Atomic Absorption Spectrometry: Flame, Graphite Furnace, Cold Vapour

Free gas-phase atoms absorb at characteristic electronic transitions. The optical measurement must isolate an appropriate analytical line and distinguish analyte absorption from other attenuation. Calibration relates the recorded response to concentration under supported preparation and atomisation conditions.

A conventional line-source instrument uses a hollow-cathode lamp or another suitable narrow-line source. In a hollow-cathode lamp, a discharge in a low-pressure fill gas sputters cathode material, producing characteristic emission. Multi-element lamps exist, and automated lamp turrets can change sources without manual replacement for each measurement.

The beam passes through the atom population and a wavelength-selecting optical system to a detector. Optical layout, modulation and detector type vary. Useful examples of analytical wavelengths include copper near 324.8 nm, lead near 283.3 nm and mercury near 253.7 nm, with the exact line and conditions selected for the method.

AAS is commonly used sequentially for individual elements, but a hollow-cathode lamp is not an unavoidable feature of all designs. High-resolution continuum-source instruments use a broad source and high-resolution spectral measurement. Analytik Jena's contrAA documentation describes a xenon short-arc source covering the relevant range.

Concentration follows calibration and specimen preparation, not wavelength alone. Different lines can offer different sensitivity or interference characteristics. A wavelength match identifies the measured transition; it does not establish the chemical form in the submitted material.

Flame AAS commonly nebulises a liquid into a burner system. Only part of the aspirated material reaches the useful observation region. Fuel–oxidant conditions, burner position and specimen properties influence atom formation. Air–acetylene and nitrous-oxide–acetylene arrangements serve different analytical needs.

Electrothermal AAS introduces a small specimen aliquot into an electrically heated atomiser, commonly a graphite tube. A programme can dry the specimen, remove some matrix, produce atoms and clean the tube. Temperatures, ramp rates, modifiers and gas flow are analyte- and matrix-specific rather than a universal recipe.

A platform can help control the relationship between specimen heating and tube conditions. Atomisation produces a transient signal, so peak measurement and background correction must suit that signal. Furnace methods often offer lower concentration detection capability than routine flame work, but no fixed hundredfold improvement applies to every element.

Cold-vapour AAS takes advantage of elemental mercury's appreciable vapour pressure near room temperature. Suitable chemistry converts mercury in the prepared specimen to elemental vapour, which is transported to an absorption cell. Other metals also have vapour pressures; the useful mercury property is not that all others have exactly zero.

Oxidation or other preparation may be needed to determine the intended mercury fraction. A reduction step alone does not necessarily release every bound or organic form. Total-mercury measurement and speciation methods answer different questions.

Hydride generation converts suitable analytes to volatile hydrides under controlled conditions, followed by transport and atomisation. Arsenic, antimony and selenium are examples. Oxidation state, pre-reduction, acid conditions and interferences affect generation efficiency. A digest cannot be assumed to respond identically regardless of its chemistry.

Detection and quantification capability depend on the full method, including dilution and specimen mass. Instrument-solution units such as µg/L should not be confused with the original blood or tissue concentration. A lower instrumental detection limit does not automatically compensate for poor recovery or contamination.

Physical effects include changes in aspiration, nebulisation and transport caused by viscosity, acid concentration or dissolved material. Matrix matching can reduce differences between standards and specimens. It should represent the relevant properties rather than merely match the nominal acid label.

Chemical effects can reduce formation of free analyte atoms. Formation of refractory calcium–phosphate material in a flame is a familiar example. A suitable releasing agent, altered atomisation conditions or another supported approach can help, but a hotter flame is not a universal remedy for all analytes.

Ionisation changes the population of neutral atoms available for a particular absorption measurement. An ionisation buffer can help stabilise relevant equilibria when appropriate. The additive and concentration follow the validated method and must be represented in standards and specimens.

Spectral attenuation includes atomic overlaps, molecular absorption and scattering. Element-specific line sources reduce some overlap problems but do not eliminate them. Review relevant interferents at realistic concentrations and examine whether the correction or alternative line is suitable.

