Spectroscopic Methods of Analysis
CSIR NET chemistry: Beer–Lambert law and deviations, multicomponent analysis, Job's method, AAS and its interferences, flame and ICP emission, fluorescence.
- 21 min
On this page · 15 sections
- Prerequisites
- The Hook: The Same Sodium Line, Two Opposite Instruments
- 1. The Beer–Lambert Law
- 2. Quantitative Spectrophotometry in Practice
- 3. Atomic Absorption Spectroscopy (AAS)
- 4. Atomic Emission: Flame Photometry and ICP
- 5. Molecular Fluorescence
- 6. Other Optical Methods in One Table
- Memory Tricks
- Don’t Confuse
- Common Mistakes
- Practice
- Quick Revision Table
- Key Takeaways
- What’s Next
Prerequisites
- Separation Methods: UV, fluorescence and atomic detectors are the eyes of every chromatograph.
- Biophysical Methods: Spectroscopy and Structure: the Life Sciences treatment of UV–visible and fluorescence, if you want a second angle on the same physics.
- Wave–Particle Duality and the Schrödinger Equation: quantised energy levels and E = hν, the reason absorption lines exist at all.
- Logarithms and exponentials: absorbance is a log of a ratio, and the Boltzmann factor is an exponential.
The Hook: The Same Sodium Line, Two Opposite Instruments
1. The Beer–Lambert Law
Monochromatic light of radiant power \( P_0 \) passes through a path b of a solution of concentration c and leaves with power P.
$$T = \frac{P}{P_0} \qquad \boxed{A = -\log_{10} T = \varepsilon\,b\,c}$$| Symbol | Name | Usual unit |
|---|---|---|
| T | transmittance (×100 for %T) | none |
| A | absorbance | none |
| ε | molar absorption coefficient (molar absorptivity) | L mol⁻¹ cm⁻¹ |
| b | path length | cm |
| c | concentration | mol L⁻¹ |
Useful pairs: T = 10% ↔ A = 1.00; T = 1% ↔ A = 2.00; T = 50% ↔ A = 0.301. Absorbance is additive: for a mixture, \( A = b\sum_i \varepsilon_i c_i \) at each wavelength.
Deviations from Beer’s law
The law assumes monochromatic light, dilute solution and absorbers that do not change identity with concentration. When a calibration plot of A against c curves, the cause is one of three kinds:
| Kind | Cause | Typical sign |
|---|---|---|
| Real | at high concentration (above about 0.01 M) solute–solute interactions and refractive-index changes alter ε | curvature at high c only |
| Chemical | the absorber takes part in an equilibrium that shifts with dilution: weak-acid dissociation, dimerisation, complex dissociation. Example: dichromate–chromate, \( \text{Cr}_2\text{O}_7^{2-} + \text{H}_2\text{O} \rightleftharpoons 2\text{HCrO}_4^- \), so an unbuffered solution does not obey Beer’s law at most wavelengths | positive or negative curvature; disappears at an isosbestic point or in a buffer |
| Instrumental | polychromatic light (finite bandwidth) where ε varies across the band; stray light reaching the detector; mismatched cells | always negative deviation at high A for stray light |
Where the reading is most precise
If the dominant error is a constant uncertainty in reading T (typical of older instruments), the relative concentration error is
$$\frac{\Delta c}{c} = \frac{0.434\,\Delta T}{T\,\log T}$$which is smallest at T = 1/e = 36.8%, i.e. A = 0.434. In practice absorbances between about 0.2 and 0.8 give the best precision; dilute or change the path length to land there.
Isosbestic points
If two absorbing species interconvert (HIn ⇌ In⁻, or free ligand ⇌ complex) and only these two are present, every spectrum in a series of different pH values or mixing ratios passes through the wavelength where their ε values are equal. This isosbestic point is evidence that only two species are involved; its loss signals a third species or a side reaction.
2. Quantitative Spectrophotometry in Practice
Calibration, sensitivity and detection limit
- External calibration: measure standards, fit A = m c + intercept, read unknowns from the line.
- Sensitivity is the slope m of the calibration line.
- Limit of detection (common IUPAC-style convention): the concentration giving a signal three times the standard deviation of the blank, \( c_{\text{LOD}} = 3s_{\text{blank}}/m \). The limit of quantitation uses 10 s.
