ’ and +d is defined as df = R(A)dA, (4.1) 123Quantum Mechanics where, R(A) is called the radhancy. The spectral dependence of K(A) has a well- defined de- pendence on the wavelength and temperature of the blackbody In Fig. 4.2 we show how the experimentally observed RA) behaves at different temperatures. The total mtensity has the Container at a temperature T Detected radiation Figure 4.1: A cavity with a tiny opening absorbs any’ radiation entering it, s0 it acts ike a blackbody. Radiation can be detected by allowing it to leave an otherwise closed cavity through « pinhole. The radiation is reflected many times within the cavity and comes to thermal equilibria with the wall Radiation leaking ‘out through the pinhole is characteristic of the radiation inside the container. Visible Radiation intensity per unit wavelength 0 500 1000 1500 Wavelength (am) Figure 4.2: (a) The spectrum of blackbody radiation at two different temperatures. At the higher temperature, ‘more radiation is emitted and the peak shifts to shorter wavelengths. behaviour I=oT*. (4.2) This is known as Stefan's law. The constant o is called the Stefan-Boltzmann constant. It is found to have a value o = 5.67 x 10-°Wm~?K*. ‘The radiancy maximizes at a certain wavelength Ames, as can be seen from Fig4.2. The temperature dependence of this wavelength is given by AmezT ® 2.898 x 10™ mK, (4.3) 124ee 4.1, BLACKBODY RADIATION This relation is known as Wien’s displacement law. According to classical physics, radiancy or intensity per unit wavelength interval 2 is IQ) = Stho§ a ke®. (a) This result is known ax the Rayleigh-Jeans formula represents our best attempt to apply clas sical physics to understanding the problein of blackbody radiation. In Fig 43, the intensity calculated from the Rayleigh-Jeans formula is compared with typical experimental results When careful experiments were carried out and the spectral der y tabulated. it was found that the Rayleigh-Jeans law was applicable only in a small frequency range In fact. as can be seen from the equation, the classical law predicts an infinite energy density at very short wavelengths-an obviously unphysical result. The failure of the Rayleigh-Jeans formula at short wavelengths is known as the ultraviolet catastrophe. It is apparent in the case of blackbody radiation that the classical theories do not work and that @ new kind of physical theory is needed. Rayleigh—Jeans Planck's g Ss function i s 2 2 Sie : § 2 . 2 oop a Pere Pee eee. 12345678910 12345678910 Wavelength (umn) Wavelength (um) @) o Figure 4.3: (a) The failure of the classical Rayleigh-Jeans formula to fit the observed intensity. At long wavelengths the theory approaches the data, but at short wavelengths the classical formula fails miserably. (b) Planck's function fits the experimentally observed data perfectly. 4.1.1 Planck’s Hypothesis Max Planck (1858-1947, Germany), his work on the spectral distribution of radiation, which led to the quantum theory, was honored with the 1918 Nobel Prize. In 1900, Planck calculated the blackbody radiation curves, using a model that represents a blackbody as a large number of atomic oscillators, each of which emits and absorbs electromagnetic waves. To obtain agreement 125| Quantum Mechanics between the theoretical and experimental curves. Planck assumed that the energy £ of an atomic oscillator could have only the discrete values of E = 0,hf,2hf,3hf, and so on. In other words, he assumed that £ = nhf.n = 0,1,2,3.... Here n is either zero or a positive integer, J is the frequency of vibration (in hertz), and h is a constant now called Planck's constant. It has been determined experimentally that Planck’s constant has a value of h = 6.62606896 x 10-*J.s. From the mathematical standpoint, the difference between Planck's calculation and the classical calculation using Maxwell-Boltzmann statistics is that the energy of an oscillator at a certain wavelength or frequency is no longer a continuous variable—it is a discrete variable that takes only the values. Based on Planck's result, the intensity of the radiation then becomes 109-5 (Cs) faery] = [er] : The perfect agreement between experiment and Planck's formula is illustrated in Figure 4.3(b). 4.1.2 Einstein’s Theory Albert Einstein was intrigued by Planck's hypothesis that the electromagnetic radiation field must be absorbed and emitted in quantized amounts. Einstein took Planck's idea one step further and suggested that the electromagnetic radiation ficld itself is quantized and that “the energy of a light ray spreading out from a point source is not continuously distributed over aan increasing space but consists of a finite number of energy quanta which are localized at Points in space, which move without dividing, and which can only be produced and absorbed as complete units.” We now call these energy quanta of light photons. According to Einstein each photon has the energy quantum E hf. (46) where f is the frequency of the electromagnetic wave associated with the light, and his Planck's constant. Notice that Equation (4.6) is consistent with Planck's relation for quantum of energy (E = nhf). The photon travels at the speed of light c in a vacuum, and its wavelength is given by Afze. (47) Im other words, Einstein proposed that in addition to its well-known wavelike aspect, amply exhibited in interference phenomena, light should also be considered to have a particle-like aspect For example, when one of these quanta, called a photon, penetrates the surface of a metal, all ofits energy may be absorbed com- pletely by a single electron. If ¢ is the energy necessary to remove an electron from the surface of a metal (¢ is called the work function and is a characteristic of the metal), the maximum kinetic energy of an electron leaving the surface will be hf — @ as a consequence of energy conservation. (Some electrons will have less than this ‘amount because of energy lost in traversing the metal.) Thus, the stopping potential should be given by 1 ey = (jm) =hf-¢. (4.8) This equation is referred to as the photoelectric effect equation. 1264.2, PARTICLES, WAVES AND PROBABILITY 4.2 Particles, Waves and Probability 4.2.1 Photons and Light waves In Maxwell’s electromagnetic theory. we had a seemingly complete description of light as an electromagnetic wave. Now we find, through the photoelectric and Compton effects that the light sometimes manifests itself as particles. What's the connection between wave and particle descriptions? Ina photoelectric experiment. the rate at which electrons are ejected depends on the light’: intensity. Since an electron is ejected when it absorbs a photon, we conclude that the number of photons in the incident light is proportional to light intensity. Now, the intensity of an electromagnetic wave depends on the square of the electric or magnetic field. ‘The fields, in turn, obey Maxwell's equations, so one aspect of a photoelectric experiment-namely, the rate of electron ¢jection-relates to Maxwell's description of light as an electromagnetic wave. ‘We can quantify the relation between waves and photons, but only in a statistical sense. The ejection of individual electrons in a photoelectric experiment is quite random. The uncertainty principle prevents us from following a photon trajectory and predicting when and where an electron will be ejected. All we can say is that electrons are more likely to be ejected where the wave intensity is greater. Specifically, the probability that an electron will be ejected is directly proportional to the intensity of the incident electromagnetic waves that is, to the square of the wave fields. More generally, the probability of finding a photon in a beam of electromagnetic waves is directly proportional to the wave intensity (Fig4.4) High amplitude Light wave high probability of finding a photon Figure 4.4: The probability of finding @ photon is directly proportional to the intensity of the electromagnetic wave. The figure i unly suggestive hecause we can't depict photons as localized particles, In this quantum-mechanical description, the fields still evolve according to Maxwell's equa- tions. For example, the fields of an electromagnetic wave undergoing double-sit interference develop regions of maximum and minimum wave intensity- the bright and dark bands of the interference pattern. But the wave fields determine only the probability that individual photons will be detected in the interference pattern. That's why a very short exposure or a low-intensity beam results not in a weak version of the interference pattern but in a seemingly arbitrary pat- tern, Only with large number of photons does the statistical pattern emerge (Fig.4.5). In quantum mechanics, then the relation between the wave and particle aspects of light is this: As long as we don’t try to detect the light, it propagates as a wave governed by Maxwell's equations. But when we detect the light, we do so through interactions involving individual photons. Those interactions are random events whose probability depends on the wave intensity that is, on the square of the wave fields. 