Lecture: Solid State Chemistry (Festkörperchemie) Part 1 H.J. Deiseroth, SS 2004 Lattice Energy and Chemical Bonding in Solids Lattice enthalpy The lattice enthalpy change H L is the standard molar enthalpy change for the following process: 0 M+(gas) + X-(gas) MX(solid) H L 0 Because the formation of a solid from a „gas of ions“ is always exothermic lattice enthalpies (defined in this way !!) are usually negative numbers. If entropy considerations are neglected the most stable crystal structure of a given compound is the one with the highest lattice enthalpy. Lattice enthalpies can be determined by a thermodynamic cycle Born-Haber cycle A Born-Haber cycle for KCl (all enthalpies: kJ mol-1 for normal conditions standard enthalpies) delete standard enthalpies of - sublimation: +89 (K) ionization: + 425 (K) atomization: +244 (Cl2) electron affinity: -355 (Cl) lattice enthalpy: x Calculation of lattice enthalpies 0 L V AB V Born VAB = Coulomb (electrostatic) interaction between all cations and anions treated as point charges (Madelung part of lattice enthalpy („MAPLE“) VBorn = Repulsion due to overlap of electron clouds (Born repulsion) Calculation of lattice enthalpies 1. MAPLE (VAB) (Coulombic contributions to lattice enthalpies, MADELUNG part of lattice enthalpy, atoms treated as point charges ) 2 V AB z z e A N 4 0 rAB Coulomb potential of an ion pair VAB: Coulomb potential (electrostatic potential) A: Madelung constant (depends on structure type) N: Avogadro constant z: charge number e: elementary charge o: dielectric constant (vacuum permittivity) rAB: shortest distance between cation and anion Calculation of the Madelung constant Cl Na typical for 3D ionic solids: Coulomb attraction and repulsion Madelung constants: CsCl: 1.763 NaCl: 1.748 ZnS: 1.641 (wurtzite) ZnS: 1.638 (sphalerite) ion pair: 1.0000 (!) 12 8 6 24 A 6 ... = 1.748... (NaCl) 2 3 2 5 (infinite summation) 2. Born repulsion (VBorn) (Repulsion arising from overlap of electron clouds, atoms do not behave as point charges) Because the electron density of atoms decreases exponentially towards zero at large distances from the nucleus the Born repulsion shows the same behaviour VAB r0 r approximation: V Born VAB B r n B and n are constants for a given atom type; n can be derived from compressibility measurements (~8) Total lattice enthalpy from Coulomb interaction and Born repulsion 0 L Min.(V AB V Born) (set first derivative of the sum to zero) 2 z z e 1 0 A N (1 ) L 4 0 r0 n Measured (calculated) lattice enthalpies (kJ mol-1): NaCl: –772 (-757); CsCl: -652 (-623) ... (measured means by Born Haber cycle based on the known enthalpy of formation !!) Applications of lattice enthalpy calculations: lattice enthalpies and stabilities of „non existent“ compounds and calculations of electron affinity data (see next transparencies) Solubility of salts in water (see Shriver-Atkins) Calculation of the lattice enthalpy for NaCl z z e 2 1 A N (1 ) L 4 0 r0 n 0 0 = 8.8510-12 As/Vm; e = 1.610-19 As (=C); N = 6.0231023 mol-1 A = 1.746; r = 2.810-10 m; 1/40 = 8.992 109 n = 8 (Born exponent) e2N = 1.54210-14 (only for univalent ions !) HL = -1.387 10-5 A/r0 (1-1/n) (only for univalent ions !) ------------------------------------------------------------------------------------------------------------Vm A2s2 VAs Dimensions: ------------------ = ----------- = J/mol As m mol-1 mol NaCl: HL‘ = - 865 kJ mol-1 (only MAPLE) HL = - 756 kJ mol-1 (including Born repulsion) Can MgCl (Mg+Cl-) crystallizing in the rocksalt structure be a stable solid ? HFormation ~ -126 kJ mol-1 (calculated from Born Haber cycle based on similar rAB as for NaCl !!) MgCl should be a stable compound !!!!! However: Chemical intuition warns that there is a risk of disproportionation: 2 MgCl2(s) Mg(s) + MgCl2(s) -------------------------------------------- HDisprop = ???????? 