Example: blank standard deviation 0.002 absorbance units, slope 0.050 per ppm: LOD = 3 × 0.002/0.050 = 0.12 ppm.
Standard addition
When the sample matrix changes the signal (viscosity in a nebuliser, other absorbers, chemical interference), standards made in pure water give the wrong slope. Standard addition puts the standards into the sample itself, so the matrix is the same.
An internal standard is different: a known amount of a different element or compound is added to every sample and standard, and the ratio of signals is plotted. It corrects for fluctuations in injection volume, nebulisation rate or flame conditions (common in flame emission, ICP and chromatography).
Two-component mixtures
Because absorbances add, two components X and Y can be determined from measurements at two wavelengths, ideally the two absorption maxima:
$$A_{\lambda_1} = \varepsilon_{X,1}\,b\,c_X + \varepsilon_{Y,1}\,b\,c_Y \qquad A_{\lambda_2} = \varepsilon_{X,2}\,b\,c_X + \varepsilon_{Y,2}\,b\,c_Y$$Two equations, two unknowns. The classic example is Cr(VI) and Mn(VII) (dichromate and permanganate) in steel analysis.
Stoichiometry of complexes
| Method | Procedure | Result |
|---|---|---|
| Job’s method of continuous variation | equimolar M and L solutions mixed in varying ratios, total moles constant; plot A (corrected for M and L) against mole fraction of L, \( x_L \) | maximum at \( x_L = n/(1+n) \) for ML_n: 0.5 for ML, 0.67 for ML₂, 0.75 for ML₃ |
| Mole-ratio method | fixed [M], increasing [L]; plot A against [L]/[M] | two straight lines meeting at the ratio n; a rounded corner means a weak complex |
| Slope-ratio method | two series, one with excess L, one with excess M | ratio of slopes gives n; suits weak complexes |
The curvature near the break or maximum also carries the formation constant: the sharper the peak, the more stable the complex.
Photometric titrations
Absorbance is followed during a titration at a wavelength where one species absorbs. Because A is linear in concentration, each branch is a straight line (after correcting for dilution) and the end point is their intersection.
| Absorbing species | Shape of A versus titrant volume |
|---|---|
| Titrant only | flat at zero, then rises after the end point |
| Product only | rises to the end point, then flat |
| Analyte only | falls to the end point, then flat at zero |
| Analyte and titrant (product does not absorb) | falls, then rises: V-shape |
Photometric end points work even when the reaction is not complete near equivalence, because extrapolation uses points away from it. Example: Cu²⁺ titrated with EDTA at a wavelength where only the Cu–EDTA complex absorbs gives a rising line followed by a flat one.
3. Atomic Absorption Spectroscopy (AAS)
The sample solution is nebulised into a flame (or dried and atomised in a graphite furnace), turning the analyte into free ground-state atoms. Light at the element’s resonance line is passed through and absorbance is measured.
Why a hollow-cathode lamp
Atomic absorption lines are extremely narrow (of the order of thousandths of a nanometre). A continuum source passed through a monochromator would present a band far wider than the line, and the tiny fraction absorbed would barely change the signal. A hollow-cathode lamp whose cathode is made of the analyte element emits that element’s lines, even narrower than the absorption lines because the lamp runs at low pressure and temperature. So each element needs its own lamp (or a multi-element lamp).
The atomic line width comes from three sources: the natural (uncertainty) width, Doppler broadening from thermal motion (grows with temperature), and pressure (collisional) broadening.