127Quantum Mechanics Figure 4.5: Developinent of a two-slit interference pattern from random photon events: (a)approximately 50 photons,(L) 250 photous.(c) 1000 photons. (d) 10,000 photon 4.2.2 Electrons and Matter waves In de Broglie’s remarkable hypothesis, that matter as well as light, exhibits both wave and particle properties. The wave-particle duality puts matter and light on essentially the same footing, and the statistical interpretation is the same for each. Fig.4.6 shows a beam of part cles anid its associated de Broglie matter wave. Just as the probability of finding a photon proportional to the wave intensity that is. the square of the electromagnetic field amplitude so the probability of finding a particle is directly proportional to the square of the matter-wave amplitude. And as with light, the particle nature of matter manifests itself only when we try to detect a particle: leave it alone, and the particle's behaviour is governed b its wave nature. If particles have wave character, then what is the appropriate wavelength? For light, the momentum of a photon is p = h/A. De Broglie used the same relationship for particles and proposed as the relevant wavelength for matter waves hh pom (4.9) This wavelength, the de Broglie wavelength, depends on the mass m and speed v of a particle. This equation used the relativistic form of the momentum, but the literature often presents this nonrelativistic approximation for the de Broglie wavelength: A Ah As -=— pom (4.10) As Figure 4.7 shows for the case of an electron, at speeds of up to 40% of the speed of light, the nonrelativistic approximation is very close to the exact result given by Eq4.9. As you can see from Figure 4.7, the de Broglie wavelength of an electron, even one moving at 10% of the speed of light, is on the order of one-tenth of a nanometer * CHECK: What are typical de Broglie wavelengths for macroscopic objects? 4.2.3 Uncertainty Relation How precisely is it possible to measure physical properties such ax location, momentum, en- ergy, and time? In addition, to what precision can two properties be measured simultaneously? This question is never considered in classical mechanics, where it is assumed that all dynamic Quantities can be measured with arbitrary precision with improved instrumentation. ‘The as- tonishing statement arising from quantuin physics is that the momentum and the position of 1284.2. PARTICLES, WAVES AND PROBABILITY. High amphiude Matter wave igh probability of finding a photon Figure 4.6: A beam of particles and its associated matter wave 0.05 ¢ 0.04 = 0.03 A : ~ 0.02 0.01 02 04 16 08 1 vic Figure 4.7 the nonrelat relength of an electron as a function of its speed. Curve (A) is the exact result; curve (B) is istic approximation, aan object cannot be measured simultaneously with arbitrary precision. The more precisely we attempt to measure an object’s momentum, the less precise the information on its position has to become, and vice versa. This physical statement is cast in mathematical terms in the form of the Heisenberg Uncertainty Relation: Az.Ap > zh (4.1) Here, Ax and Ap are the particle's to be measured position and momentum respectively. This relation was discovered in 1927 by the German physicist Werner Heisenberg (1901-1976) and caused a revolutionary change in our understanding of the measurement process, as well as of our fundamental ability to know the physical world. Now it is important to note that, if Maxwell's equations determine the behaviour of light waves, what equation describes matter waves? In 1926 the Austrian physicist Erwin Schrédinget answered this questions with his Schrédinger wave equation. In the same year Schrédinger showed that his wave theory was cquivalent to a matrix-based theory that Heisenberg, Max Born, and Pascaul Jordan had formulated in 1925. Heisenberg received the 1932 Nobel prize in physics, and Schrédinger shared the 1933 Nobel prize with Paul Dirac for their contributions to quantumn theory. 1294.3 The Schrédinger wave equation (Time independent form) ‘The Schrédinger equation is a fundamental postulate of quantum mechanics just as Newton's laws are fundamental postulate of classical mechanics. If matter does possess wave like prop- erties (matter waves), then there must lic some sort of wave equation that govern them let us consider the classical one-climiensional wave equation as Fu _ du = =US5 (4.12) ar" oa oe where ‘v’ is the velocity or speed of the wave. This equation can be solved by the method of separation of variables (x and t) and let us write the solution u(z,t) as: u(e.t) = U(x) cost, (4.13) where y(c) is the spatial amplitude of the wave, Putting Eq. (4.13) in (4.12), we get (4.14) Here w = 2nf and Af = v, then we get fey 0. (4.15) By the idea of de Broglie matter waves, the total energy of a particle is the sum of its kinetic energy and potential energy, s0 Py e=-2 +Vv@) (4.16) where p = mv is the momentum of the particle and V(z) is its potential energy, So from Eq.(4.16) p= VimE—V(a)} (4.17) As A= h/p (according to de Broglie formulation) h Jie Vor (4.18) putting Eq.(4.18) in (4.15) gives dy 2 SS Be vee) (4.19) where fi = h/2n. Eq.(4.19) can also be written as BP ed? -- + V(z)¥(z) = Ev(z). (4.20) ‘This is the time independent Schrédinger equation, a differential equation whose solution, ¥(z) namely the wave function describes a particle of mass ‘zn’ moving in a potential field described by V(z). 1304.4 Time dependent Schrédinger equation Let us consider the classical wave equation as av Or? Asu=fd~f Es (where kis the wave number and *f" is the frequency and ce = 25 fs the angular frequency of the wave). the solution for the Eq-(1.21) can be written as ule.t) = Acos(kr - wt). (422) ‘or a more gencral solution is in the form u(r.) = Act (423) As B= hf. 4 = h/p, k= 2n/d and w = 2n/f, then Eq.(4.23) becomes (zt) = Acker 2). (aay Note that, this expression exploits wave particle duality and translates the classical wave intu matter wave. By differentiating Ey.(4.24) with respect to ‘t’ gives (4.26) a (4.28) oe If V(e, ) is the potential energy then the total energy is given as P b= sV(x0) (429) am Multiplying both sides by v, we get ,. & bes Fev V(b (4.30) Putting Eq. (4.26) and (4.28) in (1.30) gives nay = - ESE ven. Sag t VOW (a3) ‘This iv known ws the time-dependent Schrodinger equation, where U(x, t) is called as the wave function, The equation (4.31) is important whenever the potential energy V(x, () is a function of time. For example, when an electron interacts with light (wn electromagnetic wave), the electron is subjected to the periodically varying electric and magnetic field of light, 1314.5 Free particle Lot us consider a quantum particle (for example electron) of mass *m’ which is experiencing no forces. This is called a free particle and this is a simplest system in quantum mechanics has the potential energy V = 0 everywhere. If we consider an one-dimensional case, with motion only in the z-direction only, then the time-independent Schrodinger equation is given a> (aw = (2) = ey, (32 2m (B) ” ) After simplification, we get - b ri qth where cme Possible solutions are, Wz) = Csin(kz) or C-eos(kx) or Ce*), (4.33) ‘The value of k or E is anything, so there is no restriction. Thus a free particle, even in quantum mechanics, can have any non-negative value of the energy. The energy levels in this ease are not quantized and correspond to the same continuum of kinetic eniergy shown by a classical particle. Also, a free particle cannot be represented by a single plane wave. It has to be represented by a wave packet. The physical solutions of the Schrodinger equation are thus given by wave packets, not by stationary solutions. 