2 MgCl 2 Mg + Cl2 HF = +252 kJ Mg + Cl2 MgCl2 HF = -640 kJ ----------------------------------------------------------------------- (twice the enthalpy of formation) (from calorimetric measurement) 2 MgCl Mg + MgCl2 HDispr = -388 kJ (Disproportionation reaction is favored) Calculation of the electron affinity for Cl from the Born-Haber cycle for CsCl Standard enthalpy of formation sublimation ½ atomization ionization Lattice enthalpy - 433.0 kJ/mol 70.3 121.3 373.6 - 640.6 HFormation = Hsubl + ½ HDiss + HIon + HEA + HLattice HEA = HF – (HS + ½ HD + HI + HL) HEA = - 433 – (70.3 + 121.3 +373.6 –640.6) = -357.6 Chemical bonding in solids bonding theory of solids must account for their basic properties as: - mechanism and temperature dependence of the electrical conductivity of isolators, semiconductors, metals and alloys (further important properties: luster of metals, thermal conductivity and color of solids, ductility and malleability of metals) ... Temperature dependence of the electrical conductivity of metals and semiconductors (isolators) key to nderstanding: „band model“ = alloy increasing resistivity resistivity below Tc = 0 !! decreasing resistivity = isolator The origin of the simple band model for solids: Band formation by orbital overlap (in principle a continuation of the Molecular Orbital model) the overlap of atomic orbitals in a solid gives rise to the formation of bands separated by gaps (the band width is a rough measure of interaction between neighbouring atoms) E << kT ~ 0.025 eV s- and p-bands in a one-dimensional solid Energy s-band p-band - Whether there is in fact a gap between bands (e.g. s and p) depends on the energetic separation of the respective orbitals of the atoms and the strength of interaction between them. - If the interaction is strong, the bands are wide and may overlap. Isolator, Semiconductor, Metal (T = 0 K) electrical conductivity requires easy accessible free energy levels above EF Energy conduction band (empty) band overlap e- EF hole:+ o valence band (filled with e-) Isolator (> 1.5 eV) Semiconductor (< 1.5 eV) Metal (no gap) (at T = 0 K no conductivity) EF = Fermi-level (energy of highest occupied electronic state, states above EF are empty at 0 K) A more detailed view for metals: The Fermi level At T = 0 K the lowest ½ N electronic states in a solid with N atoms are occupied (each level is occupied by 2 electrons with opposite spin). At T > 0 K states above EF are also occupied to a certain extend. The population P is given by the Fermi Dirac statistics a special version of the Boltzmann statistics which takes into account that no more than two electrons can occupy any level Origin and temperature dependence of electronic conductivity in metals - Electrons with energies close to the Fermi level can easily be promoted to nearby empty levels. - they are mobile and can easily move through the solid in an electric potential gradient („free electrons“) At elevated temperatures thermal vibrations of the atoms reduce the mobility of the „free electrons“ . As a consequence the electrical conductivity decreases with increasing temperature A more detailed view not only for metals: (DOS = Densities of states) E << kT ~ 0.025 eV The number of electronic stes in a range divided by the width of the range is called the density of states (DOS). - simplified, symbolic shapes for DOS representations !! Typical DOS representation for a metal Typical DOS representation for a semimetal Semiconductors conduction band band gap thermal excitation of electrons valence band T=0K T>0K Semiconductors The electrical conductivity of a semiconductor: ~ qcu [-1cm-1] q: elementary charge c: concentration of charge carriers u: electrical mobility of charge carriers [cm2/Vsec] - charge carriers can be electrons or holes (!) Semiconductors conduction band Temperature