Atomisers
| Atomiser | Features |
|---|---|
| Air–acetylene flame | most common; suitable for many metals |
| Nitrous oxide–acetylene flame | hotter and more reducing; needed for elements that form refractory oxides (Al, Si, Ti, V, Mo) |
| Graphite furnace (electrothermal) | the whole sample is atomised in a small tube in a programmed dry–ash–atomise sequence; much lower detection limits than a flame, microlitre samples, but slower and more matrix effects |
| Cold-vapour | Hg reduced to the element by SnCl₂ and swept as vapour |
| Hydride generation | As, Se, Sb, Bi, Sn converted to volatile hydrides by NaBH₄ and decomposed in a heated cell |
Interferences and their cures
| Interference | What happens | Cure |
|---|---|---|
| Spectral (line overlap) | another element absorbs at the same wavelength | choose another line |
| Spectral (background) | broad molecular absorption or scattering by particles | background correction: deuterium-lamp continuum, Zeeman effect, or Smith–Hieftje (lamp pulsing) |
| Chemical | analyte forms a compound that does not atomise: Ca²⁺ with phosphate, sulfate, silicate or aluminate gives low results | releasing agent (La³⁺ or Sr²⁺ binds the phosphate preferentially); protective agent (EDTA, 8-hydroxyquinoline forms a volatile complex with the analyte); hotter N₂O–C₂H₂ flame |
| Ionisation | in a hot flame, atoms of low ionisation energy (alkali metals) are partly ionised, so fewer neutral atoms absorb | ionisation suppressor: add a large excess of a more easily ionised element (K, Cs) to raise the electron concentration and push M ⇌ M⁺ + e⁻ back |
| Physical (matrix) | viscosity, surface tension, dissolved solids change nebulisation | matrix-matched standards or standard addition |
4. Atomic Emission: Flame Photometry and ICP
The Boltzmann distribution
At temperature T, the ratio of atoms in an excited state j to those in the ground state 0 is
$$\boxed{\frac{N_j}{N_0} = \frac{g_j}{g_0}\exp\left(-\frac{\Delta E}{kT}\right)}$$where g are statistical weights (degeneracies). Emission intensity is proportional to \( N_j \); absorption to \( N_0 \).
Consequences:
- Emission needs tight temperature control; absorption tolerates temperature fluctuations.
- Elements with low-energy excited states (Na, K, Li, Ca) emit strongly even in cool flames, which is why flame photometry is the routine method for Na and K in clinical and soil samples. Characteristic lines: Na 589 nm (yellow), K 766 nm (deep red; the flame itself looks lilac), Li 671 nm (crimson).
- Elements whose resonance lines lie deep in the UV (high ΔE, such as Zn) emit poorly in flames; AAS or a hotter source is better.
Inductively coupled plasma (ICP)
An argon plasma sustained by a radio-frequency coil reaches temperatures several thousand kelvin higher than the air–acetylene and nitrous oxide–acetylene flames used in AAS. Consequences:
- almost complete atomisation; chemical interferences (Ca–phosphate) are largely removed;
- excited populations are large, so many elements, including refractory ones, emit strongly;
- the high electron density from ionised argon suppresses ionisation interference of the analyte;
- multi-element analysis: one spectrum, many lines, measured simultaneously;
- a wide linear range (several orders of magnitude);
- the plasma ions can also be fed to a mass spectrometer (ICP-MS), giving very low detection limits and isotope ratios.
The price: spectral interferences multiply because so many lines are emitted, and the argon consumption is high.
AAS versus emission at a glance
| Feature | AAS | Flame emission | ICP emission |
|---|---|---|---|
| Measures | ground-state atoms (absorption) | excited atoms (emission) | excited atoms and ions |
| Temperature sensitivity | low | high | low (stable plasma) |
| Source needed | element-specific lamp | none | none |
| Elements per run | one (per lamp) | one or a few | many simultaneously |
| Best for | most metals at ppm (flame) or ppb (furnace) | Na, K, Li, Ca | multi-element surveys, refractory elements |
5. Molecular Fluorescence
After absorbing, a molecule relaxes to the lowest vibrational level of S₁ and may then emit (fluorescence, S₁ → S₀, nanoseconds) or cross to T₁ and emit more slowly (phosphorescence, T₁ → S₀, milliseconds to seconds). Emission occurs at longer wavelength than absorption (Stokes shift).
Intensity and concentration
$$F = k\,\phi\,P_0\left(1 - 10^{-\varepsilon b c}\right) \approx 2.303\,k\,\phi\,P_0\,\varepsilon\,b\,c \quad (\text{when } A < 0.05)$$φ is the fluorescence quantum yield. Points that follow:
- Fluorescence is linear in c only at low absorbance. At higher c the excitation beam is absorbed before it reaches the viewed region (inner-filter effect) and the plot bends over.