4.6 Particle in an infinite potential well Let us consider an electron moving along the z-axis of aligned metallic tubes held at different potentials as shown in Fig.4.8. If the central tube is grounded, so a test charge or an electron inside the tube has zero electric potential energy and experiences no electric force. When both center metallic tubes @) and @) are held at higher electric potential V, strong electric fields arise in the gaps at 0 and L. As the applied voltage V is increased without limit and the gaps are simultaneously reduced to zero, then the idealization is known as the infinite potential well or “box” potential (see Fig.4.8) sespess seepete Una ® Ly, THF THF 4 v 7 i td © Figure 4.8: (2) Aligned metallic tubes serve to confine a charged particle. The charge ¢ or the electron moves, fraty ‘rithin the iniddle tube but encounter electric forces iu ue naps: (L) A charge whove energy sles than aV is cont 1994.6. PARTICLE IN AN INFINITE POTENTIAL WELL © | y=o 0 — x Figure 49° As U = qV, and if V is increased and the gaps between the cylinder of Fig.4.8(a) are narrowed, the potential energy approaches that of the infinite square well ‘The potential energy U(z) is that of the square well in Fig.4.9 is: for UL (4.34) Since there is no chance that the particle inside the well (between 0 < x < L) can penetrate the rigid walls at z = 0 and at 2 = L, the wave function W(z) is to be zero in the region x L. Inside the box U(x) = 0. Now, this particle has a wave function whose time independent part is given by the Schrédinger equation (time independent type): Ray Soma U(a)¥(x) = Ev(z). (4.35a) = &y | 2m Ta + le - Ue@lve@) =0. (4.350) ‘As within the well U(x) = 0, then Eq.(4.35b) becomes (4.36a) or (4.366) with (4.37) ‘The most general solution of this equation is: ¥(z) = Asinkz + Bouskz, (4.38) mE where k = As the wave function y is continuous, and Y = 0 outside the well (the shaded region), y must be zero at z = 0 and x = L: So (0) = 0 and ¥(L) = 0. These are the boundary conditions. Now Eq.(4.38) becomes: 0 = ¥(0) = Avind + Bcos0 = 0+ B 133a Quantum Mechanics So B must be zero, then Eq. (438) becomes W(z) = Asinkz. (4.39) As =Oatx=L, then w(L) = AsinkL = 0. (4.40) As the amplitude A # 0, then to satisfy Eq.(4.40),, sinkL = 0 which means kL = nm, where ‘n’ is an integer (n = 1,2,3....). Now nt a-7 (441) 2x? mE Sok? = 7 = (from Eq.(4.37)) or nthe B= gop tah (4.42) where fi = h/2n, Thus a particle trapped in a rigid box can have only discrete or quantized energies, The lowest energy (ground state) has n = 1 and is given by BR "* BmE?” then the next highest energy (n = 2) is Ey = 4E; and for higher energies (Ref. Fig.4.19), (4.43) Ey =9E) E, = WE, The integer ‘n’ is called the quantum number of state. 0 0 Figure 4.10: Possible energy levels for a particle in an infinite potential well ‘Correspondingly, the wave function for each of the quantum state is va = Asin (Fa). (4.44) 1344,7. PARTICLE IN A THREE DIMENSIONAL BOX The value of ‘A’ can be determined by normalizing °¢," by i vninde = [ WalayPae or or cos 21) oF (4.45d) ‘Thus the wave function is given as Unla) = Ean (2) (4.46) ‘The wave functions and the probability distributions for a particle in an infinite potential well for the states with n = 1,2,3... is shown in Fig.4.11 oo oo eo oc ¥ n=3 n=3 hel? n=2 n=2 n=1 1 0 L 0 L Aer x Figure 4.11: Standing wave type patterns of the wave function w and |v. 4.7 Particle in a three dimensional box Let us consider a quantum particle (for example electron) of mass ‘mn’ constrained to move freely within the space of the rectangular box (a metallic crystal), for example with edges of length a.b and c along X, Y and Z-axes as shown in Fig.4.12. 135y' Figure 4.12: A three dimensional rectangular box Here the potential barrier which exists at the surface of the crystal will be in the form of rectangular box for the frec electron inside the crystal. The origin is O as shown in Fig.4.12. Now. a three dimensional time-independent Schrédinger’s equation in the box has to be solved. The 3D Schrédinger equation is given as (4.47) As the potential energy V = 0 inside the box and it is infinity (V = 00) at the boundaries of the box, then Eq.-(4.47) becomes, FY YAY IME Ox? Sptaet we (4.48) where tis the function of the three variables , y and z. Let this wave function is equal to the product of three wave functions X, Y and Z as w(z,y,2)=X(2)¥(y)4(z)_ (or) w= XYZ. (4.49) If we substitute Eq.(4.49) in (4.48) and after simplification, we get C2 mE is oo = RE = 1, (4.50) where eae K? = K2+k2+K? aS (4.51) So, Eq.(4.48) becomes, 1&X an Xae ook (4.528) 1ey : rare -K}, (4.52) 1€Z_ ya aaa 7K (4.52c) 1364.7. PARTICLE IN A THREE DIMENSIONAL BOX Similar to the solution for one-dimensional Schrédinger’s wave equation and applying boundary conditions, we get e : 2 nar X= wan a (4.53a) - 2. nyzy Ye Vian 5 (4.53b) 2 1z Z = ff ein BZ (4.53) one (4.53) or ete (2eeenenas 2 wf nemz a = Bain - Vie Fain EE (4.53¢) From Eq.(4.51), ne a E=7- (K? + Kj +B?) (4.54a) or - mgk\? (My)? (net)? e-F {()'+ (C8) +()} (asa) As nz, ny and n, takes different values, So Hf ymey? , (my)? , (Me)? : Es Bay Bua = 3 { (2) +(#)'+(®) }. (4.55) where nz = Ny = 1, = +23. Eq.(4.55) gives the eigenvalues of energy of the particle. These values are called the energy levels of the particle. 4.7.1 Degeneracy ‘An interesting feature of the solutions for a particle in a three-dimensional box is obtained when the edges of length a, and c along X,Y and Z-axes are equal to L. Then Eq.(4.53(a)) 3 at (e/2 myTE Mey. Nahz v= (V3) sin 3 sin “TY sin, (4.56) (4.57) becomes and Eq.(4.55) becomes We 2 2 2 Exons = gouge ta +73 + 13). Here, the quantum numbers nm, 2, 73 are positive numbers, Now consider the case, my = m2 = 3 2)" nr, my. me = (f2) sin ZZ sin ™ sin 4.58 v (v2) sin TF sin sin, (4.58) and a i (4.59) ng = 1, thenQuantum Mechanics However, the first excited state is associated Iymy = 2,my = mz = 1 then hny = 2m = ny = nz = Qn = ny = L peepee eetereeae de 3 hare (f2) sin 2 sin 22 sin 222 viaa= (Vz) sin Zsin Bsn and the energy eigen value is 3 2\" ae day. nz via = (fF) sin sin 4 sin S, Gh? anBs (4.60) Faia = Byaa = Buaa= ‘We see that, although the wavefunctions are different, they are degenerate, meaning that they correspond to the same energy. In this case, in which there are three degenerate wavefunctions, the energy level (647/82 L*) is coumnon; hence, is there is the threefold degeneracy in the first exited state. The degree of degeneracy increases in higher-lying energy levels. The occurrence of degeneracy is related to the syminetry of the system and for the present case it is a cubic box. * Exercise: Consider a particle in a cubic box. What is the degeneracy of the level that has an energy (14/3) times that of the lowest level’ (Ans: 6). 1384.8. CORRESPONDENCE PRINCIPLE 4.8 Correspondence principle ‘The correspondence principle states that at sufficiently high energies, quantum mechanics agrees with classical mechanics. For example. if a particle in an infinite potential well case, the plots of |W). the probability: densities can be approximated as some constant value Tt ineans at larger 'n’, the particle or the electron were prescut in all regions of the potential well (between length 0 << L) with equal probability (see Fig.4.13). wr Wwe wr n=2 . oe ugh Figure 4.13 4.9 Barrier penetration and quantum tunneling Let us consider a quantum particle of mass m (like electron) and energy E travelling to the right along the 2-axis in free space where the potential energy U = 0. So the energy is purely kinetic energy.(Refer Fig 4.14(region 1)) U 0 x=L x— Figure 4.14: ‘The wave function of the approaching particle in region 1, where U = 0 is sinusoidal. If the particle is a classical one, then the particle would be reflected at z = 0. However, as like in the finite potential well case, the wave function of the particle decays exponentially within the barrier region (Il). If the thickness of the barrier is not large, the wave function may not decay completely to zero on the otherside (region II). So for x > L there is again a sinusoidal wave function, as U = 0 and E > 0. However in region III, the appeared sine wave is having 139___ Quantum Mechanics lesser amplitude than that of wave in region | (Refer Fig 4.15). As |W? is non zero bevond the barrier. so there is non zero probability that the particle penetrates the barrier of region I ‘This process is called tunneling through the barriet or barrier penetration This type of process +s used in scanning tunneling electron microscope U x x— Figure 4.15: A particle with energy less than the height uf the potential barrier has x probabilty of tunneling through the barrier. If the barrier thickness is ‘L’, then the tunneling probabilty is described with a transmission coefficient T, which is ( for small T’ values) Tac™, (4.61) where (4.62) Note that from Eq.