Encycopedia of Materials Characterization (surfaces_ interfaces_ thin films) - C. Brundle_ et al._ (BH_ 1992) WW Part 8 pptx

A major thrust in the hture will be the use of contactless modulation methods like PR or RDS (together with scanning ellipsometry) for the in-situ monitoring and control of growth and processing, including real-time measurements These methods can be used not only during actual growth at elevated temperatures but also for in-situ post growth or processing at room temperature before the sample is removed from the chamber Such procedures should improve a material’s quality and specifications, and also should serve to reduce the turn-around time for adjusting growth or processing parameters The success of PR as a contactless screening tool for an industrial process, i.e., heterojunction bipolar transistor structures, certainly will lead to more work on real device configurations There also will be improvements in instrumentation and software to decrease data acquisition time Changes can be made to improve lateral spatial resolution For example, if the probe monochromator is replaced by a tunable dye laser spatial resolutions down to about 10 pm can be achieved Related Articles in the Encyclopedia RHEED, VASE References Semiconductorsand Semimetah (R K Willardson and A C Beer, eds.) Academic, New York, 1972, Volume z Proceedings of the First International Conference on Modulation Spectroscopy Su$ Sci 37, 1973 D E Aspnes In: Handbook on Semiconductors.(T S MOSS, North ed.) Holland, New York, 1980, Volume 2, p 109 E H Pollak h c SOC Photo-OpticalImtz Eng 276,142, 1981 E H Pollak and 0.J Glembocki h c SOC Photo-OpticalImtz Eng 946, 2, 1988 D E Aspnes, R Bhat, E Coles, L T Florez, J I? Harbison, M K Kelley, V G Keramidas, M A Koza, and A A Studna Proc SOC Photo-Optical Imtz Eng 1037,2,1988 D E Aspnes, J I? Harbison, A A Studna, and L T F1orez.J Vac Sci Tecbnol A6, 1327, 1988 E H Pollak and H Shen J Crystal Growth 98,53,1989 R Tober, J Pamulapari, R K Bhattacharya, and J E Oh J Ekmonic Mater 18,379, 1989 B Drevillon Proc SOC Photo-OpticalInstz Eng 1186,110, 1989 10 7.2 Modulation Spectroscopy 399 11 E H Pollak and H Shen./ E&ctronicMat 19,399,1990 i z Proceedings of the International Conference on Modulation Spectroscopy Proc SOC Photo-OpticalIns& Eng 1286,1990 13 14 M H Herman h o c SOC Photo-OpticalInstr Eng 1286,39,1990 R E.Hummel, W Xi, and D R Hagmann./ E&cmchm SOC 137, 3583,1990 400 VIS!BLE/UV EMISSION, REFLECTION, Chapter 73 VASE Variable Angle Spectroscopic Ellipsometry J O H N A WOOLLAM A N D P A U L G S N Y D E R Contents Introduction Basic Principles Applications Conclusions Introduction The technique of ellipsometry was introduced in the 0 ~ ~ until computers but became available, it was painllly slow to perform.' Rapid advances in small computer technology have made ellipsometric data acquisition rapid and accurate t Most impormnt, k personal computers make possible quick and convenient analysis of data from complex material structures Early work in ellipsometry focused on improving the technique, whereas attention now emphasizes applications to materials analysis New uses continue to be found; however, ellipsometry traditionally has been used to determine frlm thicknesses (in the range 1-1000 nm), as well as optical constants.14 Common systems are oxide and nitride films on silicon wafers, dielectric films deposited on optical suhces, and multilayer semiconductor structures In ellipsometry a collimated polarized light beam is directed at the material under study, and the polarization state of the reflected light is determined using a second polarizer To maximize sensitivity and accuracy, the angle that the light makes to the sample normal (the angle of incidence) and the wavelength are controlled.u The geometry of a typical ellipsometry set up is shown in Figure Ellipsometry is a very powerfd, simple, and totally nondestructivetechnique for determining optical constants, film thicknesses in multilayered systems, surface and 73 VASE 401 I I Figure Planar structure anumedfor ellipsometric analysis: 4is the complex index of refraction for the ambient medium; n, is the complex index for the substrate medium; 0, is the value of the angles of incidence and reflection, which define the plane of incidence interfacial roughness, and material microstructures (An electron microscope may alter surkces, as may Rutherford backscattering.) In contrast to a large class of surface techniques such as ESCA and AUGER, no vacuum chamber is necessary in ellipsometry Measurements can be made in vacuum, air, or hostile environments like acids The ability to study surfices at the interface with liquids is a distinct advantage for many disciplines, including surface chemistry, biology and medicine, and corrosion engineering Ellipsometry can be sensitive to layers of matter only one atom thick For example, oxidation of freshly cleaved single-crystal graphite can be monitored from the first monolayer and up The best thicknesses for the ellipsometric study of thin films are between about nm and 1000 nm Although the spectra become complicated, films thicker than even pm can be studied Flat planar materials are optimum, but surface and interfacial roughness can be quantitatively determined if the roughness scale is smaller than about 100 nm Thus ellipsometry is ideal for the investigation of interhcial surfaces in optical coatings and semiconductor structures?’ In some applications lateral homogeneity of a sample over large areas needs to be determined, and systems with stepper driven sample positioners have been built Use of focused ellipsometer beams is then highly desirable As normally practiced, the lateral resolution of ellipsometry is on the order of millimeters However, the light beam can be focused to 100 pn if the angle of incidence variation is not criti d For smaller focusing the beam contains components having a range of angles of incidence that may alter the validity of the data analysis Depth resolution depends on the (spectrallydependent) optical absorption coefficient of the material Near-surface analysis (first 50 nm) frequently can be per43 - 402 VISIBLE/UV EMISSION, REFLECTION, Chapter b x, y components E Propagation direction Figure (a) Representation of a linearly polarized beam in its x- and p or (p and s-) orthogonal component vectors The projection plane is perpendicularto the propagation direction; (b) lows of projection of electric vector of light wave on the projection plane for elliptically polarized light-a and b are the major and minor axes of the ellipse, respectively, and a is the azimuthal angle relative to the x-axis formed using short wavelength light (2300 nm) where absorption is strongest, and infiared radiation probes deeply (many pm) into many materials, including semiconductors Basic Principles Light Waves and Polarization Light is an electromagnetic wave with a wavelength ranging from 350 nm (blue) to 750 nm (red) for visible radiation.8 These waves have associated electric (E) and magnetic ( H ) components that are related mathematically to each other, and thus the Ecomponent is normally treated alone Figure 2a shows the electric field associated with linearly polarized light as it propagates in space and time, separated into its x- and y-vector components In the figure the x- and ycomponents are exactly in phase with each other thus the electric vector oscillates in one plane, and a projection onto a plane perpendicular to the beam propagation direction traces out a straight line, as shown in Figure 2a When the vector components are nor in phase w t each other, the projection of ih the tip of the electric vector onto a plane perpendicular to the beam propagation direction traces out an ellipse, as shown in Figure 2b A complete description of the polarization state includes:' The azimuthal angle of the electric field vector along the major axis of the ellipse (recall the angle a in Figure 2b) relative to a plane of reference 73 VASE 403 The ellipticity, which is defined by e = b / a The handedness (righthanded rotation of the electric vector describes clockwise rotation when looking into the beam) The amplitude, which is defined by A = (a2 62)45 + The absolute phase of the vector components of the electric field In ellipsometry only quantities and (and sometimes 3) are determined The absolute intensity or phase of the light doesn't need to be measured, which simplifies the instrumentation enormously The handedness information is normally not critical All electromagnetic phenomena are governed by Maxwell's equations, and one of the consequences is that certain mathematical relationships can be determined when light encounters boundaries between media Three important conclusions that result for ellipsometry are: '3 The angle of incidence equals the angle of reflectance 80 (see Figure 1) z Snell's Law holds: nl sin el = sin 0, (Snell's Law), where nl and are the complex indexes of refraction in media and media 0, and the angles , and00 are shown in Figure The Fresnel reflection coefficients are: ~ d - nlmSeOnocosel p nl coseo noCOSel + r =-P Edp where refers to the light vector component perpendicular to the plane of incidence, p refers to the component parallel to the plane of incidence, and rand i refer to reflected and incoming light The plane of incidence is defined by the incoming and outgoing beams and the normal to the sample The complex indices of refraction for media and are given by and nl The relations r, and rp are the complex Fresnel reflection coefficients Their ratio is measured in ellipsometry: r P A p = - = (tanY)e i Since p is a complex number, it may be expressed in terms of the amplitude factor tan Y, and the phase factor exp jA or, more commonly, in terms of just Y and A Thus measurements of Y and A are related to the properties of matter via Fresnel coefficients derived from the boundary conditions of electromagnetic theory ', 404 VISIBLE/UV EMISSION, REFLECTION, Chapter There are several techniques for measuring Y and A, and a common one is discussed below Equations l a and 1b are for a simple two-phase system such as the air-bulk solid intehce R a materials aren't so simple They have natural oxides and surface el roughness, and consist of deposited or grown multilayered structures in many cases In these cases each layer and interface can be represented by a x matrix (for isotropic materials), and the overall reflection properties can be calculated by matrix multiplication.' The resulting algebraic equations are coo complex to invert, and a major consequence is that regression analysis must be used to determine the system's physical parameters.'' 2, In a regression analysis Y t and A t are calculated from an assumed model for the structure using the Fresnel equations, where Y and A in Equation are now indexed by c, to indicate that they are calculated, and by i, for each combination of wavelength and angle of incidence The unknown parameters of the model, such as film thicknesses, optical constants, or constituent material fractions, are varied until a best fit between the measured Yi" and Aim and the calculated Y t and A i is found, where m signifies a quantity that is measured A mathematical function called the mean squared error (MSE) is used as a measure of the goodness of the fit: 53 MSE = N -C (Y;-Y?) N +( A ~ ~ - A ~ ~ ) (3) i= where N is the number of wavelength and angle of incidence combinations used The MSE is low if the user's guess at the physical model for the system was correct and if starting parameters were reasonably close to correct values The model-dependent aspect of ellipsometric analysis makes it a difficult technique Several different models fit to one set of data may produce equivalently low MSEs The user must integrate and evaluate all available information about the sample to develop a physically realistic model Another problem in applying ellipsometry is determining when the parameters of the model are mathematically correlated; for example, a thicker fdm but lower index of