- Signal is proportional to \( P_0 \): a stronger lamp gives more signal. This, and measuring against a dark background rather than a small difference between two large signals, is why fluorimetry detects concentrations 10 to 1000 times lower than absorption spectrophotometry.
- Fluorescence is favoured by rigid, planar, conjugated structures (fluorescein, quinine) and by electron-donating substituents; heavy atoms (Br, I) and paramagnetic species (O₂) promote intersystem crossing and quench fluorescence. Inorganic analytes are measured through fluorescent chelates (Al³⁺ with morin or 8-hydroxyquinoline-type ligands).
Quenching: the Stern–Volmer equation
For dynamic (collisional) quenching by Q:
$$\boxed{\frac{F_0}{F} = 1 + K_{SV}[\text{Q}] = 1 + k_q\,\tau_0\,[\text{Q}]}$$\( \tau_0 \) is the lifetime without quencher and \( k_q \) the bimolecular quenching constant. A plot of \( F_0/F \) against [Q] is a straight line of slope \( K_{SV} \).
| Test | Dynamic (collisional) | Static (ground-state complex) |
|---|---|---|
| Lifetime | decreases with [Q]: \( \tau_0/\tau = F_0/F \) | unchanged |
| Temperature rise | more quenching (faster diffusion) | less quenching (complex dissociates) |
| Absorption spectrum of fluorophore | unchanged | may change |
A fitted \( k_q \) far larger than the diffusion limit (of the order of 10¹⁰ L mol⁻¹ s⁻¹ in water) shows that collisional quenching alone cannot explain the data: some static quenching or binding is present.
6. Other Optical Methods in One Table
| Method | Measures | Use |
|---|---|---|
| Turbidimetry | decrease in transmitted light due to scattering by suspended particles | sulfate as BaSO₄ suspension |
| Nephelometry | scattered light at 90° | very dilute suspensions |
| Chemiluminescence | light from a chemical reaction, no lamp | luminol (with H₂O₂ and a catalyst), NO with O₃ |
| X-ray fluorescence | characteristic X-rays after inner-shell ionisation | non-destructive elemental analysis of solids |
Structural spectroscopies (IR, NMR, EPR, Mössbauer, mass spectrometry) are separate units of the Chemical Sciences syllabus; here we treat spectroscopy as a tool for how much, not what structure.
Memory Tricks
Don’t Confuse
| This | vs | That |
|---|---|---|
| Absorbance A = −log T (no units) | transmittance T (fraction) or %T | |
| Molar absorption coefficient ε (L mol⁻¹ cm⁻¹) | specific absorptivity (L g⁻¹ cm⁻¹, mass-based) | |
| Chemical deviation (equilibrium shift on dilution) | instrumental deviation (stray light, polychromatic light) | |
| Releasing agent (binds the interferent) | protective agent (binds the analyte) | |
| Ionisation suppressor (K, Cs added in excess) | releasing agent (La, Sr) | |
| Standard addition: corrects matrix effects on slope | internal standard: corrects instrument fluctuations by signal ratio | |
| AAS: element lamp, ground-state atoms | flame emission: no lamp, excited atoms | |
| Dynamic quenching: τ falls | static quenching: τ unchanged | |
| Fluorescence (S₁ → S₀, ns) | phosphorescence (T₁ → S₀, ms to s) |
Common Mistakes
Practice
All questions below are original practice questions written for this page. They are not from any CSIR NET paper.
Level 1: Foundation
- What absorbance corresponds to 1% transmittance?
- Name the light source used in AAS and say why a continuum source is unsuitable on its own.
- In a Job’s plot, the maximum lies at mole fraction of ligand 0.75. What is the stoichiometry?
Answers
- A = −log(0.01) = 2.00.
- A hollow-cathode lamp of the analyte element. Atomic absorption lines are far narrower than the bandpass of a monochromator, so a continuum source would be absorbed only in a tiny fraction of the band and the absorbance would be very small and non-linear.
- ML₃, since \( n/(n+1) = 0.75 \) gives n = 3.