(4.62), increasing the height of the barrier Up or increasing the thickness of the barrier L, will reduce the value T. 4.10 Tunneling microscope A microscope called as scanning tunneling microscope(STM), works under the principle of quantum barrier tunneling. This type of microscope is used to resolve individual atoms in a given material specimen. In a STM, a pointy metal tip is placed near the sample. A potential difference is established between the tip and the surface of the sample as shown in Fig.4.16. If the tip is at a lower potential than the sample, the electrons in the tip are attracted to the surface. The vacuum between the tip and the sample surface is acting as an insulator or as 1 potential barrier. Now it is possible for an electron to tunnel through this barrier. As the probability of tunneling depends on the width of the barrier and width of the barrier is the length of the path the electron covers from the tip to the sample. A longer path means lower tunneling probability (Fig.4.17). The tip is scanned over the sample, changing the path length By measuring the current that results from clectron tunneling as the tip is scanned, an image of the surface of the sample is revealed Interestingly, the strong electric field Letween the tip and the sample can also be used to move individual atoms along the surface of the sample to create quantuin structure. 140Figure 4.16: Schematic diagram of the STM Tunnelling is more a likely for shorter path Atoms of the J tip Tien je ‘Vacuum (barrier) Surface atoms _7 ofthe sample Figure 4.17: Electrons tunnel through vacuuin(barricr). 4.11 Electron microscope ‘Transmission electron microscopes (TEM) are microscopes that use a particle beam of electrons to visualize specimens and generate a highly-magnified image. TEMs can magnify objects up to 2 million times. Electron microscopy (EM) is mainly used for obtaining high resolution images of biological and non-biological specimens. The high resolution of EM images results from the use of electrons (which have very short wavelengths) as the source of illuminating radiation. The transmission electron microscope is used to view thin specimens (tissue sections, molecules, etc) through which electrons can pass generating a projection image. The TEM is analogous in many ways to the conventional (compound) light microscope. In a conventional transmission electron microscope (CTEM, or TEM for short) (Fig. 1.18). a thin specimen is irradiated with an electron beam of uniform current density. The acceleration voltage of routine instruments is 100-200 kV. Medium-voltage instruments work at 200-500 KV to provide better trans- mission and resolution, and in high-voltage electron microscopy (HVEM) the acceleration voltage reaches 500 kV-3 MV. Electrons are emitted in the electron gun by thermionic, Schottky, or field emission. The latter are used when high gun brightness and coherence are needed. A three- or four-stage condenser-lens system permits variation of the illumination aperture and the arca of the spec ‘men illuminated. Specifically, TEM employs an electromagnetic lens which focuses the electrons into a very fine beam. The electron- intensity distribution behind the specimen is imaged with 4 lens system, composed of three to eight lenses, onto Auorescent screen. An image of the 41specimen with its assorted parts shown in different shades according to its density appears on the screen. The image can further be recorded by direct exposure of a photographic emulsion or an image plate in-side the vacuum, or digitally via a fluorescent screen coupled by a fiber-optic plate to a CCD camera Electrons interact strongly with atoms by elastic and inelastic scattering. The specimen must therefore be very thin, typically of the order of 5-100 nim for 100 keV electrons, depending. on the density and elemental composition of the object and the resolution desired. Special preparation techniques are needed for this, Electron gun Condenser 1 ens Cy Aperture Condenser 2 lens Cz Aperture Scan coils ‘Specimen Objective lens Objective Aperture Back focal-plane of objective Ditlraction Aperture First intermediate image plane Ditfraction lens Intermediate lens Projector 1 lens Projector 2 lens Image Figure 4.18: Simplified diagram of a transmission electron microscope. Exercises : 1. An electron has a kinctic energy of 3.00eV. Calculate its wavelength, Solution: 1424.11, ELECTRON MICROSCOPE KE £ E. (or) p= VamE 6.626 x 10s 7.09 10 “in = 0.709 9.11 x 10°31 kg)(3)(1.60 x 10-9 J)} 2. Calculate the de Broglie wavelength for a proton moving with a speed of 1.00 x 10% m/s. Solution: Js Eg)( x 105 m/s) 3. An electron is confined between two impenetrable walls 02am apart. Determine the energy levels for the states n = 1,2 and 3. = 3.97 . Taking the reciprocals of these give 1,0,0, and reduce them to the lowest common set of integers. we get (100). The central pane (ii) is the (110) plane and lower right most panel (ji) is the (111) plane. Ap important consequence of this is that the {11]] direction in the crystal is the normal direction to the (111) plane. Similarly, the (100) and [110] directions are normal to the (100) and (110) planes, respectively. Hence, the Miller indices define a plane through its nor! al vector. If a plane vote the axis on the negative side of the origin, corresponding index is negative. Thus, for gn intercept along the negative axis, the indices of the plane are represeuted by taking the reciprocals of the intercept and placing a bar (-) over the integer (intercept value) as shown in Figure 5.9(b)(iv - vi). intersections One characteristic of a crystal that can be determined is the distance between nearest equivalent parallel planes. Another characteristic is the surface concentration of atoms number per square centimeter (#¢/cm?), that are cut by a particular plane. Aguty & single-crystal pe enductor is not infinitely large and must terminate at some surface. The surface density aretoms may be important, for example, in determining how another material such as an insulator, will “fit” on the surface of a semiconductor material, Figure 5.8: A representative crystal plane 156EXAMPLES OF CRYSTAL STRUCTURES : ‘in : z on E ls peor” 9 ’ PY Y : 100 111 oe Wie > fe A, é iH c y x HH 7 ZY Ml x : Foo 010 007 (iv) w) (wi) (b) in cubic unit cell, Here, Miller indices, a shorthand notation to id planes in @ unit cell are indicated. Figure 5.9: Representative crystal pla describe certain crystallographic direction 5.6 Examples of crystal structures sstal family, whose three and face-centered cubic Most metals and semiconductor materials belong to the cubic © members are the simple cubic (SC), body-centered cubic (BCC) (FCC)—and the related diamond structures (See 510). The number atoms associated with each unit cell can be calculated by considering an atom’s location and atont sharing with ad- jacent unit cells. The simple formula by which, the number of atoms per unit cell, N’, can be 157,Et _Grystal Physics computed as N, NaN4M 3 (5.1) where, N, is the mmber of inter 5 of comer atom Mer of interior atoms. 1; is the umber of fare atoms ane! Ny is the number For SCN, = 0. Ny=0 and Ne=8; therefore, N=1 For BCC: N, = 1, Ny=0 and N.=8; therefore, N=2, For FOC: Ny 0, Ny -6 and Ne 8; therefore, N- 4 For Diamond: N, = 4, Ny=6 and N, therefore, N=8, fa) sc Bcc Foc DIAMOND Figure 5.10: (a) The basic cubie erystal structures (reduced-sphere unit cell): simple eubie (SC), body- centered cubic (BCC), and face-centered eubic (PCC). (b) The diamond structure (left) aud x tetrahedron a its building block (right) Although the simple cubic lattice is conceptually the simplest of all lattices, real crystals of atoms are rarely simple cubic, except polonium (of all the chemical elements, polonium is the only one which can form a SC lattice with a single atom basis). When describing crystalline structures, atoms (or ions) are thought of as being solid spheres having well-defined diameters that weakly attract each other and therefore try to pack close together. ‘This is termed the atomic hard sphere model in which spheres representing nearest-neighbor atoms touch one another (refer Figure 5.11). Mostly, BCC and FCC lattices are realized more frequently in nature than SC. For example, the elements Al, Ca, Au, Pb, Ni, Cu, Ag (and many others) are FCC whereas the elements Li, Na, K, a-Fe, Mo, Cs (aud many others) are BCC. 158sc BCC FCC Figure 5.11: (a) Packing of hard spheres in basie cubic erystal structnres. (hj A cutaway of the conventional runt cell exposing how the unit cell lattices have empty space 5.6.1 Co-ordination number and atomic radius Co-ordination number is the number of nearest neighbours to au atom in its atomic arrange- ment. Assuming that atoms are spherical in shape and are in contact in a crystal, atomic radius can be defined as half the distance between the centres of two neighbouring atoms. The atomic radius is represented by r and is usually expressed in terms of length of edge (a) in a unit cell. By knowing the structure of the crystal, atomic radius may be calculated as follows (Refer Figure 5.12) d Ai Ab 8 (@) SC structure (©) BCC structure (@ Foe structure Figure 5.12: (a) SC, (b) BCC and (c) FCC structures. (a) Simple cubic structure: From Figure 5.12(a), let a denote the lattice parameter (length of the cube edge) and r the atomic radius. Then a = 2r and r = a/2 159Crystal Physics (b) BCC structure: Let a denote the lattice parameter and 5 the atomic radius. From re 5.12(b), AG = r+ 2r-+r = dr and from Figure 5.120), EG? = a 2a. As EG? + AE? = 2a? + a? = 3u®, then Mir? = 3a? and ¢ = via/4 AG? jeter and r. the atomic 202, Hence. (c) FCC structure: From Figure 5 12(c), let a denote the lattice param radius, For the triangle ABG. AG? = AB? + BG? = a2 + a? = 2a?. So, (167)? r= aV2/4, or r = a/2V2 = 0.35350. 