refraction might give the same MSE as some other combinations of index and thickness That is, the answer is not always unique Access to the correlation matrix generated during the regression analysis is thus important's to determine which, and to what degree, variables are correlated It is common for the user of an ellipsometer-mistakenly make five wrong (correlated) to measurements of an index of refraction and film thickness at, say, 632.8 nm and then to average these meaningless numbers In reality all five measurements gave nonunique values, and averaging is not a valid procedure-the average of five bad numbers does not yield a correct number! The solution to the correlation problem 73 VASE 405 tr.f4 Toughness ta,fa t4,fZ tl Substrate Figure Common structure assumed for ellipsometric data analysis: t and lj are the l thicknessas of the two deposited films, for example; and t, are interfacial and surface roughness regions; is the fraction of film t mixed with film lj in l an effective medium theory analysis of roughness-film f3 could have void (with fraction 1-41 dispersed throughout; and f, is the fraction o t, mixed f with the ambient medium to simulate surface roughness is to make many measurements at optimum wavelength and angle combinations, and to keep the assumed model simple yet realistic Even then, it is sometimes inherently not possible to avoid correlation In this case especiallyit is important to know the degree of correlation Predictive modeling can be performed prior to making any measurements to determine the optimum wavelength and angle combinations to use, and to determine when there are likely to be correlated variables and thus nonunique an~wers.~’ A typical structure capable of being analyzed is shown in Figure 3, consisting of a substrate, two films (thicknesses tl and t3), two roughness regions (one is an interfacial region of thickness %, and the other is a surface region of thickness t4) One of the films t l or t3 may consist of microscopic (less than 100 nm size) mixtures of two materials, such as SiO, and Si3N4 The volume ratios of these two constituents can be determined by ellipsometry using effective medium theory lo This theory solves the electromagneticequations for mixtures of constituent materials using simplifying approximations, resulting in the ability of the user to determine the fraction of any particular species in a mixed material Likewise the roughness layers are modeled as mixtures of the neighboring media (air with medium for the surface roughness, and medium with medium for interfacial roughness, as seen in Figure 3) The example in Figure is as complex as is usually possible to analyze There are seven unknowns, if no indices of refraction are being solved for in the regression analysis If correlation is a problem, then a less complex model must be assumed For example, the assumption thatf2 andf4 are each fixed at a value of 0.5 might reduce correlation The five remaining unknowns in the regression analysis would then be tl,%, t3, t4, andff In practice one first assumes the simplest possible model, then makes it more complex until correlation sets in, or until the mean squared error fails to decrease significantly 406 VISIBLE/UV EMISSION, REFLECTION, Chapter Polarization Measurement Manual null ellipsomerry is accurate but infrequently done, due to the length of time needed to acquire sufficient data for any meaningful materids analysis Automated null ellipsometers are used, for example, in the infrared, but are still slow Numerous versions of kt automated ellipsometers have been built 1-3 Examples are: Polarization modulation z Rotating analyzer Rotating polarizer The most common versions are and 3, and the rotating analyzer system will be briefly described here." Such a system consists of a light source, monochromator, collimating optics, and polarizer preceding the sample of Figure 1, and a rotating polarizer (called the analyzer) and detector following the sample The intensity of the light measured at the detector oscillates sinusoidally according to the relation I = + acos2d+ PsinZA where a and p are the Fourier coefficients, and A is the azimuthal angle between the analyzer "fast axis" and the plane of incidence There is a direct mathematical relationship between the Fourier coefficients and the Y and A ellipsometric parameters The actual experiment involves recording the relative light intensity versus A in a computer The coefficients 01 and p, and thus Y and A, can then be determined By changing the angle of incidence and wavelength, the user can determine N sets of Y j and Ai values for the regression analysis used to derive the unknown physical properties of the sample The polarizer and analyzer azimuthal angles relative to the plane of incidence must be calibrated A procedure for doing this is based on the minimum of signal that is observed when the fist axes of two polarizers are perpendicular to each other For details the reader can consult the literature.l1 Applications In this section we will give some representative examples Figure shows the regression procedure for tan Y for the glass/Ti02/Ag/Ti02 system The unknowns of the fit were the three thicknesses: TiO2, Ag,and the top TiO2 Initial guesses at the thicknesses were reasonable but not exact The final thicknesses were 33.3 nm, 11.3 nm, and 26.9 nm, and the fits between measured 'Pi" and Aim and calculated (from Fresnel equations) Y/ and A/ were excellent This means that the assumed optical constants and structure for the material were reasonable Because Y and A can be calculated for any structure (no matter how complex, as long as planar parallel interhces are present), then the user can predictive modeling Figure shows the expected A versus wavelength and angle of incidence for a 73 VASE 407 Fundamental information from vibrational spectra is important for understanding a wide range of chemical and physical properties of surfaces, e.g., chemical reactivity and forces involved in the atomic rearrangement (relaxation and