7.2 Infrared (FTIR) and Raman Spectroscopy: Vibrational Modes and Instrumentation

7.2 Infrared (FTIR) and Raman Spectroscopy: Vibrational Modes and Instrumentation

Quick Answer: Vibrational spectroscopy measures transitions between molecular vibrational energy levels in the mid-infrared region ($4000-400\text{ cm}^{-1}$). Infrared absorption requires a net change in molecular dipole moment ($\frac{d\mu}{dq} \neq 0$), whereas Raman scattering requires a change in polarizability ($\frac{d\alpha}{dq} \neq 0$). For centrosymmetric molecules, the Rule of Mutual Exclusion prevents any mode from being both IR and Raman active. Modern FTIR utilizes a Michelson interferometer with a He-Ne laser for optical path tracking, providing Fellgett (multiplex), Jacquinot (throughput), and Connes (calibration) advantages over dispersive optics. Attenuated Total Reflectance (ATR) eliminates sample preparation by probing an evanescent wave at the crystal-sample boundary, while Raman scattering excels in aqueous systems where water absorbs IR strongly.

---

Physics of Molecular Vibrations: Harmonic and Anharmonic Models

In classical mechanics, molecular vibration between two bonded nuclei is approximated by a simple harmonic oscillator governed by Hooke's law, where restoring force ($F$) is proportional to displacement ($x = r - r_e$) from equilibrium bond distance ($r_e$):

$F = -k \, x \implies V(x) = \frac{1}{2} k \, x^2$

Applying the Schrödinger equation yields quantized vibrational energy levels:

$E_v = \left( v + \frac{1}{2} \right) h \nu_0, \quad v = 0, 1, 2, 3\dots$

where $v$ is the vibrational quantum number and $h$ is Planck's constant ($6.626 \times 10^{-34}\text{ J}\cdot\text{s}$). Even at absolute zero ($T = 0\text{ K}$), a molecule retains a finite zero-point vibrational energy of $E_0 = \frac{1}{2} h \nu_0$.

Potential Energy Curves:

  V(r) ^                 Anharmonic Morse Potential (Dissociation De)
       |      /\           . - - - Harmonic Parabola V = 1/2 k x^2
       |     /  \        . '
       |    /    \     . '   v = 3 ---------------------
       |   /      \  . '     v = 2 ---------------------
       |  /        '         v = 1 ---------------------
       | /       . '         v = 0 ---------------------
       +--------+----------------------------------------> Interatomic r
               r_e (Equilibrium Distance)

The Hooke's Law Equation for Vibrational Frequency

The fundamental frequency (wavenumber $\tilde{\nu}$ in $\text{cm}^{-1}$) of a diatomic harmonic oscillator is:

$\tilde{\nu} = \frac{1}{2\pi c} \sqrt{\frac{k}{\mu}}$

where $c$ is the speed of light ($2.998 \times 10^{10}\text{ cm/s}$), $k$ is the bond force constant in dynes per centimeter ($1\text{ N/m} = 10^3\text{ dyn/cm}$), and $\mu$ is the reduced mass of the two oscillating atoms:

$\mu = \frac{m_1 \, m_2}{m_1 + m_2}$

• Force Constant ($k$) Dependency: Bond strength directly dictates force constant: single bonds ($k \approx 5 \times 10^5\text{ dyn/cm}$) < double bonds ($k \approx 10 \times 10^5\text{ dyn/cm}$) < triple bonds ($k \approx 15 \times 10^5\text{ dyn/cm}$). Consequently, stretching frequencies increase systematically: $C-C$ ($\,\sim 1200\text{ cm}^{-1}$) < $C=C$ ($\,\sim 1650\text{ cm}^{-1}$) < $C\equiv C$ ($\,\sim 2150\text{ cm}^{-1}$).
• Reduced Mass ($\mu$) Dependency (Isotope Effects): Heavy atoms vibrate at lower frequencies. For example, replacing a hydrogen atom with deuterium in a $C-H$ bond approximately doubles the reduced mass ($\mu_{C-H} \approx 0.923\text{ amu}$ vs. $\mu_{C-D} \approx 1.714\text{ amu}$), shifting the stretching frequency from $\sim 3000\text{ cm}^{-1}$ down to $\sim 2200\text{ cm}^{-1}$: $\frac{\tilde{\nu}_{C-D}}{\tilde{\nu}_{C-H}} = \sqrt{\frac{\mu_{C-H}}{\mu_{C-D}}} = \sqrt{\frac{0.923}{1.714}} \approx 0.734$