5.6.2 Atomic packing fraction eat each lattice point and by choosing the radius of the spheres to be such that neighboring spheres just touch (Refer 5.11). Then then packing fraction is the fraction of the volume of all the space which is enclosed by the union of all the spheres ‘That is, the ratio of the volume of the spheres to the total volume. Thus the atomic packing fraction for the case of SC, BCC and FCC structures are calculated 2s follows: If we consider a lattice with a sph unit cell is Vaeu = a. For the case of SC, each cell corre: ) of this sphere is a/2, s0 its volume is Vephere (i) The volume of a convention sponds to a single atom or sphere. The radits (r 4 = (#) (a/2)8. Thus the packing fraction is Viphere/Veeu = 7/6 = 0.52 tentional unit cell is Vi«u = a2. For the case of BCC, each con- hhich are at a distance av/3/2 apart from each ) (iii) The volume of a convent ional unit cell is Veer = a3. For the case of FCC, each con- ventional unit cell contains four lattice points which are at a distance av/2/2 apart from each ; (ii) The volume of a conv ventional unit cell contains two lattice points w other. The radius (r) of the each sphere is aV3/4, 0 its volume is Vephere = (47/3) ( ~ ‘Thus the packing fraction is 2Vsphere/ Veet = aV¥3/8 = 0.68. other. The radius (r) of the each sphere is aV/2/4, so its volume is Vephere = (4n/3) ( Thus the packing fraction is 4Vspnere/ Verw = #/(3V2) = 0.74. 5.6.3. Hexagonal Closed-Packed (HCP) Crystal Structure ‘The third common metallic crystal structure is the hexagonal close-packed (HCP) structure shown in Figures §.13(4-c). Many metals have the HCP structure, including magnesium (Mg), titenium (Ti), zine (Zn), and cadmium (Cd). In crystallization of metals, the metal atoms can attain a lower energy and more stable condition by forming the HCP strueture of Figure 5.13(a). Here, the hexagonal close-packed (IHCP) structure consists of one atom each at the all comers of the hexagonal unit cell, one atom each at the center of the two hexagonal planes ‘and three atoms in the center of the unit cell in a triangular fashion such that one atom lies in alternate triangular prisms as depicted in Figure 5.13{a-c). Like in FCC erystal structure, in HCP crystal structure, each atom is surrounded by 12 atoms (co-ordination number = 12) with the difference in their atomic packing. Hence, we have 160PLES OF CRYSTAL STRUCTURES _ Hexagonal P lattice + two atoms —> Hexagonal close-packed (HCP) structure, ors would effectively contribute 2 Each corner atom contributes 1/6 atom, thus the 12 cor atoms. The two atoms at the opposite faces, contributing half atom each, would contribute 1 atom. Thice atoms forming a triangle in the middle, would contribute 3 atoms. This makes total contribution of 24143 = 6 atoms. ‘The ration of the hexagonal height. «. to the basal side, a,c/a is called as c/a ratio. It is 1643 for (ideal) HCP made of uniform spheres. Metals with c/a tatio more than 1.633, have their atom elongated along the e-axis, and those with ef ratio less than 1,633 have atoms compressed along the c-axis. Usually Mg and Co. each having this c/a ratio as 1.623. Metals having larger de are Cd (1.890) and Be (1.568) ‘Top layer of atoms @ Sena ayer Bottom layer of Toatoms. B8y (@) Figure 5.13: (a) Three dimensional hexagonal structure lattice (hard-sphere model), (b) Schematic of the chose packed HOP crystal structure (c) Cut seston of unit cal (HCP), (d) Representative postions of inner atoms 161Atoms Jadye length / ano a Figure 5.14: (a) Relation betwen eg kength an aoa ras (b) Basal triangle BOR (3) ¢.a Ratio: The three imier atoms, located at the middle of the HCP unit cell. form an equilateral triangle of side a, same as the edge length of the HCP unit cell. Moreover. these atoms aro so symmetrically located that a normal from the face-centered atoms mect at the centroid of this triangle. Also, the distance of the face-centered atoms from each of these three atomis is same as that of the edge length of the HCP unit cell. Figure 5.13(b) shows: the face- centered atoms as A and B, and the three inner atoms, M, N and P. Figure 5.13(c) depicts the cut section of a unit cell and Figure 7?(d) shows the representative positions of inner atoms, AO is a normal on the equilateral triangle with O as the centroid. Hence PO SPE (5.2) From Figure 5.13(d), PB = VPP= MEF = Sat) = (eT = 53) Hence, (5.4) Now from Figure 5.13(d), we have AO = c/2 = VAP? — OP? = fa — (0/3) (5.8) cfa= (5.6) (ji) Atomic packing fraction of HCP unit cell: Atomic packing fraction = (Volume of atoms in HP unit cell)/(Volume of HCP unit cell). «Volume of atoms in HCP unit cell = 6 x Fae =6x fnta/2)* = nat ‘Atomic radius r = a/2 as the atoms touch each other along the HCP edge length (5.16(a) Hence, volume of HCP unit cell = Basel area x Height = 6x Area of ABQR x 1625,7. REAL CRYSTAL STRU ‘The base of the HCP unit cell is made of six identical equilateral triangles. One of those triangles ABQR, is shown in Figure 9.16(b). Area of ABQR = 3 Base « Height Hence, Volume of HCP unit cell is (58) So, atomic packing fraction is (5.9) 5.7 Real crystal structures 5.7.1 Diamond structure, FCC with a multiple-atom basis A single structure dominates the semiconductor industry. ‘The elemental semiconductors (St, Ge, and gray Sn) share the diamond cubic structure shown in Figure 9.15. The structure differs from its FCC and BCC counterparts primarily because it has a multiple-atom basis, even though all of its atoms consist of the same element Bravais lattice with two atoms associated with cach lattice point ‘A key feature of this structure is that it accommodates the tetrahedral bouding configuration ‘The structure of diamond crystal unit cell is considered as the interpenetration of two FCC i carbon unit cells along their body diagonals by a distance of 7th the diagonal length. The origin of one unit cell (basis, first-atom coordinate) is at_ (000) and the origin of another unit ‘This structure is built on an FCC- nd cight atoms per unit cell. | 1 cell (basis, second-atom coordinate) is at li 454, is| ‘The projection of lattice points on bottom face of the cube are shown in Fig.5.15(d). The fractions denote height of lattice points above the bottom face, interms of cube edge. Because of the tetrahedral bonding geometry of the diamond cubic structure, the atoms lie along body diagonals. Inspection of Figure 5.15 indicates that this orientation of atoms leads to the equality (if r is the radius of the atom and «is the unit cell edge length): 163Crystal Physics, Interior toms located a positions one ‘quarter ofthe distance along the body diagonal fa) (e) Tw atoms per — aevce pont (d) () Figure 5.15: (a) Diamond cubic unit cell showing atom positions (Refer Fig.5.7 and 5.10). The diamond ying FCC lattice with a two-ntom basis (note the representative solid circles). Each structure has an under atom is tetrahedrally coordinated dashed lines shows one tetrahedral unit, (b) Projection of diamond lattice points on bottom face, (c) The actual packing of full-size atoms associates with the unit eel, (J) Tetrabeadral unit of the diamond lattice. The erystal Ved 62 |8x2=8, Typical semiconductors 18x 5=8, Typical Inet The atoms are comiected by covalent bonds. The cube outlined by the structure: diamond eubie, Bravais lattice: FCC, Atoms/unit cell: 4 Si, Ge, Gray Sn. 1 2r = F(body ~ diagonal) = ‘The unit-cell volume then is, Bs = (4.62)4r5 = 98 55 The volume of the eight atoms in the unit cell is %=8x free 33.5r3, which gives an atomic packing factor of 1645,7. REAL CRYSTAL STRUCTURES 5r8 Vue 98.578 APF = = 0.310 ‘This result represents a very open structure compared with the tightly packed metals’ struc: (c.g. APF = 0.74 for FOC and HCP metals) tures amond cubic erystal structure, then If-we consider silicon crystal structure which is having di unit cell is if the radius of a silicon atom. r= 0.117nm, then the volume of the Vue = 98.5r3 = 98.5(0.11710)! = 0. 