reconstruction) of solid surfaces Practical applicationsof H E E L S indude studies OE Functional groups on polymer and polymer film s& u Phonon modes of metals and films, semiconductors, and insulators Concentrations of free charge carriers in semiconductors Adsorbed species (and even underlayer atoms) on singlecrystalsurfices Kinetics ofsurfice processes when used in a time-resolved mode Of these, the most extensive use is to identify adsorbed molecules and molecular intermediates on metal single-crystal surfaces On these well-defined surfaces, a wealth of information can be gained about adlayers, including the nature ofthe surface chemical bond, molecular structural determination and geometrical orientation, evidence for s&-site specificity, and lateral (adsorbateadsorbate) interactions Adsorption and reaction processes in model studies relevant to heterogeneous catalysis, materials science, electrochemistry, and microelectronics device failure and fibrication have been studied by this technique The first vibrational spectrum of adsorbed molecules obtained by inelastic scattering of low energy electrons was obtained by Propst and Piper in 1967 Over the next ten years, Ibach in Jiilich achieved much higher resolution and, alongwith several other research groups, developed and used the new technique (HREELS) to study a wide range of surfice vibrations, including polyatomic molecules adsorbed on surfices The enthusiasm existed mainly because HREELS could be used to study adsorbates on low su&e-area, opaque, metal single-crystal samples, something that could not be done with infrared and Raman spectroscopy It is now wellestablished as an important vibrational spectroscopyat surfaces, and its utility as an analytical tool with extreme surface sensitivity is rapidly being extended Basic Principles Electron Scattering At sufficientlyhigh resolution, quasi-elastically scattered electrons have an inelastic scattering distribution from exciting surfice vibrational modes such as surfice phonons and adsorbate vibrations (See, for example, Figure 1, the case of CO adsorbed on a Rh surface.) These modes have excitation energies below 0.5 e V (4000cm-l) The basis of the kinematic (single scattering) description of electron scattering from surfices is conservation of energy and momentum parallel to the surface These conservation laws define the scattering possibilities and, along with the vibrational energies, determinethe peak energies in HREELS The mechanisms 83 HREELS 443 I I I I I I I 2E,=%-AE - Rhodium Figure 500 10 00 1500 ENERGY LOSS (on-') ( Schematic of electron energy-loss scattering process for electrons of energy Ei striking a Rh single-crystal surface with adsorbed CO molecules present The actual energy-loss spectrum, due to excitation of CO vibrations, is shown also of electron inelastic scattering determine the scattering cross sections (probabilities), and therefore the intensities of vibrational energy loss peaks Two different electron inelastic scattering mechanisms are most important for explaining the intensities and angular distribution of peaks observed in HREELS: dipole and impact scattering Dipole scattering is due to the long-range part of the electrostaticinteraction between an incoming electron and the dipolar electric field of the vibrating (oscillating) atoms at the surface Energy exchange takes place at a long distance (50A) from the surface This leads to small angle scattering and so the scattered intensity is strongly peaked near the specular direction, with a half-angle A0 0.1-1 ' The dominance of this mechanism for detection at the specular direc tion leads to the dipole selection rule in HREELS, that is identical to that associated with reflection infrared spectroscopy: Only vibrational modes which produce a - 444 VIBRATIONAL SPECTROSCOPIES Chapter dynamic (oscillating) dipole moment perpendicular to the metal surface are dipoleactive, i.e., will produce energy loss peaks on-specular The origin of this selection rule is the screening of a charge on a conducting surface by an image charge induced in the free electrons Dynamic dipoles parallel to the surface generate no long-range dipole field, while dynamic dipoles perpendicular to the surface generate a longrange dipole field that is enhanced by a factor of At large scattering angles away from the specular direction, one enters the impact scattering regime This mechanism is due to a short range electrostatic interaction between an incoming electron and the ion cores of the adsorbate and substrate lattice Impact scattering is usually several orders of magnitude weaker than dipole scattering at the specular direction in the low-energy regime (below 10 ev) Broad angular scattering distributions having intensities proportional to vibrational amplitudes are characteristic Selection rules are much less restrictive and are based on adsorbate site symmetry and vibrational mode polarization in relation to the scattering plane of incidence The impact scattering regime extends to several hundred eV, and the theory of these interactions must include multiple scattering Measurements of inelastic scattering cross sections in this regime have a great potential for structural analysis, but have been made experimentally only recently A type of molecular resonance scattering can also occur from the formation of short-lived negative ions due to electron capture by molecules on su&es While this is frequently observed for molecules in the gas phase, it is not so important for chemisorbed molecules on metal surfaces because of extremely rapid quenching (electron transfer to the substrate) of the negative ion Observations have been made for this scattering mechanism in several chemisorbed systems and in physisorbed layers, with the effects usually observed as small deviations of the cross section for inelastic scattering from that predicted from dipole scattering theory While the underlying mechanisms of HREELS are pretty well understood, many important details relating to selection