dependence of the electrical conductivity ( [-1cm-1]) (Arrhenius type behaviour): band gap 0 Ea e kT Ea ln ln 0 kT valence band ln = f(1/T) linear Typical band gaps (eV): C(diamond) 5.47, Si 1.12, GaAs 1.42 Semiconductors Electrical conductivity as a function of the reciprocal absolute temperature for intrinsic silicon. T,K A semiconductor at room temperature usually has a much lower conductivity than a metallic conductor because only few electrons and/or holes can act as charge carriers , -1m-1 1/T,K-1 Ea ln ln 0 kT slope gives Ea A more detailed view of semiconductors: Intrinsic and extrinsic Semiconduction Intrinsic semiconduction corresponds to the mechanism just described: charge carriers are based on electrons excited from the valence into the conduction band (e.g. very pure silicon). conduction band Acceptor band valence band band gap Extrinsic semiconduction appears if the semiconductor is not a pure element but „doped“ by atoms of an element with either more or less electrons (e.g. Si doped by traces of phosphorous [ntype doping] or traces of boron [p-type doping] Intrinsic and extrinsic Semiconduction Intrinsic range slope: -EA2/kT Ea ln ln 0 kT Extrinsic range slope: -EA1/kT Semiconductors: Comparison of conductivity to metals and insulators Advanced structural chemistry Metal structures based on sphere packings ccp hcp bcc deviations from the ideal symmetry (e.g. , elongation or compression along the body diagonal (c-axis) in case of ccp (hcp)) Unusual structures of metals without direct relation to simple sphere packings 1. -Ga ~270 pm 247 pm strongly distorted („puckered“) hexagonal layers with short dGa-Ga = 247 pm between adjacent layers Special structures of metals without direct relation to simple sphere packings 2. -Mn complex network of CN 16 polyhedra (Friauf polyhedra) interpenetrated by additional atoms (similar -Mn) (notice: in metals and intermetallic compounds cordination numbers 12, 14, 16 and more are common) Further examples for „anomalies“ in the structural chemistry of metallic elements are: polymorphism of manganese (-Mn, -Mn, -Mn, -Mn) polymorphism of iron (-Fe, -Fe, -Fe, -Fe) (important for steel processing) the cubic primitive structure of -Po Possible reasons for structural anomalies of metallic elements: special interactions between magnetic atoms (e.g. Fe, Mn) deviations of atoms from spherical symmetry (special electronic interactions due to d-orbitals) The structures of some representative non metallic elements - Phosphorous: several polymorphs with complex stability relations (only black phosphorous thermodynamical stable under normal conditions) - Boron: electron deficiency, icosahedra or symmetrically related polyhedra - Carbon: polymorphism fulleren – graphite – diamond (thermodynamic relations and stability) - Sulfur: variety of molecular structures P(white): a = 18,51 Å, I-43m, Z = ?? 2,3 Å P (black): a=3.314 Å, b=10.478 Å, c=4.376 Å Z = 8, Cmca 2,2 Å 3,6 Å P (black): a=3.314 Å, b=10.478 Å, c=4.376 Å Z = 8, Cmca The purple modification of phosphorous system of tubes based on five membered rings; each P atom acts with three covalent bonds Boron The structural chemistry of boron is dominated by icosahedral B12 units or fragments of such icosahedra B12 icosahedra with the topology of a close packed layer in tetragonal boron-T50 (Z=50) B12-icosadron in the center of B60-cage with the topology of a fulleren (Boron-R105) Further advanced aspects of structures of non-metallic elements: thermodynamical relations between the polymorphs of carbon (p/T diagram of carbon) Structures of sulfur molecules S S S S S S S6 S S S S SS S S S 8 S S S S S S S S7 S10, S11, S12 ... Tin (Sn) -Sn (non-metal) d = 5,75 gcm-3 CN = 4 (281 pm) cubic diamond structure 13 oC -Sn (metal) d = 7,3 g cm-3 CN = 4+2 (302, 318 pm) tetragonal compressed diamond structure