158nm°. By considering 8 atoms per unit cell, the calc ulated density is. 2.36 g/cm 5.7.2. Compound,multiple element,crystals Is. such as the III-V and II-VI semiconductor es with historical names such as depicted in Figure 5.16. In the case of binary compoun s. the crystal structures have multiple-atom snultiple- element bass as the NaCl structure, the Zn$ structure, and the CsCl structure . “” . — - 2 . . 4 ob oe () ions with small spheres Na* ions) (a). ZnS filling half the tetrahedral holes. There are coll origin in. Figure 5.16: (a) Illustration of the NaCl (Big spheres are C1" (The zine blende structure isan tice of anions, with cations oon ai a and four anions per cell) (b) and wurtzite (c) unit cells Asterisks denote the 1 each case, (a) Nacl lattice ‘The NaCl structure is also called the rocksalt structure. Examples of materials that crystallize this way include PbS, PbSe, and PbTe, ‘These are all interesting materials from a thermoelec- is FCC with a two-atom/two-element tat tric or photo- voltaic applications. The underlying lattice clement and basis. These atoms are located at (0, 0, 0) for the 4 5 for the second element, Note that the NaCl structure can alternatively be viewed as two interpenetrating 11 FOC lattices offset by | 5a. 54: 2a The NaCl structure contains 8 atoms per unit cell. with 4 atoms from the first element aud 4 from the second. This can be seen in Figure 6. 16(a), where there are # corner and G edge atoms of the first element. In tandem. there are 12 edge atoms 165Crystal Physics and 1 interior atom from the second. (b) ZnS or zine blende lattice The ZnS lattice is also referrad to wy the zine blend: (ZB) or sphaleite structure. | Binary compounals that crystallize with this atomic arrangement include ZnS, GaAs, ZnTe. and CATs It is identical to the diamond structure by possessing two-atom/two-element basis with atom: located at (0, 0, 0) (first element) and ze ie 7 (second element). Note that, just as with nS structure as two interpenetrating FCC the diamond unit cell, itis possible to visualize the lattices offset. by qe ‘The Zn$ structure contains 8 atoms per unit cell, of which 4 belong to the first element and 4 to the second. This can be seen in Figure 5.16(b), where 4 atoms of one clement are located completely inside the unit cell. The remaining 4 atoms reside as 8 corner atoms and 6 face atoms. Figure 5.17: (a) The (110) plane of the zinc-blende lattice an (b) the arrangement of different atoms within (210) plane. Unit cell length is a. Consider Figure 5.17, where a (110) plane of the zinc-blende lattice with the atomic positions of the atoms in the unit cell is shown, The significance of this plane is that all three primary directions [100], [110], and [111] lie in this plane. The unit cell is rectangular with length a along [100] and av2 along |110), which form the two perpendicular axes for this structure. The [111] direction points diagonally across the rectangle at an angle of approximately 35.26° with respect to the [110] direction. 1665.8. CARBON STRUCTURES (c) Wurtzite Lattice ‘The wurtzite structure is the compound material version of the single-element hexagonal close- packed structure. Compounds that crystallize with this arrangement include CdS and CdSe ‘The underlying lattice is hexagonal with a four-atom two-clesnent basis. These atoms are lo- oie) 201 cated at (0.0.0) and | 5a. 52, 5c} for the first clement and (0.0.4) and | 5a. 3 second element, where 5 is just an off: set along the = direction of the unit cell. The unit cell exist as 8 corner atoms is shown in Figure 5.16(c) and contains four atoms per unit cell. They and 1 interior atom of the first element along with | edge atoms and 1 interior atom of the second, 1 sc +6] for the xr 5.8 Carbon structures ure, carbon-based materials have Carbon being one of the most fundamental elements in nat All organistns on Earth consist of attracted a great deal of attention during the past century carbon-based structures, and many materials important to human society are based on carbon. Carbon is the Group 14 element that has four valence electrons and forms different types of chemical bonds (covalent bonds) using several hybridizations (sp. sp. andsp*). Carbon’s im- portance currently extends beyond abstract science into the technology of electronics. 5.8.1 Graphite Carbon atoms can bond together to produce of diamond, In graphite, bonding involves the orbital set sp? + p. The three sp? orbitals are directed in a plane at angles of 120°C, and the p orbitals overlap. Thus, the p electrons are delocalized—that is, not restricted to the region between two C’ atoms but shared among many C atoms within a plane of C atoms. This type of bonding produces the crystal structure shown, in Figure 5.18. Each carbon atom forms strong covalent bonds with three neighboring carbon ‘atoms in the same plane, giving rise to layers of carbon atoms in a hexagonal arrangement. Bonding within layers is strong, but the intermolecular forces between layers are the much weaker van der Waals forces. We can see this through bond distances. The CC bond distance within a layer is 142 pm; between layers, it is 335 pm, a solid with properties very different from those ae Sch 25 385 pm i de oe ee Figure 5.18: The layered crystal structure of graphite, the stable room temperature form of carbon. 167Crystal Physics Jecause bonding Its unique crystal structure gives graphite some distinctive properties. Beca on s weak. the layers can glide over one another rather easily ‘Asa result, graphite between ee Tha mild pressure is applied is a good lubricant, cither in ars ea a in A 7 ne re Pants pec As. boca the peletons av deca ny migrate throw he pans Of carbon atoms when an electric field is apphed: graphite conducts electricity. An im at use of graphite is as cloctrodes in batteries and in industrial electrolysis. In contrast, ¢ as is not an electrical conductor because all its valence vlectrons are localized or permanently ft into single covalent bonds. ppens when we USe 2 5.8.2. Graphene Graphene is an isolated (single) sheet of graphite. ‘This sheet of carbon atoms is only one atom thick. An example is shown in Figure 5.19. Thus the layers of graphite are graphene. This system has recently been studied extensively for its unique electronic properties, especially alter the discovery in 2004 that it conild he isolated using Seotch tape. This discovery Ted to the 2010 Nobel Prize in Physies for Andre Geim and Konstantin Novoselov. Graphite Graphene Figure 5.19: Mlustration of au exfoliated grapline sheet ‘A carbon nanotube is a sheet of graphene rolled into a cylinder, A fullerene is obtained when an appropriate number of hexagonal rings in graphene are replaced by pentagonal rings; the presence of the pentagonal rings causes the flat sheet to pucker aud form a spherical ball ‘The relationship between graphene and other forms of carbon is illustrated in Figure 5.20 Graphene, a single atomic-layer material is exceptionally strong. Graphene has very interesting electronic properties Lecause electrons in these sheets are moving very quickly—at approxi. mately 1/300 of the speed of light. Graphene’s electrical properties provide the potential for the increasing miniaturization of integrated circuits. Its high conductivity combined with high optical transparency suggests the potential for transparent conducting electrodes. It is expected that graphene will play an important role in the develo spiment of electronic devices, possibly replacing silicon, : 1685.9. CRYSTAL IMPERFECTIONS Fullerene Carbon nanotube Figure 5.20: The relationship betwwen graphene and the other forms of carbon 5.9 Crystal Imperfections dimensional perfectly ordered arrangements of Theoretically, (perfect) crystals are defined as 3- but contains imperfections atoms or ions. In engineered materials, the lattice is not perfect or defects; that is, the perfect geometric periodicity is disrupted in some manner and there will be some missing of atoms ions or misalignment of unit cell in an otherwise perfect crystal. From the electrical properties’ point of view, single crystals are preferred over poly-er talline ‘materials because imperfections tend to alter the electrical properties of a material and, in some cases, electrical parameters can be dominated by these defects or impurities. Also, even in a single crystal, deviations from perfect long-range order are