rules and scattering cross sections remain unknown Vibrationsat Surfaces Vibrations give molecular information by identifylng which atoms are chemically bonded together The frequency of a vibrational mode is related to the bond force constant and reduced mass of the vibrating atoms Reasonable correlations exist between the number of bonds, bond energy, and force constant for vibrations within molecular species at surfaces and for adsorbat-ubstrate vibrations It would be extremely useful if one could determine chemical bond strengths from vibrational spectra, but the accurate determination of this quantity is quite ambiguous Frequency shifts occur as the concentration is changed in adsorbed layers as a result of local bonding variations due to changing sites or bond energies, dipole dipole coupling leading to collective vibrations of the overlayer, or other factors 83 HREELS 445 The intensity of a vibrational mode in H E E L S on-specular is given by the ratio of the inelastic to elastic intensities S=~ ~ C E ~ ~ ~ N P ~ F ( ~ , e) arccose (1) where Sis the dipole scatteringcross section, Eiis the primary beam energy, Nis the number of surface oscillators, Pp is the perpendicular component of the dynamic dipole moment, and f l a y e) is an instrument and geometry factor This intensity (sensitivity)can be optimized by considering the detected inelastic current, I: (2) I = IoR(Ei)KS where is the incident beam current, R(EJ is the metal surfice reflectivity (typi0 cally 1-10%) for the primary beam energy E;, and Kis the analyz,er transmission constant The hnction Sis a smoothly decreasing function of energy and is propor; ' tional to E for a layer of adsorbed dipoles, but R exhibits significant variation as a function of E+Operationally, the incident energy is tuned to give an intense elastic peak at a low energy (< 10 ev) where Sis large The sensitivity of H E E L S is largest for molecules, materials, or particular vibrations, that have large S Normalizingfor the other variables, such as concentra tion, leads to a general rule that vibrational modes that are observed as strong bands in infrared spectroscopyof gases or of condensed phases will give rise to intense loss peaks in HEELS For example, carbonyl (C = 0)stretching modes have larger intensities than hydrocarbon C-H stretching modes in IR of bulk phases and also in HREELS on surfaces Orientational effects will affect this sensitivity in accord with the dipole selection rule Quantitative analysis and determination of the concentration of surface species requires measuring peak intensities and accounting for vibrational cross sections This is a difficult task Careful analysis of Kand analyzer angular characteristics is required when determining S(EJ To obtain information on the charges on atoms at the surface, one could perform a calculation of dynamic dipole moments (or effective charges) from knowledge of S, in principle In practice, one usually assumes a point dipole model and gas phase polarizabilitiesfor surface species, and this has lead to anomalouslylow values in the case of adsorbed molecules on metals Chemical bonding effects greatly enhance the electronic polarizability of chemisorbed molecules and enhance the dielectric screening by the adlayer, reducing the predicted vibrational (inelastic peak) intensity The width and shape of the energy loss peaks in HREELS are usually completely determined by the relatively poor instrumental resolution This means that no information can be obtained from HREELS about such interesting chemical physics questions as vibrational energy transfer, since the influence of the time scale and mechanism of vibrational excitations at surfaces on the lifetimes, and therefore the line widths and shapes, is swamped (Adsorbates on surfaces have intrinsic vibra446 VIBRATIONAL SPECTROSCOPIES Chapter - SHIELDING MONO? HROYATOR ANALYZER ELECTRON MULTIPLIER I I L L Figure Schematic of a 127" high-resolution electron energy-loss spectrometer mounted on an 8-in flange for studies of vibrations at surfaces tional line widths of typically c - (0.6 mev) and Lorentzian line shapes.) A m' practical matter is that this poor resolution is insufficient to resolve closely overlapping vibrational frequencies Instrumentation In HREELS, a monoenergetic beam of low energy electrons is focused onto the sample surface and the scattered electrons are analyzed with high resolution of the scatteringenergy (e 10 meV or 80 cm-') and angle (A0 = 2-5") This is achieved by using electrostatic spectrometers, typically with 127" cylindrical dispersive elements Hemispherical and cylindrical mirror analyzers have been used also Some typical analyzer parameters are 25-mm mean radius, 0.1-mm slit width, and 0.5-eV pass energies Refinements also include the addition of tandem cylindrical sectors to the monochromator and analyzer A number of commercial versions of spectrometers are capable of routine and dependable operation A simple spectrometerthat we have used successfully is shown in Figure Electrons from an electron microscope hairpin tungsten filament are focused with a n Einzel lens onto the monochromator entrance slit, pass through the monochromator and exit slit, and are focused on the sample's surface by additional electrostatic 8.3 HREELS 447 lenses (in this case a double plate lens system) The incident beam energy is usually below 10 eV, where dipole scattering cross sections are strong, and the beam current to the sample is typically 0.1-1 nA Electrons that are reflected from the sample's surfice are focused on the analyzer entrance slit and energy analyzed to produce an electron energy loss (vibrational) spectrum An electron multiplier and pulse counting electronics are used due to the small signals Count rates in HREELS are typically 104-106 counts/sec fbr elastically scattered electrons and 10-103 counts/sec for inelastically scattered electrons Scan times typically range from about five minutes to several hours The exact incident and scattering angles (+60"from the