Anharmonicity, Overtones, and Fermi Resonance

Real chemical bonds diverge from ideal harmonic behavior at higher vibrational energies due to internuclear repulsion at short distances and homolytic bond dissociation ($D_e$) at long distances, modeled by the Morse potential ($V(r) = D_e[1 - e^{-a(r-r_e)}]^2$). Key consequences of anharmonicity include:

1. Relaxation of Harmonic Selection Rules: While harmonic oscillators permit only fundamental transitions ($\Delta v = \pm 1$), anharmonicity allows weak overtone bands ($\Delta v = \pm 2, \pm 3$, appearing at approximately $2\tilde{\nu}_0, 3\tilde{\nu}_0$ with $10-100\times$ lower intensity) and combination bands ($\tilde{\nu}_1 + \tilde{\nu}_2$) or difference bands ($\tilde{\nu}_1 - \tilde{\nu}_2$).
2. Fermi Resonance: Occurs when a fundamental vibrational mode accidentally coincides in energy and possesses identical quantum mechanical symmetry with an overtone or combination band of another vibration. Quantum mechanical mixing causes energy repulsion: the two transitions split into a prominent doublet with altered frequencies and borrowed intensities. Classic examples include the carbon dioxide Raman doublet ($1285$ and $1388\text{ cm}^{-1}$) and the split carbonyl band in benzoyl chloride and cyclopentanone.

---

Vibrational Degrees of Freedom and Normal Modes

A molecule composed of $N$ atoms possesses $3N$ total degrees of freedom, representing three-dimensional Cartesian coordinates ($x, y, z$) for every nucleus:

• Translational Degrees of Freedom: 3 (movement of the molecular center of mass along $x, y, z$).
• Rotational Degrees of Freedom:
  • Non-Linear Molecule: 3 (rotation about $x, y, z$ axes).
  • Linear Molecule: 2 (rotation about two axes perpendicular to the internuclear axis; rotation about the molecular bond axis imparts zero moment of inertia).

$\text{Fundamental Vibrational Modes} = \begin{cases} 3N - 6 & \text{for non-linear molecules} \\ 3N - 5 & \text{for linear molecules} \end{cases}$

Classification of Normal Modes: Stretching vs. Bending

1. Stretching Vibrations ($\nu$): Continuous rhythmic movement along the bond axis, expanding and compressing internuclear distances. Stretching requires higher energy than bending:
   • Symmetric Stretch ($\nu_s$): Both bonds expand and contract simultaneously.
   • Asymmetric Stretch ($\nu_{as}$): One bond extends while the adjacent bond compresses. Asymmetric stretching occurs at higher wavenumbers than symmetric stretching due to electronic dipole repulsion.
2. Bending Vibrations ($\delta$): Deformation of bond angles between three bonded atoms without altering bond lengths; lower force constants dictate that bending modes appear at lower wavenumbers ($<1500\text{ cm}^{-1}$):
   • In-plane deformations: Scissoring (two atoms move toward each other) and Rocking (atoms sway in unison in the same planar direction).
   • Out-of-plane deformations: Wagging (atoms swing together above and below the molecular plane) and Twisting (atoms rotate in opposite directions out of the plane).

---

Infrared vs. Raman Selection Rules and the Rule of Mutual Exclusion

The Rule of Mutual Exclusion

For any molecule possessing a center of inversion ($i$; centrosymmetric), such as carbon dioxide ($O=C=O$), ethylene ($H_2C=CH_2$), trans-1,2-dichloroethylene, or benzene ($C_6H_6$):

$\text{No vibrational mode can be both Infrared active and Raman active. }$

• Vibrations that are symmetric with respect to inversion through the center of symmetry are gerade ($g$); they involve polarizability changes and are Raman active but IR inactive.
• Vibrations that are antisymmetric with respect to inversion are ungerade ($u$); they induce dynamic dipole moments and are IR active but Raman inactive.
• In $CO_2$, the symmetric stretch ($1388\text{ cm}^{-1}$) maintains zero dipole moment ($d\mu/dq = 0$) but expands electron cloud polarizability ($d\alpha/dq \neq 0$), making it exclusively Raman active. The asymmetric stretch ($2349\text{ cm}^{-1}$) and degenerate bends ($667\text{ cm}^{-1}$) generate oscillating dipole moments ($d\mu/dq \neq 0$) without altering polarizability symmetry, rendering them exclusively IR active.
• Conversely, for non-centrosymmetric molecules lacking an inversion center (such as water, $H_2O$, $C_{2v}$), fundamental modes can be both IR and Raman active.