Specific aspects of the structural chemistry of alloys (intermetallic compounds) - classical alloys do not obey simple valence rules (8-N, noble gas configuration etc.): Zr4Al3, Cu5Zn8 ... - alloys in many cases are characterized by variable chemical compositions („homogeneity ranges“) with statistical atom distribution phase diagrams order-disorder-transitions - electronegativity differences are generally small ! however, if larger electronegativity differences occur they favor smaller homogeneity ranges (e.g. CsAu = Cs+Au-, „polar alloy“ with CsCl structure, NaTl = Na+Tl-, ZINTL-phase ) The most important groups of alloys 1. Hume-Rothery phases: (valence electron concentration favors certain structures, e.g. brass) 2. Laves-phases: relative atomic sizes favor certain structure types (classical examples: MgCu2 a cubic Laves-phase) 3. Frank-Kasper-Phases: 3D interpenetration of „Frank-Kasperpolyhedra“ favors complex alloy structures which correspond to a 3D-packing of (distorted) tetrahedra only. (e.g. Zr4Al3) 4. Zintl Phases: alloys of metals with moderate electronegativity differences show complete electron transfer (idealized) from the less electronegative to the more electronegative partner. NaTl = Na+Tl- (Tl- as „pseudo element“ of main group 4 and forms a diamond structure Brass: Cu – Zn (Hume-Rothery-Phases bcc fcc complex hcp Cu Zn The Laves phase MgCu2 Mg (160 pm) Cu (128 pm) r(Mg)/r(Cu) = 1.25 3/ 2 = 1.22 (ideal value) from R.C. Evans, Crystal Chemistry Frank Kasper-phases (topological close packings) (not to be confused with classical close packings) distorted icosahedra in the Frank-Kasper-phase Zr4Al3 icosahedra must be distorted due to symmetry reasons 5-fold symmetry is impossible in crystalline solids „Classical“ close packings: - Space filling: (<74%) - 3D systems of interpenetrating cuboctahedra and anti-cuboctahedra (both CN 12) - defines a system of tetrahedra and octahedra „Topological“ close packings (Frank-Kasper-phases): („tetrahedral close packings“) - Space filling: ??? (>74%) - 3D systems of interpenetrating Frank-Kasper-polyhedra (CN 12, 14, 15, 16) - defines a system of distorted tetrahedra Zintl phases Na - If one assumes a complete electron transfer from Na to Tl the latter one becomes a „pseudo element“ of group 14 and forms a diamond structure. - Typical for Zintl phases: the „pseudo element“ forms a structure (1D, 2D or 3D) which is found in real elements of the respective group of the periodic table. (LiIn, NaSi, Ba3Si4, NaP ...) Tl NaTl = Na+Tl-: Tl- forms a diamond structure (Tl- is a „pseudo element“ of group 14) - The Zintl model is an idealized assumption which accounts for the structural but in many cases not for the physical properties of the respective alloys The perovskite structure (CaTiO3) BaTiO3 BaTiO3 is a dielectric material that shows „Ferroelectricity“ Ti4+ O2- - Displacive phase transition below the ferroelectric Curie temperature (Tc = 393 K for BaTiO3) - Piezoelectricity (pressure induced), Pyroelectricity (temperature induced) are dielectric properties with a similar physical origin in other materials Ba2+ - In an electric field and below 393 K all cations (Ba2+, Ti4+) move in one and the anions (O2-) move in the opposite direction (see arrows); the structure becomes tetragonal (inversion center gets lost !) and shows a permanent electrical polarisation Different views of the perovskite structure The structural relation between the structure of the high temperature superconductor YBa2Cu3O7-x (x<1) and the perovskite structure oxygen atoms are omitted perovskite unit „BaCuO3“ „YCuO3“ „BaCuO3“ oxygen deficient !! The spinell structure: MgAl2O4 - In a ccp arrangement of O2- the Mg2+ occupiy 1/8 of the tetrahedral and Al3+ ½ of the octahedral holes - TA2+ O(B3+)2 O4: normal spinell - TB3+ O(A2+B3+) O4 : inverse spinell