surfgcenormal) are not critical, but a specular scattering geometry must be attainable It is also very useM to be able to observe a nonspecular scattering angle either by rotating the crystal about an axis perpendicular to the scatteringplane or by rotating one of the analyzers in the scattering plane Magnetic shielding must surround the spectrometer because the magnetic field of the earth and any nearby ion pumps will distort the trajectories of the electrons within the spectrometer, because of their small kinetic energies The spectrometer is mounted in a ultrahigh-vacuum chamber and analysis must be carried out at pressures below lo4 torr This requirement exists because of the sensitivity of the electron filament and slits to reactions with background gases Higher pressure gas phase molecules also will cause inelastic scattering that obscures the surface spectra The applicability of the HREELS technique can be greatly extended by combining it with a high-pressure reaction chamber and sample transfer mechanism We have previously used this type of system to study hydrogen transfer in adsorbed hydrocarbon monolayers at atmosphericpressure Interpretationof Vibrational Spectra In the following discussion, heavy emphasis is made of examples from studies of adsorbed layers on metal single-crystal samples These illustrate the power of the HREELS technique and represent the main use of HREELS historically Certainly HREELS has been used outside of the single-crystal world, and mention is made concerning its use on "practical" materials This latter use of HREELS represents a true frontier IdentXcation o Adsorbed Species f Determination of surfgce functional groups, e.g., -OH, -C C-, and >C= 0,and identification of adsorbed molecules comes principally from comparison with vibrational spectra (infrared and Raman) of known molecules and compounds Quick qualitative analysis is possible, e.g., stretching modes involving H appear for v(C-H) at 3000 cm-' and for v(0-H) at 3400 a - In addition, the vibrational n' energy indicates the chemical state of the atoms involved, e.g., v(C=C) 1500a n' and v(C=0) 1800 cm-' Further details concerning the structure of adsorbates - + 448 VIBRATIONAL SPECTROSCOPIES Chapter Mode assignment CH@+@3, CH3C-Rh (111) v,(CH~)/V,(CD~) 2930 (m)/2192(w) e 2920 (vw)/2178 ( w e v) vS(CH3)/vs(CD3) 2888 (m)/- al 2880 (w)/2065(vw) a] 1420 (m)/1031 (w) e 1420 ( w / e v)- Gs(CH3)/Gs(CD,) 1356 (m)/1002 ( w a1 v) 1337 (s)/988 (w) al VKC) 1163 (m)/1882(ms) a1 1121 (m)/1145(m)al p(CH3)/P (CD3) 1004 (s)/828(s) e 972 (w)/769 (vw) e vS(M-C) 401 (mV393(m)a1 435 (w)/419 (w) a1 G,(CH3)/Sa(CD3) * Inremiria of the spectral bands are given in parenthesesfollowing the band frequencies using the following abbreviations: vs = very strong, s = strong, ms = medium mong, m = medium, w = weak, and vw = very w& Symmetry assignments h each of rhe vibrational r modes are also indicated after &e band ikquencies Table Comparison of the vibrational frequenciee ( n ' of the ethylidyne surface a-) species formed on Rh (111)with those of the ethylidyne cluster compound comes from comparison to vibrational spectra of ligands in metal cluster compounds whose X-ray crystal structure is known Isotopic substitution is extremely important in confirming vibrational assignments Only H / D substitution can be carried out, due to the low resolution, but this is useful for an enormous range of adsorbed molecules, including hydrocarbons Substituting D for H causes an isotofor those modes wirh largeamplitude H motion pic shift of As an example, Figure 3a shows the H E E L S spectra after the adsorption of ethylene (H2C = CH2) on Rh(ll1) at 310 IC5 Comparison with gas phase ethylene inhred spectra shows that large changes occurred during adsorption, e.g., v(C-H) = 2880 cm-', indicative of aliphatic C-GH bonds, rather than the o expected v(C-H) * 3000 cm-1 f r olefinic C=C-H bonds The complete agreement with frequencies, intensities, and H / D shifts observed in IR spectra (and normal mode analysis) of an organometallic complex, CH~CCO(CO),,~ allowed for the detailed assignment of the loss peaks to vibrational modes of a surface ethylidyne (CCH3) species, as shown in Figure 3b and Table The lack of a welldefined s ular direction for polycrystalline metal samples decreases the signal levels by 10y103, and restricts the symmetry information on adsorbates, but many studies using these substrates have proven use!il for identifying adsorbates Charging, beam broadening, and the high probability for excitation of phonon modes of the substrate relative to modes of the adsorbate make it more difficult to carry out adsorption studies on nonmetallic materials But, this has been done previously fbr a number o metal oxides and compounds, and also semiconf h 83 HREELS 449 a XI183 (CH3 1387 R h (I1 I1 0K h 1ooo m o 3000 ENERGY LOSS (cm-') RhOll) + ethylidyne Figure 450 (a) Specular spectra in HREELS obtsinedfollowing exposure of ethylene or C2D4)on Rh(l11) a t 310 K to form the ethylidyne (CCH,) surface speck.* Ib) The atomic structure (bond distances and angles) of ethyfidyne as deaermined by LEED crystallography VIBRATIONAL SPECTROSCOPIES Chapter ductors like Si, InP, and diamond Dubois, Hansma, and Somorjai fabricated model supported metal catalysts by evaporating rhodium onto an oxidized aluminum substrate and studied CO adsorption by using HREELS We also have used HREELS to characterize lubricating carbon films on small samples of actual magnetic recording disk heads Determination of Adsorption Geometry The symmetry of an adsorbed molecule and its orientation relative to the surface plane can be established using group theory and the dipole selection rule for specular scattering The angular variations of loss intensities determines the number and frequenciesof the dipole active modes Only those modes that belong to the totally symmetric representations of the point group which describes the symmetry of the adsorbed complex will be observed as fimdamentals on-specular The symmetry of the