possible due to unavoidable random thermal vibrations of atoms. However, the average positious around which the atoms vibrate do form a crystal structure. Other causes of deviations from perfect crystallinity are the formation of defects of various types. There are three basic types of imperfections: point defects, line defects (or dislocations), ‘and surface defects. These imperfections only represent defects in or deviations from the perfect ‘or ideal atomic or ionic arrangements expected in a given crystal structure, The material is not considered defective from a technological viewpoint. In many applications, the presence of such defects is useful. Often the defects may be created intentionally to produce @ desired set of electronic, mag- netic, optical, or mechanical properties. For example, pure iron is relatively soft, yet, when we add a small amount of carbon, we create defects in the crystal structure of iron and transform it into a plain carbon steel that exhibits considerably higher strength. Similarly, a crystal of pure alumina is transparent and colorless, but, when we add a small amount of chromium, it creates ‘a special defect, resulting in a beautiful red ruby crystal. The effect of point defects, however, is not always desirable. When we want to use copper as a conductor for microelectronics, we use the highest purity available. This is because even small levels of impurities will cause an orders of magnitude increase in the electrical resistivity of copper!. (a) Point defects (zero dimensional defects): The important point defects are: 169. Crystal Physics (i) Vacancies: Whenever one or more as shown in Figure §.21(a). the defect cansed is known as vacancy (i) Interstitial ~ defects: Whenever an extia atoin (generally sinaller than the parent atoms) occupies interstitial position in the crystal system withont dislodging the parent atom as shown in Figure 5.21(b), the defect caused is known as interstitial defect (it) Substitutional — defect: Whenever an impurity atom (other than the parent atoms) occupies a position which was initially meant for a parent atom, as shown in Figure 5.21(c.d), the defect caused is known as substitutional defect (iv) Frankel — defect: It consists of an ato 1 displaced from its regular site to an interstitial site. The atom in the interstitial site and the vacancy caused by the displacement of atom from the regular site (Figure 5.21(¢)) together are named Frankel pair (v) Schottky ~ defect: In ionic crystals if there isa vacancy in positive ion site, then charge neutrality of crystal can be maintained by creatin site (Figure 5 ‘4 atoms are missing from a normally occupied position, 1 vacancy in the neighlouring negative ion 21(f)), such a pair of vacant site is called Schottky defect (d) (e) (f) Figure 5.21: Point defects: (a) vacancy, (b) interstitial atom, (c) small substitutional atom, () large substi- tutional atom, (e) Frenkel defect, and (f) Schottky defect. All of these dofects alter the perfect arrangement of the surrounding atoms. (b) Line defects (one dimensional defects): These defects are mostly due to misalignment of atoms or presence of vacancies along a line. Line defects in crystalline solids are termed as dislocations. The dislocations are classified into two categories (i) Edge dislocation and (ii) Screw dislocation. By referring Figure 5.22(a), for the case of edge dislocation, an extra portion of a plane of atoms, or half plane, the exlge of which terminates within the crystal. However, 1705.9. CRYSTAL IMPERFECTIONS for the case of screw dislocation, one part of eryst ee re ee ee en eaenats Sinlocation line (Figure &:22(b), Both acrew end cage ellocation may ee ceaeeiibe crystal forming a curved dislocation, ‘To describe the size and dite sar ot the mai lattice distortion caused by a dislocation, the Burger veetor Wy is intrvduced. To fine! the Burger vector wwe make a circuit from atom to atom counting the same number of atomic distances in all directions, If the circnit encloses a dislocation it will not close. The vector that closes the loop is the Burger vector i Screw dislocation \ ie “hh wn to ‘ I alee) Edge dislocation @) (b) Figure 5.22: Dislocations: (a) The line along which shearing occurs 8 Sore dislocation. A Burgers veetor P evrequired to elese a loop of equal atom spacings around the sew dislocation (b) A Burgers veetor b is required to close a loop of equal atom spacings around the edge dislocation Dislocations are most significant in metals and alloys since they provide a mechanism for Plastie deformation refers to irreversible deformation or change in shape that occurs when the force or stress that caused it is removed. The applied stress causes dislo- cation motion, and the cumulative effect of slip ‘of numerous dislocations is plastic deformation. (c) Surface defects: Surface defects are the boundaries, or planes, that separate a material into regions. For example, each region may have the same crystal structure but different orientations (i) Material ~ Surface: The exterior dimen the crystal abruptly ends. Each atom at the surface no longe eimber, and atomic bonding is disrupted. The exterior surface 1A also be very rough, may sain tiny notches, and may be much more reactive than the bulk of the material. (i) The microstructure of many engineered ceramic and metallic materials consists of many grains, A grain isa portion of the material within which the arrangement of the atoms is neatly entical; however, the orientation of the atom arrangement, or cxystal structure, is different foreach adjoining grain. Three grains are shown schematically in Figu'6 5.2%; the arrangement of atoms in each grain is identical but the grains are oriented differently. A grain boundary, plastic deformation. fons of the material represent surfaces at which ¢ has the proper coordination imrystal Physics 1s, is a narrow zone in which the atoms are not the surface that separates the individual gr jaterials are properly spaced. Aside from the electronics dustry, most. practical engineering polycrystalline rather than in the form of single crystals. The predominant micros! feature of many engineering materials is the grain structure. Many materials’ properties are highly sensitive to such grain structures. Grain boundaries are more chemically reactive than the grain themselves. Grain boundaries usually occur when two crystals begin growing SD rately and then meet. Grain size varies inversely with the grain boundary area, ‘The growth of grains depends on temperature. structural @000000 @0000000 ©000000 e@ e000 @00000 aus equilibriuin spacing or arrangeme' 5.10 Crystal growth The development of and improvement in the formation or growth of pure single-crystal semi- conductor materials is responsible for the success in fabricating very large scale integrated (VLSI) circuits. For these VLSI circuits, semiconductors are needed in purest form. Silicon, than 1 part in 10 atoms, The high for example, as concentrations of most impurities of | ry in the growth and the treatment of purity requirement means that extreme care is necessar the material at each step of the fabrication process 5.10.1 Growth from a melt ‘A common technique for growing single-crystal materials is called the Czochralski method. In this technique, a small piece of single-crystal material, kuown as a seed. is brought into contact with the surface of the same material in liquid phase, and then slowly pulled from the ‘melt. As the seed is slowly pulled, solidification occurs aloug, the plane between the solid tiquid interface. Usually the erystal is also rotated slowly as it is being pulled, to provide a slight stirring action to the melt, resulting in a more uniform temperature. Controlled amounts of specific impurity atoms, such as boron or phosphorus, may be added to the melt so that the grown semiconductor crystal is intentionally doped with the impurity atom. Figure 5.24 shows 172.CRYSTAL GROWTH cee a schematic of the Cvochralski growth process and a silicon ingot or boule grown by this process ‘After the semiconductor is grown, the boule is mechanically trimmed te the proper diameter axl a flat is gronid over the entire length of the boule ty denote the crystal arientation, The flat is perpendicuior to the {110} sineetion or indicates the (110) plane (See Figs then allows the individual chips to be fabricated along, given crystal planes so that the chips can be sawed apart mote easily: The boule is then sliced into wafers, The wafer must Ie thick enough to mechanically support itself A mechanical two-sided lapping operation produces a ilat wafer of uniform thickness. By polishing, « smooth