e.g. Fe3O4 (magnetite) TFe3+ O(Fe2+Fe3+) O4 Different view of the spinell structure The rutile structure (TIO2) Scanning Electron Microscopy Photo emission spectroscopy Impedance spectroscopy Moessbauer spectroscopy CamScan 44 Interaction of a high energy electron beam with material Primary Beam Backscattered Electrons Cathode Luminescence Auger Electrons Characteristic X-Ray Spectrum Secondary Electrons X-Ray Retardation Spectrum Specimen Absorbed Current Electron Gun Magnetic Lens Anode Scanning Coils Energy Dispersive X-Ray Detector Objectiv Lens Backscattered Electron Detector Wave length Dispersive X-Ray Detector Secondary Electron Detector Stage with Specimen to the Vacuum Pump Comparison of W-, LaB6-, and Field emission-cathods W LaB6 FE work function /eV 4,5 2,7 4,5 crossover /µm (important for high resolution images) 20-50 10-20 3-10 Tcathod /K 2700 <2000 300 1-3 25 105 gun Brightness /A/cm2 sr 105-106 107 109 vacuum /mbar 10-5 10-7 10-9 service life /h 40-100 1000 >2000 emission current density /A/cm2) Electron Gun (W-Cathode) Cathode Heating Current Tungsten Wehnelt Cylinder (500 V more negative than the Cathode,„Bias“) 20000 V Crossover (20-50 µm) Electron beam source Anode grounding The crossover is projected on the sample in a reduced size by the electronic-optical system (minimal diameter of the beam: ca. 5 nm) LaB6-Cathode (also: CeB6): Graphite LaB6 single crystal - indirect heating (because of the low conductivity of LaB6) - lower work function than the W-cathode ( higher brightness) - demageable by ionic shooting ( high vacuum necessary) - expensive! Field Emission-Cathode - W – cathode with a fine apex - two anodes: 1. one to bring up the work function 2. one for the acceleration - high brightness - high vacuum necessary Interaction volume of the electron beam (pear-like) primary beam Auger-electrons (up to 0,001m)* Secondary Electrons (up to 0,01m)* Backscattered Electrons (up to 0,1m)* X-Ray (1-5m)* *information depths Dependence of the information depth on the acceleration voltage and material (simulations) Fe (10 kV) Au (20 kV) Fe (20 kV) Al (20 kV) 1m Fe (30 kV) Secondary Electrons: - inelastic scattered PE (Primary Electrons) - energy loss by interaction with valence electrons or with the atomic nucleus - Energy: < 50 eV - maximal emission depth: 5-50 nm - leads to high resolution images Backscattered Electrons: - elastic and inelastic scattered PE - Energy: 50 eV – Energy of the PE (e.g. 20 keV) - maximal emission depth: 0.1 - 6µm (dependent on the specimen) - Intensity depends on the average atomic number of the material ( material contrast images) - high interaction volume ( low resolution images) Cu-wire imbedded in solder SE-Image (high resolution) BE-Image (high Z-contrast) Cu Cu Cu Cu ZPb > ZSn > Z Cu Backscattered electron images are less sensitive on charging: BE-image SE-image reason: the average energy of the backscattered electrons is higher Auger Electrons: - Energy characteristic for the Element Auger Electron Spectroscopy (AES) 3. EAuger-Electron = E1-E2-E3 E3 E2 1. 2. E1 Comparison Auger / X-Ray Rate of yield Auger X-Ray Z small X-Ray yield for light elements (B, C, N, O, F) Cathode Luminescence - not for metals - visible and UV radiation - special detectors necessary PE h conduction band valence band X-Ray Retardation Spectrum Intensity Energy /keV - primary electrons are retarded by the electron clouds of the atoms - Emax of the X-Ray‘s: e × Uaccelerating voltage Characteristic X-Ray Spectrum (without fine structure) L O L N K M L K K L M M Energy range of the main series as a function of the atomic number: energy K K L L M Mosley‘s law: 1/ = K (Z-1)2 atomic number (K: constant, Z = atomic number) Typical (characteristic) X-Ray spectrum (EDX) intensity Spectrum of Ag6GeS4Cl2 energy /keV - sulfur (Z = 16) and chlorine (Z = 17) easily