adsorbate-surface complex is then determined by comparing the intensity, number, and frequency of dipole-active modes with the correlation table of the point group of the gas phase molecule In the previous example of adsorbed ethylidyne, observation of an intense symmetric C-H bending (6,CH3) loss peak and weak antisymmetric C-H bending (6,CH3) loss peak establishes C symmetry , for the surface complex The adsorption of nitrogen dioxide ( N O on metal surfaces is a beautifid example of linkage isomerism, as illustrated in Figure 4, that was discovered by using a HREELS.' G s phase NO2 has C2v symmetry which is retained in the top two binding geometries (Figures 4b and c) since the asymmetric O N stretching (v,J is not observed on-specular The symmetry is reduced to C, when NO2 is bonded as the bridging isomer (Figure 4a) and V, is dipole-active Confirmation of these bonding geometries (and the correct assignment of the rwo C2, isomers) comes from comparison with transition metal complexes containing the nitrite (NO23 ligand Determination of Adsorption Site While there is a general pattern of decreased metal-atom stretching frequency with increasing coordination of the adsorption site for the same metal-adatom combination, no site assignments can be made simply by observing the vibrational frequency Surface chemical bonds clearly control the site dependent vibrational shifts of adsorbed species Detailed studies that often involve impact scattering can determine the adatom adsorption site in some cases For polyatomic molecules, bonding to one or more metal atoms at the surface can not be distinguishedin general Good correlation does exist between the G O stretching frequency (VCO) adsorbed for CO and the adsorption site: vco > 2000 cm-' indicates an atop site (bonding to a single metal atom); 1850 cm-' > vco > 2000 cm-' indicates a bridgesite (bonding to two metal atoms); and vco < 1850 a - indicates a threefold or fourfold bridge n' sire (bonding in a region between three or four metal atoms) This correlation has 8.3 HREELS 451 L vs 1180 I x125 m 0N\ 0 800 0.75 ML O/Pt(lll) Figure4 1000 Energy loss / cm-l 2000 Specular spectra in HREELS of NO2 adsorbed in three different bonding geometries? been established by measuring the vibrational spectrum in conjunction with determining the adsorption site for CO by LEED crystallographycalculations, and also by examining the correlations for CO bonded in organometallic clusters Similar correlations are likely to exist for other diatomic molecules bonded to surfaces, e.g., NO, based on correlations observed in organometallic clusters but this has not been investigated sufficiently Figure shows the utility of H E E L S in establishing the presence of both bridge-bonded and atop CO chemisorbed on Pt( 111) and two SnPt alloy surfaces, and also serves to emphasize that H E E L S is very useful in studies of metal al10ys.~ The vcopeaks for CO bonded in bridge sites appear at 1865,1790, and 1845 cm-l on the Pt(l1 I), (2 x 2) and f i surfaces, respectively The vco peaks for CO 452 VIBRATIONAL SPECTROSCOPIES Chapter 1'' '' "' '' 1000 2000 3000 ENERGY LOSS (cm-1) Figure5 HREELS of the saturation coverage of CO on P t ( l l l and the (2 x 2) and x R30" Sn/Pt surface alloys.' (b ) bonded in atop sites appear at 2105,2090, and 2085 cm-' on the Pt( 11l), (2 x 2) and surfaces, respectively.Also,lower frequencyvp,co peaks accompany each of the vco peaks As discussed previously, the peak intensities are not necessarily proportional to the concentration ofeach type of CO species and the exact vco fiequency is determined by many k o r s h OthwAppliestiO~ Many other surfaces can be investigated by HEELS As larger molecule and nonsingle-crystal examples, we briefly describe the use of HREELS in studies of polymer suhces The usefulness of HRJZEU specifically in polymer surface science 83 HREELS 453 q-; A I -CH 800 1600 ,IC00 scissor 1800 Wave Number [cm-'l Figure Vibrational spectra of polymers (a) Transmission infrared spectrum of polyethylene; (b) electron-inducedloss spectrum of polyethylene; (c) transmission infrared spectrum of polypropylene.'0 applications has recently been reviewed by Gardella and Piream.' H E E L S is absolutely nondestructive and can be used to obtain information on the chemical composition, morphology, structure, and phonon modes of the solid surfice Many polymer surfaces have been studied, including simple materials like polyethylene, model compounds like Langmuir-Blodgett layers, and more complex systems like polymer physical mixtures Figure shows an H E E L S spectrum from polyethylene [CH3-(CH,),-CH,] Assignment of the energy loss peaks to vibrational modes is done exactly as described for adsorbates in the preceding seaion One observes a peak in the C-H stretching region near 2950 cm-', along with peaks due to C-C stretching and bending and C-H bending modes in the "fingerprint" region between 700-1500 cm-' from both the -CH3 (which terminate the chains)and -CHz groups in the polymer Since the CH3/CH2 ratio is vanishingly s d l in the bulk of the polymer, the high intensity of the -CH3 modes indicate 454 VIBRATIONAL SPECTROSCOPIES Chapter Figure HREELS vibrational spectra of the interface formation between a polyimide film and evaporated aluminum: (a) clean polyimide surface; (b) with 1/10 layer of AI; (c) with1/2 layer of AI.” that they are located preferentially in the extreme outer layers of the polymer surface lo HREELS is useful in many interfacial problems requiring monolayer sensitivity The incipient formation of the interface between a clean cured polyimide film and deposited aluminum has been studied using HREELS,ll as shown in Figure The film was PMDA-ODA [poly-N,N’-bis(phenoxyphenyl)pyromellitimide], shown I schematically in Figure At low A coverage, the v(C=O) peak at 1720 cm-l is affected strongly, which indicates that Al reacts close to the carbonyl site At higher AI coverage, new peaks at 2950 and 3730 cm-’ appear which are due to aliphatic -CH, and -OH groups on the surface