Level 2: Part B Style
Practice Q1. A solution with \( \varepsilon = 1.2 \times 10^4 \) L mol⁻¹ cm⁻¹ shows A = 0.45 in a 1 cm cell. Its concentration and percentage transmittance are:
(a) 3.75 × 10⁻⁵ M, 35.5% (b) 3.75 × 10⁻⁵ M, 64.5% (c) 5.4 × 10⁻³ M, 35.5% (d) 1.63 × 10⁻⁵ M, 35.5%
Answer: (a)
\( c = 0.45/(1.2 \times 10^4 \times 1) = 3.75 \times 10^{-5} \) M. \( T = 10^{-0.45} = 0.355 \). Option (b) gives the fraction absorbed; (d) uses ln instead of log (0.45/2.303 ÷ 1.2 × 10⁴ ≈ 1.6 × 10⁻⁵ M).
Practice Q2. In the AAS determination of calcium in a phosphate-rich sample, adding LaCl₃ to all solutions:
(a) suppresses ionisation of calcium (b) binds phosphate preferentially and releases calcium for atomisation (c) corrects for background absorption (d) increases flame temperature
Answer: (b)
La³⁺ is a releasing agent: it forms a more stable phosphate than Ca²⁺ does, freeing calcium to atomise. Ionisation suppression (a) needs K or Cs; background correction (c) needs a D₂ lamp or Zeeman system.
Practice Q3. Readings for a single standard addition: 10.0 mL sample diluted to 50.0 mL, A = 0.240; 10.0 mL sample + 5.00 mL of 10.0 ppm standard diluted to 50.0 mL, A = 0.400. The analyte concentration in the original sample is:
(a) 1.50 ppm (b) 5.00 ppm (c) 7.50 ppm (d) 15.0 ppm
Answer: (c) 7.50 ppm
\( 0.400/0.240 = (10.0c_x + 50.0)/(10.0c_x) \Rightarrow 16.67c_x = 10.0c_x + 50.0 \Rightarrow c_x = 7.50 \) ppm. Option (a) is the concentration in the 50 mL flask (7.50 × 10/50).
Practice Q4. Which statement about flame atomic emission compared with AAS is correct?
(a) Emission is less sensitive to flame temperature because it measures ground-state atoms. (b) Emission needs a separate lamp for each element. (c) A small rise in flame temperature changes the emission signal far more than the absorption signal. (d) Alkali metals emit weakly because their excited states lie high.
Answer: (c)
The excited population depends exponentially on T while the ground-state population is almost all the atoms. (a) and (b) describe AAS backwards; (d) is false because alkali-metal resonance lines lie in the visible and are easily excited.
Level 3: Part C Style
Practice Q5. A mixture of X and Y in a 1.00 cm cell gives A = 0.530 at λ₁ and A = 0.420 at λ₂. Molar absorption coefficients (L mol⁻¹ cm⁻¹): X, 8000 at λ₁ and 2000 at λ₂; Y, 1000 at λ₁ and 6000 at λ₂. The concentrations are:
(a) [X] = 6.0 × 10⁻⁵ M, [Y] = 5.0 × 10⁻⁵ M (b) [X] = 5.0 × 10⁻⁵ M, [Y] = 6.0 × 10⁻⁵ M (c) [X] = 6.6 × 10⁻⁵ M, [Y] = 7.0 × 10⁻⁵ M (d) [X] = 6.0 × 10⁻⁵ M, [Y] = 7.0 × 10⁻⁵ M
Answer: (a)
Multiply the first by 6: \( 48000x + 6000y = 3.180 \). Subtract the second: \( 46000x = 2.760 \), so \( x = 6.0 \times 10^{-5} \) M. Then \( 1000y = 0.530 - 0.480 = 0.050 \), \( y = 5.0 \times 10^{-5} \) M.
Check: \( 2000(6.0 \times 10^{-5}) + 6000(5.0 \times 10^{-5}) = 0.120 + 0.300 = 0.420 \). Option (c) treats each wavelength as if only one component absorbed there (0.530/8000 and 0.420/6000), ignoring the overlap.
Practice Q6. For the sodium 589 nm line (\( g_j/g_0 = 3 \)), consider the statements:
- At 2500 K the ratio \( N_j/N_0 \) is about 1.7 × 10⁻⁴.
- Raising the flame from 2500 K to 2600 K raises the emission intensity by roughly 45%.
- The same rise changes the AAS absorbance by roughly 45%.
- An ICP source would give a larger excited-state fraction than the flame.
