Lesson 2 · 30 min

The spectrophotometer

Source, monochromator, cell, detector: a virtual instrument scans permanganate, copper ammine and dichromate. Learn to zero it, pick the analytical wavelength, and determine manganese and chromium in steel at once.

In this lesson

Name the five parts of a UV–visible spectrophotometer — source, monochromator, sample cell, detector, readout — and what each does
Explain why we zero on a blank, record a baseline and choose the cell material
Choose the analytical wavelength and keep readings in the absorbance range of least photometric error
Determine two components at once from absorbances at two wavelengths

A steelworks needs the manganese content of every melt. Dissolve 0.5 g of steel turnings in acid, oxidise all the manganese to purple permanganate with potassium periodate, and the depth of colour is proportional to the manganese. To turn “depth of colour” into a four-figure number takes an instrument, the spectrophotometer. What it does is simple: pick one wavelength out of white light, send it through the sample, and compare the light before and after.

Predict first

To measure a (purple) potassium permanganate solution, where should the monochromator be set?

Pick an answer and lock it in before you look. A wrong guess teaches more than no guess.

440 nm (violet-blue) 526 nm (green) 640 nm (red) Anywhere — absorbance does not depend on wavelength

Five parts of the instrument

Source: a tungsten–halogen lamp for the visible (about 350–1000 nm), a deuterium lamp for the ultraviolet (about 190–400 nm).
Monochromator: entrance slit, diffraction grating and exit slit. Turning the grating sends a different wavelength onto the exit slit; the slit width sets the spectral bandwidth, the range of wavelengths passed.
Sample cell (cuvette): usually 1.00 cm of path. Glass and plastic start to absorb below about 340 nm, so ultraviolet work needs quartz.
Detector: a photodiode, a photomultiplier, or a diode array that records a whole spectrum at once.
Readout: the instrument compares $P$ with the reference $P_0$ and shows %T or A.

A single-beam instrument is first zeroed on a blank — the same solvent, reagents and cell — which defines $P_0$ as the light through the blank; reflections from the cell and absorption by solvent and reagents are subtracted that way. Lamp output and detector sensitivity change with wavelength, so a new wavelength needs a new zero, and a spectrum scan starts by recording a baseline. A double-beam instrument sends the light alternately through sample and reference and subtracts continuously.

Single-beam instrument · tungsten–halogen lamp 350–900 nm · 1.00 cm cell

UV–visible spectrophotometer

Sample Concentration = 35.4 mmol/L

Wavelength λ = 526 nm Bandwidth

Show the noise-free model

Not zeroed yet: the reading includes the cell and solvent themselves.

Scanned spectrum
Noise-free model
Single readings

Take a reading before zeroing: it includes the cell and the solvent. Zero at 600 nm, move to 380 nm and see how far off you are without re-zeroing, then press “Record a full baseline”. Next compare CuSO₄ with [Cu(NH₃)₄]²⁺: with ammonia the band moves from 800 nm to about 605 nm and ε grows about fourfold — the reason copper is determined as its ammine.

Where readings are most precise

Detector noise usually appears as a roughly constant error $\sigma_T$ in transmittance. Translated into concentration, the relative error grows both when A is small (so little is absorbed that the difference is tiny) and when A is large (almost no light gets through); its minimum is at $A = \lg e \approx 0.434$ . In practice, keep absorbances between about 0.1 and 1.0; dilute a solution that is too dark, or use a shorter cell. Above A = 2 less than 1 % of the light reaches the detector and stray light and noise take over.

\frac{\sigma_c}{c} = \frac{\sigma_A}{A} = \frac{\sigma_T}{T\,\lvert\ln T\rvert}

Setting the derivative with respect to T to zero gives the best transmittance, T = 1/e = 36.8 % (A = 0.434).

$\sigma_c/c$: relative standard deviation of the concentration
$\sigma_T$: standard deviation of a transmittance reading (absolute)
$T$: transmittance

Three rules for choosing the analytical wavelength: sit on a band maximum (high sensitivity, flat top); avoid wavelengths where other components absorb; keep the slit narrower than about a tenth of the band width. Steel solutions often also contain dichromate from oxidised chromium, which absorbs strongly at 440 nm and hardly at all at 545 nm.

Worked example

Manganese and chromium together

A steel solution in a 1.00 cm cell reads A₄₄₀ = 0.405 and A₅₄₅ = 0.712. Molar absorptivities: MnO₄⁻ 95 at 440 nm and 2350 at 545 nm; Cr₂O₇²⁻ 369 at 440 nm and 11 at 545 nm (L mol⁻¹ cm⁻¹). Find both concentrations.

Concept check

When determining iron as its phenanthroline complex, what should the blank be?

An empty cell (full of air) Distilled water A solution with every reagent (hydroxylamine, phenanthroline, buffer) but no iron The most concentrated standard

Try the calculation

The transmittance noise is σ_T = 0.005 (0.5 %T). What is the relative error in concentration (%) for a sample of absorbance 1.00?

Key takeaways

Source → monochromator (slits + grating) → cell → detector → readout; the ultraviolet needs a deuterium lamp and quartz cells.
Zero on a blank containing every reagent; a single-beam instrument must be re-zeroed at each wavelength and needs a baseline before a scan.
Pick the analytical wavelength on a band maximum, away from interferences; keep A between 0.1 and 1.0, best near 0.43.
Absorbances add: two components need two wavelengths and a pair of linear equations.

Progress is stored only in this browser and never uploaded.

Data and references

D. C. Harris, Quantitative Chemical Analysis, 9th ed. (2016), ch. 18–20 (spectrophotometers, analysis of mixtures).
D. A. Skoog, F. J. Holler, S. R. Crouch, Principles of Instrumental Analysis, 7th ed. (2018), ch. 13–14 (photometric error, instrument components).
D. A. Skoog, D. M. West, F. J. Holler, S. R. Crouch, Fundamentals of Analytical Chemistry, 9th ed. (2014), ch. 26 (simultaneous determination of Mn and Cr: ε(MnO₄⁻) 95 and 2.35 × 10³, ε(Cr₂O₇²⁻) 369 and 11 at 440 and 545 nm).