This is evidence for bond scissions in the polymer skeleton In general, the main problems with the analysis of bulk polymers has been charging and rough surfaces The latter characteristic makes the specular direction poorly defined, which causes diffuse and weak electron scattering Preparation of the polymer as a thin film on a conducting substrate can overcome the charging problem Even thick samples of insulating polymers can now be studied using a “flood gun” technique Thiry and his coworkers’2 have shown that charging effects can be over- - 83 HREELS 455 r -'n L PMDA Figure ODA Structure o PMDA-ODA f come by using an auxiliary defocused beam of high-energy electronsto give neutralization of even wide-gap insulators, including AlZO3, MgO, SiO2, LiF, and NaC1 Comparison to Other Techniques Information on vibrations at surfaces is complementary to that provided on the compositional analysis by AES and SIMS, geometrical structure by LEED, and electronic structure by X P S and UPS Vibrational spectroscopy is the most powerful method for the identification of molecular groups at surfaces, giving information directly about which atoms are chemically bonded together These spectra are more directly interpreted to give chemical bonding information and are more sensitive to the chemical state of surface atoms than those in UPS or X P S For example, the C( 1s) binding energy shift in XPS between C=Oand G O species is 1.5 eV and that between C=Cand C-C species is 0.7 eV, with an instrumental resolution of typically eV In contrast, the vibrational energy difference between C=O and G O species is 1000 cm-' and that between C=C and G C species is 500 cm-', with an instrumental resolution of typically 60 cm-' Vibrational spectroscopycan handle the complications introduced by mixtures of many different surface species much better than UPS or XPS Many other techniques are capable of obtaining vibrational spectra of adsorbed species: infrared transmission-absorption (IR) and infrared reflection-absorption spectroscopy (IRAS),& s enhanced Raman spectroscopy (SERS), inelastic electron tunneling spectroscopy (IETS), neutron inelastic scattering (NIS), photoacoustic spectroscopy (PAS), and atom inelastic scattering (AIS) The analytical characteristicsof these methods have been compared in several reviews previously The principle reasons for the extensive use of the optical probes, e.g., IR compared to HREELS in very practical nonsingle-crystal work are the higher resolution ( 02 cm-') and the possibility for use at ambient pressures HREELS could be &ectively used to provide high surfice sensitivity and a much smaller sampling depth (e nm) and wider spectral range (50-4000 c - )than many of these other methm' ods 456 VIBRATIONAL SPECTROSCOPIES Chapter HREELS is used extensively in adsorption studies on metal single crystals, since its high sensitivity to small dynamic dipoles, such as those of C-C and C-H stretching modes, and its wide spectral range enable complete vibrational characterization of submonolayer coverages of adsorbed hydrocarbons.l The dipole selection rule constraint in IR, IRAS, and HREELS can be broken in HREELS by performing off-specular scans so that all vibrational modes can be observed This is important in species identification, and critical in obtaining vibrational frequencies required to generate a molecular force field and in determining adsorption sites Conclusions HREELS is one of the most important techniques for probing physical and chemiih cal properties of suhces The future is bright, w t new opportunities arising fiom continued fundamental advances in understanding electron scattering mechanisms and from improved instrumentation, particularly in the more quantitative aspects of the te~hnique.’~ better understanding of the scattering of electrons fiom surA faces means better structure determination and better probe of electronic properties Improvements are coming in calculating HREELS cross sections and surface phonon properties and this means a better understanding of lanice dynamics Extensions ofdielectric theory of HREELS could lead to new applications concerning interface optical phonons and other properties of superlattice interfaces Novel applications of the HREELS technique include the use of spin-polarization of the incident or analyzed electrons and time-resolved studies on the ms and sub-ms time scale (sometimes coupled with pulsed molecular beams) of dynamical aspects of chemisorption and reaction Studies of nontraditional surfaces, such as insulators, alloys, glasses, superconductors, model supported metal catalysts, and “technical” surfaces (samples of actual working devices) are currently being expanded Many of these new studies are made possible through improved instrumentation While the resolution seems to be limited practically at 10 an-’, higher intensity seems achievable Advances have been made recently in the monochromator, analyzer, lenses, and signal detection (by using multichannel detection) New configurations, such as that utilized in the dispersion compensation approach, have improved signal levels by factors of 102-103 Related Articles in the Encyclopedia EELS, IR, FTIR, and b a n Spectroscopy References 8.3 H Ibach and D L Mills Ek-ctron Energy Loss Spectroscopy andSu$ace vibrations Academic, New York, 1982 An excellent book covering all aspects of the theory and experiment in HREELS HREELS 457 ... mathematical details of polarization in optics D E Aspnes In: Handbook of Optical Constana of Solid (E Palik, ed.) Academic Press,Orlando, 1 985 Description of use of ellipsometry to determine optical... 31,99, 1 988 Review of use of variable-angle spectroscopicellipsometer (VASE) for semiconductors J A Woollam and I? G Snyder M a t Sci Eng B5,279,1990 Recent review of application of VASE in materials. .. Bu-Abbud, N M Bashara, and J A WooUarn Thin Solid Film 1 38, 27, 1 986 Description of Marquardt algorithm and parameter sensitivity correlation in ellipsometry i o D E Aspnes Thin Solid Film 89 ,249,