distinguishable (in contrast to X-Ray diffraction) Electron Detectors 1. Secondary Electron (SE-) Detector Szintillator-Photomultiplier-Detector (Everhart-Thornley-Detector) Amplifier PhotoMultiplier Light conductor VideoSignal Scintillator Metal net (+ 400V) SE-Detektor: SE-D SE-D Principle of an EDX-Detector P- i- n-conducting Si (Li) + 3,8 eV h + Si Si+ + e- - e.g. Mn K: 5894 eV X-Ray - + high voltage 5894/3.8 = 1550 electron hole pairs WDX-Spectrometer: Scanning of a -range with one monochromator crystal Rowland circle Detector Axis of Crystal movement Monochromator Crystal Specimen (X-Ray source) WDX Detector 4 Monochromator crystals Proportional counter Comparison EDX - WDX Irel Ga L Ga L EDX Ge L Ge L Irel Ga L WDX Ga L Ge L Ge L (compound: GeGa4S4) 1.1 1.2 energy /keV Detector‘s for WDX: two proportional counter switched in series 1: FPC, Flow proportional counter (for low energy X-Ray‘s) - the counting gas (Ar / CH4) flows through the counter (very thin polypropylen window, not leak-free for the counting gas) 2. SPC, Sealed proportional counter (for high energy X-Ray‘s) - counting gas: Xenon / CO2 Comparison of EDX and WDX EDX WDX spectral resolution 110-140 eV 10 eV specimen current 10-10 A 10-7 A analysis time 1-2 min 30-100 min simultaneous sequential spectrum develops Sensitivity of EDX and WDX atomic percent EDX Be - window WDX atomic number Applications: I) High resolution Images II) Qualitative and quantitative analysis Principle of the image formation synchronised scanning coils Beam Amplifier Television image Signal-Detector (SE or BE) Specimen Twinning of Crystals PbS Na2Zn2(SeO3)3 3H2O Morphology of crystals CuGa3S5 K2In12Se19 SnIn4S4 CsIn3S5 Quality control of small technical objects Compact disc Cantilever of an AFM Large area mapping (X-Ray-images) SE-Image Cu-Kmapping Ni-Kmapping Zn-Kmapping 256x256 pixel, moving of the specimen holder typical sample holder equipment typical preparation of small crystals conducting tabs (adhesive plastic with graphite) Special preparation for insulating material: -metallisation with gold (sputtering process) better for high resolution images -carbon deposition (evaporation process) better for quantitative analysis Impedance spectroscopy (basic aspects) Purpose: Exploring the electrical behavior of a microcrystalline solid sample as function of an alternating current (ac) with a variable frequency. three basically different regions for the exchange interactions between current and sample: a) inside the grains („bulk“) b) at grain boundaries c) surface of the electrodes the elecrcical behavior is simulated by a suitable combination of RC circuits: R = resistivity, C = capacity Impedance spectroscopy (electrical quantities) The direct current (dc) resistance (R) is defined as: U R I U: applied voltage (V), I: curent (A) R: resistance () The alternative quantity for an alternating current (ac) is the impedance (Z*) which, however, is a more complex quantity than R because there is a phase shift between U and I in general. ac-voltage: U(t) = U0sint I(t) = I0 sin(t+) (t: time, = 2 with : frequency) (: phase angle) Impedance spectroscopy (U(t), I(t)) If only capacities are present the phase shift between voltage and current amounts to /2 = 900. Photoemission spectroscopy UV or X-Ray e- UPS spectrum UHV DOS solid Band structure The energy spectrum of the emitted electrons is analyzed energy Moessbauer Spectroscopy the nucleus of the specific isotope of an atom embdedded in a solid (e.g. 57Fe) is excited by -rays emitted by an instable isotope of a neighbor element (e.g. 57Co) source: 57Co e.g. (tunable) absorber: e.g. 57Fe frequently applied for 57Fe, 119Sn, 127J ... chemical surrounding (symmetry, coordination number, oxidation state, magnetism) of atoms with these nuclei in a solid can be probed in a highly sensitive way (~10-8 eV)