surface om which deviees may be fabricated or further growth processes may be carried out. This final seaniconduetor wafer is called the substrate material Seed Crystal Figure 5.24: Schematic of the Czochralski erystal growth process axial growth ‘A common and versatile growth technique that is used extensively in device and integrated Circuit fabrication is epitaxial growth. Epitaxial growth is a process whereby a thin, single- crystal layer of material is grown on the surface of a single-crystal substrate. In the epitaxial process, the single-crystal substrate acts as the seed, although the process takes place far below the melting temperature. When an epitaxial layer is grown on a substrate of the same material, the process is termed homoepitaxy. In a heteroepitaxy process, although the substrate and epitaxial materials are not the same, the two crystal structures should be very similar if single-crystal growth is to be obtained and if a large number of defects are to be avoided at the epitaxial-substrate interface. One epitaxial growth technique that has been used extensively is, called chemical vapor-phase deposition (CVD). Silicon epitaxial layers, for example, are grown on silicon substrates by the controlled deposition of silicon atoms onto the surface from a chemical vapor containing silicon. In one method, silicon tetrachloride reacts with hydrogen at the surface of a heated substrate. The silicon atoms are released in the reaction and can be deposited onto the substrate, while the other chemical reactant, HCI. is in gaseous form and is 173Crystal Physics swept out of the reactor. A sharp demarcation between the inpurity doping in the substrate and in the epitaxial layer can be achieved using the CVD process. This technique allows great flexibility in the fabrication of semiconductor deviecs. A more versatile technique for growing epitaxial layers is the molecular beam epitaxy (MBE) process, A substrate is held in vacuum at @ temperature normally in the range of 400” to 800°C a relatively low temperature compared with many semiconductor processing steps. Semiconduc- tor and dopant atoms are then evaporated onto the surface of the substrate. In this technique. the doping can be precisely controlled resulting in very complex doping profiles. Complex ternary compounds, such as AlGaas, can be grown on substrates, such as GaAs, where abrupt changes in the erystal composition ure desired. Many layers of various types of epitaxial com positions can be grown on a substrate in this manner. These structures are extremely beneficial in optical devices such as laser dioules. 5.11 Diffraction from crystalline materials It is important to note that an atom can scatter x-rays, and if many atoms are together then the scattered waves from all the atoms can interfere. If the scattered waves are in phase (coherent), they interfere in a constructive way and we get diffracted beams in specific directions. These \Girections are governed by the wavelength 2 of the incident radiation and the nature of the crystalline sample. Bragg’s law, formulated by W. L. Bragg in 1913, relates the wavelength of the x-rays to the spacing of the atomic plan ‘To derive Bragg’s law, we begin by assuming that each plane of atoms partially reflects the incident wave much like a half-silvered mirror. ‘The x-rays are not really being reflected but they are actually being scattered. he diffracted bean is composed of waves that have been scattered by all the atoms of the crystal lying in the path of the incident beam. Reflection of visible light is a surface phenomenon occurring in « layer about 1/2 thick Consider the diffracted wave in Figure 5.25. It is assumed to make the same angle, 8, with the atomic planes as does the incident wave. Now the criterion for the existence of the diffracted wave is that the scattered ("reflected") x-rays should all be in phase across a wavefront such as BB’. For this to be so, the path lengths between wavefronts AA’ and BB! for the rays shown must differ by exactly an integral number (11) of wavelengths ’. Therefore the path difference, 6, must be 174Figure 5.25: Dihaction of X-rays by a crystal b=n, (5.10) where n is an integer. Now since lines CC’ and CD in Fig. 5.25 are also wavefronts, we can write 6 = DE+ EC’ = 2EC". (5.11) From elementary trigonometry, 6 =2CEsind. (5.12) and because CF is the interplanar spacing a’, we can write 5 = 2d'sind (5.13) So, we get nd = 2d sind (5.44) This equation is known as Bragg’s law. ‘The diffraction of x-rays of a single wavelength, takes place only at certain angles, (known as the Bragg angle), where there is constructive interference and Bragg’s law is thus satisfied. However, the reflection of visible Tight takes place at any angle of incidence. ‘Thus, Brage’s law gives the angular positions of the reinforced Fiffrected beams in terms of the wavelength 2 of the incoming X-rays and of the interplanar spacings dof the crystal planes. It is extremely important in indexing x-ray diffraction patterns tnd, hence, for determining things like the crystal structure of the materiel. The parameter 7 is Known as the order of reflection and is the path difference, in terms of number of wavelengths, retwaen waves scattered by adjacent planes of atouts, as indicated by Eq. (6.14). & first-order reflection oceurs when n = 1 and the scattered and incident waves have a path difference of one wavelength. When n> 1, the reflections are called higher-order. So where, as defined in Fig5.25. d! = diay corresponds to the spacing between crystal planes of indices (ikl) and the parameter a’/u corresponds to the spacing between planes (uhinknl). 80 175,Crystal Physics we can consider a higher-order reflection as a first-order reflection from planes spaced at # distance of 1/n of the previous spacing. By setting d = d'/n and substituting into Eq, (5.15). wwe can write Bragg’s law as A= 2dsind. (5.16) As an example, the -HO reflection ean be considered a fourth-order reflection from the (ilo) 10) planes of spacing d = d'/2, oF convention is planes of spacing d or a second-order reflection from the even as a first-order reflection from the (440) planes of spacing d = d'/4. This consistent with our definition of Miller indices. (Note: Higher-order reflections are very impor tant for clectron diffraction in the transmission electron microscope.) It is important to note that, when the condition expressed by Bq.(5.14) is not satisfied destructive interference occurs and uo reinforced beam will be produced. Constructive i terference occurs when the condition given by Eq.(5.14) is satisfied leading to peaks in the intensity distribution. For solids which crystallize in cubic structures, the interplanar spacing dua, between two closest parallel planes with Miller indices (hkl) is given by 2 (5.17) h? + aki where a represents the lattice constant. In order to calculate the angles of diffraction we substitute Eq,(5.17) in the Brage’s law [Eq.(5.14)] to obtain 4 sind = nd (5.18) If we consider first order reflections only (n = 1), Eq.(5.18) can be written in the form d sind = VN, (5.19) 2a where, N = h?+k?+12, Now, for a simple cubic lattice, all values of (hkd) are possible implying the possible values of N = 1,2,3,4,5,6,7,...(SC). Similarly, for a BCC lattice h + k +1 must be even implying N = h? +k? +2? = 2,4,6,8, 10,.... Finally, for an FCC lattice, Miller indices are either all even or all odd implying N = A? + k? +? = 3,4,8,11,12, 16,19... Hence, for a e and for given values of d and a, one can now easily calculate the different values given structure of 8. 5.12 X-ray diffractometer X-rays are clectromagnctic waves with very short wavelengths. Note that diffraction effects ‘occur with different types of electromagnetic waves. For example, the atoms in a crystal are arranged in a periodic array, forming planes of atoms. ‘These planes reflect electromagnetic radiation, leading to interference of the reflected waves. For diffraction from the atomic planes in a crystal, the effective slit spacing is the distance between atomic planes, which is typically 3X107!m and hence much smaller than the spacing in a diffraction grating. To observe diffr tion using the atomic planes, one therefore needs to use wavelengths much shorter than visible light, and X-rays with wavelengths on the order of 10-!m are a convenient choice. Experi- ments of this type are called X-ray diffraction. By measuring the angles that give constructive interference of reflected X-rays, one can determine the spacing between atomic planes. just as the spacing between the slits in a diffraction grating can be determined from the diffraction 176