Which are correct? (Take h = 6.626 × 10⁻³⁴ J s, c = 2.998 × 10⁸ m s⁻¹, k = 1.381 × 10⁻²³ J K⁻¹.)
(a) 1, 2 and 4 only (b) 1 and 3 only (c) 2 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 4 only
ΔE = hc/λ = 3.37 × 10⁻¹⁹ J.
- 2500 K: ΔE/kT = 9.77; \( 3e^{-9.77} = 1.71 \times 10^{-4} \). Statement 1 true.
- 2600 K: ΔE/kT = 9.39; \( 3e^{-9.39} = 2.50 \times 10^{-4} \). Ratio 2.50/1.71 = 1.46. Statement 2 true.
- N₀ changes from 0.99983 to 0.99975 of the atoms: the absorbance is essentially unchanged. Statement 3 false. (Atomisation efficiency could still change, but not through the Boltzmann factor.)
- The plasma is much hotter, so the exponential factor is much larger. Statement 4 true.
Practice Q7. The fluorescence of a dye (lifetime τ₀ = 5.0 ns in the absence of quencher) falls to F₀/F = 2.5 when 0.010 M of an iodide salt is added. The Stern–Volmer plot is linear. Which conclusions follow?
- K_SV = 150 L mol⁻¹.
- If quenching were purely dynamic, k_q would be 3.0 × 10¹⁰ L mol⁻¹ s⁻¹.
- This k_q exceeds the typical diffusion limit in water, suggesting a static (binding) contribution.
- Purely static quenching would reduce the measured lifetime to 2.0 ns.
(a) 1, 2 and 3 only (b) 1 and 4 only (c) 2, 3 and 4 only (d) 1, 2, 3 and 4
Answer: (a) 1, 2 and 3 only
- \( K_{SV} = (2.5 - 1)/0.010 = 150 \) L mol⁻¹. True.
- \( k_q = K_{SV}/\tau_0 = 150/(5.0 \times 10^{-9}) = 3.0 \times 10^{10} \) L mol⁻¹ s⁻¹. True.
- This is above the order-of-magnitude diffusion limit of about 10¹⁰ in water, so collisions alone cannot account for all the quenching. True (a lifetime measurement would settle it).
- Static quenching leaves the lifetime of the uncomplexed fluorophore unchanged at 5.0 ns. A lifetime of 2.0 ns (= 5.0/2.5) would be the purely dynamic case. False.
Quick Revision Table
| Idea | Formula or fact |
|---|---|
| Beer–Lambert | A = −log T = εbc; additive over components |
| Conversions | T 10% → A 1; T 1% → A 2; T 50% → A 0.301 |
| Best precision | A = 0.434 (T = 36.8%) for constant ΔT; work at 0.2–0.8 |
| Deviations | real (high c), chemical (equilibria), instrumental (polychromatic, stray light) |
| Stray light | A’ = −log[(T + s)/(1 + s)]; true 2.00 with 1% stray reads 1.70 |
| LOD | 3s_blank/slope; LOQ uses 10s |
| Standard addition | spike the sample; corrects matrix (slope) effects |
| Job’s method | max at x_L = n/(n + 1) |
| AAS source | hollow-cathode lamp of the analyte |
| AAS interferences | chemical: La/Sr releasing, EDTA protecting; ionisation: K/Cs suppressor; background: D₂, Zeeman |
| Boltzmann | N_j/N_0 = (g_j/g_0) e^(−ΔE/kT); Na 589 nm at 2500 K ≈ 1.7 × 10⁻⁴ |
| Flame photometry | Na 589, K 766, Li 671 nm |
| ICP | Ar plasma, very hot; multi-element, few chemical interferences, many spectral lines |
| Fluorescence | F ∝ φP₀εbc at A < 0.05; inner-filter at high c |
| Stern–Volmer | F₀/F = 1 + K_SV[Q]; K_SV = k_q τ₀; dynamic lowers τ, static does not |
Key Takeaways
What’s Next
- Electroanalytical Methods: concentration from potential, current and charge instead of light.
- Thermoanalytical Methods: concentration and composition from mass and heat.
- Back to Separation Methods for how these detectors sit at the end of a column.
- Reading calibration plots and straight-line fits: Graphs of Functions and Scientific Plots.