---

FTIR Spectrometer Architecture: The Michelson Interferometer

Modern FTIR instruments replace dispersive diffraction gratings with a Michelson interferometer, which modulates mid-infrared radiation into the time/spatial domain.

Michelson Interferometer Optical Train:

                    Moving Mirror (Moves at velocity v_M)
                           [ === ]
                              ^ 
                              |  (Optical Path Difference δ = 2(d_M - d_F))
                              v
   IR Source  ------>  [ Beam Splitter ]  ------>  Sample Cuvette  ------>  Detector
   (Globar)            (KBr/Ge Film)          (Transmission or ATR)       (DTGS/MCT)
                              ^
                              |
                           [ === ]
                    Fixed Stationary Mirror

Interferometer Mechanics and the Interferogram

1. Broadband Source: A ceramic silicon carbide rod (Globar, heated resistively to $1200-1500\text{ K}$) or Nichrome wire coil emits continuous infrared radiation.
2. Beam Splitter: A semi-reflecting flat optical plate (typically a polished potassium bromide $KBr$ substrate coated with a submicron film of germanium) positioned at a $45^\circ$ incident angle. It divides the incoming collimated beam into two beams of equal radiant power ($50\%$ reflected, $50\%$ transmitted).
3. Fixed and Moving Mirrors: One half-beam strikes a fixed stationary mirror; the second half-beam strikes a moving mirror translating smoothly along an air-bearing carriage at constant velocity ($v_M$).
4. Optical Path Difference (Retardation, $\delta$): Because the moving mirror travels a physical distance $\Delta d$, the total optical path traveled by that beam increases by twice that displacement: $\delta = 2(d_{\text{moving}} - d_{\text{fixed}}) = 2 \, v_M \, t$
5. Interference Recombination: When the two reflected beams recombine at the beam splitter:
   • Constructive interference occurs whenever retardation equals an integer multiple of wavelength: $\delta = n\lambda$.
   • Destructive interference occurs whenever retardation is a half-integer multiple: $\delta = (n + 1/2)\lambda$.
6. The Interferogram: At zero path difference ($\delta = 0$, ZPD), all wavelengths interfere constructively simultaneously, producing a massive intensity spike designated the centerburst. As $\delta$ increases away from ZPD, the composite signal decays into a complex interference waveform designated the interferogram ($I(\delta)$).
7. The Fast Fourier Transform (FFT): The interferogram represents signal intensity as a function of optical retardation ($I(\delta)$). The instrument's digital signal processor computes the continuous inverse Fourier transform, resolving the composite spatial wave into individual frequency intensities, generating a single-beam spectrum ($S(\tilde{\nu})$): $S(\tilde{\nu}) = \int_{-\infty}^{+\infty} I(\delta) \cos(2\pi \tilde{\nu} \delta) \, d\delta$

Canonical Instrumental Advantages of FTIR

1. Fellgett (Multiplex) Advantage: In a dispersive instrument, individual resolution elements are scanned sequentially across an exit slit, requiring prolonged scan times. In FTIR, all spectral frequencies reach the detector simultaneously throughout the entire duration of the scan. For a spectrum with $M$ resolution elements, the signal-to-noise ratio ($S/N$) improves by a factor of $\sqrt{M}$ for equal acquisition times, or the measurement speed is accelerated by a factor of $M$ for equivalent $S/N$.
2. Jacquinot (Throughput) Advantage: Dispersive spectrometers require narrow entrance and exit slits to achieve spectral resolution, attenuating $>99\%$ of source radiant power. The Michelson interferometer possesses no slits, utilizing a wide circular Jacquinot stop aperture. Throughput of optical radiation reaching the detector is $10$ to $100$ times higher, drastically improving $S/N$.
3. Connes Advantage (Laser Wavenumber Precision): Modern FTIR systems pass the beam of a continuous internal Helium-Neon (He-Ne) gas laser (emitting at precisely $\lambda_0 = 632.8\text{ nm}$) through the exact same interferometer path. The sinusoidal interference fringes generated by the He-Ne laser trigger digital analog-to-digital converter (ADC) sampling at exact, reproducible mirror displacement intervals. This provides an absolute, invariant internal frequency calibration with wavenumber accuracy exceeding $\pm 0.01\text{ cm}^{-1}$, eliminating baseline calibration drift.

---

Mid-IR Sampling Techniques: Transmission vs. Attenuated Total Reflectance (ATR)

1. Transmission Sampling Methods

• Potassium Bromide ($KBr$) Pellets: $1.0-2.0\text{ mg}$ of finely powdered solid sample is ground with $100-200\text{ mg}$ of spectroscopic grade, dry $KBr$ in an agate mortar to a particle diameter smaller than the IR wavelength ($<2\text{ }\mu\text{m}$ to prevent Christiansen scattering). The mixture is compressed in an evacuated die under $8-10\text{ tons}$ hydraulic pressure, producing a transparent disk. Limitation: $KBr$ is highly hygroscopic; atmospheric moisture contamination introduces spurious broad $O-H$ bands at $3450\text{ cm}^{-1}$ and $1640\text{ cm}^{-1}$.
• Nujol Mulls: Solid sample is ground with a drop of heavy mineral oil (Nujol, a branched paraffin). The viscous suspension is pressed between polished $NaCl$ or $KBr$ salt plates. Limitation: Nujol exhibits intense aliphatic $C-H$ stretching ($2950-2850\text{ cm}^{-1}$), $C-H$ bending ($1462, 1377\text{ cm}^{-1}$), and methylene rocking ($720\text{ cm}^{-1}$) that completely obscure sample functional groups in those diagnostic regions.
• Liquid Cells: Fixed-pathlength cells ($0.015-1.0\text{ mm}$) bounded by water-soluble $NaCl$ (usable down to $650\text{ cm}^{-1}$) or $KBr$ ($400\text{ cm}^{-1}$) windows. Aqueous solutions cannot be analyzed in these cells as water dissolves the crystal plates.

2. Attenuated Total Reflectance (ATR) Spectroscopy

ATR has revolutionized modern analytical laboratories by permitting direct, non-destructive analysis of solids, liquids, pastes, and polymers without sample preparation.

Attenuated Total Reflectance (ATR) Crystal Interface:

                 Sample Clamped in Intimate Optical Contact (n2)
  ========================================================================
               ~~~~~ Evanescent Wave (Penetration dp) ~~~~~
  ------------------------------------------------------------------------
         /\                                                /\
        /  \                                              /  \  Total Internal
       /    \                                            /    \ Reflection
      /      \                                          /      \
  -->/   θ    \----------------------------------------/        \--> To Detector
     High-Refractive-Index IRE Crystal (Diamond, ZnSe, Ge; n1)

• Principle of Internal Reflection: The sample is clamped in intimate physical contact against an Internal Reflection Element (IRE) crystal possessing a high refractive index ($n_1$). The infrared beam travels through the crystal, striking the sample interface at an angle of incidence ($\theta$, typically $45^\circ$) exceeding the critical angle ($\theta_c$): $\theta_c = \arcsin\left( \frac{n_2}{n_1} \right)$
• IRE Crystal Materials:
  • Diamond ($n_1 = 2.4$): Chemically inert, diamond-coated tungsten carbide substrate; impervious to strong acids, bases, abrasive powders; extremely durable.
  • Zinc Selenide ($ZnSe$, $n_1 = 2.4$): Economical, transparent $4000-600\text{ cm}^{-1}$; toxic if fractured, easily scratched, attacked by complexing agents and extreme $pH$ ($<5$ or $>9$).
  • Germanium ($Ge$, $n_1 = 4.0$): Very high refractive index, producing shallow penetration depth; ideal for highly absorbing, dark samples like carbon-black filled elastomers.
• Evanescent Wave and Penetration Depth ($d_p$): Total internal reflection generates an evanescent standing electromagnetic wave that penetrates perpendicularly across the crystal boundary into the sample. The electric field amplitude decays exponentially with distance. The depth of penetration ($d_p$, distance where electric field decays to $1/e = 36.8\%$ of its surface value) is governed by: $d_p = \frac{\lambda_0}{2\pi n_1 \sqrt{\sin^2\theta - (n_2/n_1)^2}}$ where $\lambda_0$ is the vacuum infrared wavelength. Typical penetration depths range from $0.5\text{ }\mu\text{m}$ to $3.0\text{ }\mu\text{m}$.
• Wavelength Dependence and ATR Correction: Crucially, $d_p$ is directly proportional to wavelength ($\lambda_0 = 1/\tilde{\nu}$). Consequently, the beam penetrates deeper into the sample at lower wavenumbers (longer wavelengths). In raw ATR spectra, absorption bands at $1000\text{ cm}^{-1}$ appear roughly three times more intense than bands at $3000\text{ cm}^{-1}$ relative to transmission spectra. Modern FTIR software applies a mathematical ATR correction ($A_{\text{trans}} \approx A_{\text{ATR}} \times \tilde{\nu}$) to match standard transmission library databases.

---

Raman Spectroscopy: Scattering Mechanics and SERS

1. Inelastic Scattering Mechanics

When a sample is irradiated with high-intensity monochromatic laser light ($h\nu_0$, typically $532\text{ nm}, 785\text{ nm}$, or $1064\text{ nm}$), the electric field polarizes the molecular electron cloud, exciting the molecule to a transient, non-quantized virtual state. De-excitation yields three scattering possibilities:

1. Rayleigh Scattering (Elastic): The molecule relaxes back to its initial vibrational ground state ($v=0$). The scattered photon retains the exact frequency of the excitation laser ($h\nu_s = h\nu_0$). Rayleigh scattering accounts for $>99.999\%$ of scattered photons and is blocked in spectrometers using holographic notch or edge filters.
2. Stokes Raman Scattering (Inelastic, Red-Shifted): The molecule, initially in the vibrational ground state ($v=0$), absorbs vibrational energy ($h\nu_{\text{vib}}$) and relaxes to an excited vibrational level ($v=1$). The emitted photon emerges with lower energy and lower frequency: $h\nu_s = h\nu_0 - h\nu_{\text{vib}} \implies \Delta\tilde{\nu} = \tilde{\nu}_0 - \tilde{\nu}_s > 0$
3. Anti-Stokes Raman Scattering (Inelastic, Blue-Shifted): The molecule, already residing in an excited vibrational state ($v=1$) due to thermal excitation, relaxes down to the vibrational ground state ($v=0$). The scattered photon extracts vibrational energy, emerging with higher frequency: $h\nu_s = h\nu_0 + h\nu_{\text{vib}} \implies \Delta\tilde{\nu} = \tilde{\nu}_0 - \tilde{\nu}_s < 0$

Energy Level Transitions in Raman and Rayleigh Scattering:

                 Virtual States (Transient Polarization)
  =========================================================================
       ^   |            ^         |              ^              |
       |   |            |         |              |              |
       |   |            |         |              |              |
  hν0  |   | hν0   hν0  |         | hνs =        | hν0          | hνs =
       |   |            |         | hν0 - hνvib  |              | hν0 + hνvib
       |   v            |         v              |              v
  -----+----------------+------------------------+------------------------- v = 1
       |                |         (Stokes)       |              (Anti-Stokes)
  -----+----------------+------------------------+------------------------- v = 0
         Rayleigh                 Stokes                   Anti-Stokes

Stokes vs. Anti-Stokes Intensity Ratio and Temperature Probing

Because Anti-Stokes scattering requires the initial population of an excited vibrational state ($v=1$), its intensity relative to Stokes scattering is governed strictly by the Boltzmann distribution:

$\frac{I_{\text{Anti-Stokes}}}{I_{\text{Stokes}}} = \left( \frac{\nu_0 + \nu_{\text{vib}}}{\nu_0 - \nu_{\text{vib}}} \right)^4 \exp\left( -\frac{h \nu_{\text{vib}}}{k_B T} \right)$

At room temperature ($298\text{ K}$), the exponential term ensures that Stokes scattering is orders of magnitude more intense than Anti-Stokes scattering. However, as sample temperature rises, the Anti-Stokes intensity increases exponentially, permitting non-contact optical temperature sensing.

2. Surface-Enhanced Raman Spectroscopy (SERS)

Normal Raman scattering exhibits very low cross-sections ($10^{-30}\text{ cm}^2/\text{molecule}$). When target analyte molecules are adsorbed onto or positioned within nanometer proximity ($<5\text{ nm}$) of nanoscale roughened noble metal substrates (colloidal silver or gold nanoparticles, nanovoids, or lithographic nanopillars), Raman scattering intensities are amplified by factors of $10^6$ to $10^{14}$, enabling single-molecule detection.

• Electromagnetic Mechanism ($10^4-10^8$ enhancement): Resonant excitation of Localized Surface Plasmon Resonances (LSPR) in the metal nanostructure amplifies the local incident electric field ($E_{\text{local}}$). SERS intensity scales as the fourth power of the local electric field enhancement: $I_{\text{SERS}} \propto |E_{\text{local}}|^4$.
• Chemical Mechanism ($10^1-10^2$ enhancement): Charge-transfer complexes form between the metal Fermi levels and analyte frontier molecular orbitals, substantially increasing analyte polarizability derivatives ($d\alpha/dq$).