Standard addition can compensate for a suitable multiplicative matrix effect when the added analyte behaves like the native analyte. It does not correct every nonlinear effect, unresolved spectral contribution, contamination or analyte loss before addition. Its negative intercept requires an appropriate model and consistent volume treatment.

Use matrix-relevant reference materials, preparation blanks and other controls to investigate bias. A certificate must actually provide an appropriate assigned value for the analyte and material. Not every element listed for a reference material has a certified value, and a successful recovery does not rule out all forms of interference.

Background correction estimates attenuation that is not attributable to the intended atomic absorption. It is particularly important when matrix absorption or scattering is substantial. Correction performance must be appropriate for the spectral structure and timing of the measurement.

Deuterium correction compares a broad-source response with the narrow analytical-line response under suitable conditions. It works best when the background estimate represents the background at the analyte line. Structured background or optical mismatches can limit the correction.

Zeeman approaches use magnetic-field effects on atomic transitions and suitable polarisation or modulation arrangements to distinguish analyte and background contributions. Implementations differ. They do not guarantee correction of every structured overlap or remove preparation and chemical-interference errors.

Smith–Hieftje correction uses source self-reversal under different lamp operating conditions. High-resolution continuum-source systems can use spectral information around an analytical feature for background assessment. Neither architecture makes independent quality control unnecessary.

Examine corrected and relevant uncorrected signals, warnings and control performance. A plausible final absorbance can conceal excessive or inappropriate correction. Changing correction settings to obtain an expected result without evidence would undermine the measurement.

Define whether the method measures total, total-recoverable, dissolved or another operationally defined elemental fraction. Preparation determines which material reaches the atomiser. Digestion, dilution or direct-introduction methods must be suitable for the specimen and target.

Contamination from collection devices, reagents, vessels or the laboratory can be important at trace concentrations. Use suitable blanks and handling controls. Volatile losses and incomplete dissolution can cause low results; a clear digest alone does not prove complete recovery.

Closed-vessel microwave preparation can reduce some losses and improve control, but it also requires compatible chemistry, equipment and operating limits. It is not risk-free or mandatory for every biological matrix. Some supported methods use dilution or other preparation rather than complete digestion.

Calibration standards must cover the required working range and represent relevant response conditions. Evaluate the model with residuals, back-calculated concentrations and independent controls. ISO/IEC 17025 does not prescribe one universal R², recovery range, duplicate frequency or blank-replicate count for every AAS method.

Assess selectivity, bias, precision, stability, carryover and detection or quantification capability as relevant. Verify dilution and the calculation back to the original specimen. A supported result below the quantitative range should not be converted into a precise concentration merely because a signal was detected.

Report specimen, measurand, units, method and relevant limitations. Include uncertainty where applicable under the reporting framework. A CRM result is important evidence about performance but does not alone transform an otherwise deficient result into legally conclusive evidence.

Blood lead, urinary arsenic, mercury and tissue metals illustrate different specimen–question relationships. The appropriate specimen depends on the element's kinetics, chemical form, exposure pattern and intended use. Hair and nails are vulnerable to external contamination and do not provide an exact exposure calendar from segment length alone.

In the United States, CDC currently uses a blood lead reference value of 3.5 µg/dL for young children. It identifies children with higher blood lead than most peers and supports follow-up; it is not a safe-versus-toxic boundary or a universal treatment threshold. Other countries and programmes have their own guidance.

Arsenic interpretation may require speciation because dietary organic arsenic and inorganic exposure can affect total results differently. Mercury specimen selection also depends on the form of exposure: urine is generally more relevant to elemental or inorganic exposure than to methylmercury assessment. A total result should not be assigned a form without evidence.

Detection of a nonessential element does not automatically establish poisoning. Background exposure, contamination, specimen timing and analytical capability matter. Neither a thallium detection nor a metal concentration independently establishes intent, a particular source or cause of death.

US, UK, Indian and other laboratories may use AAS, ICP-MS or other validated methods according to capability and the question. Confirm services with the actual laboratory rather than rely on an unsupported national instrument inventory. Legal conclusions remain separate from the chemical measurement.

Related reading: ICP-OES and ICP-MS: Plasma-Based Elemental Analysis.

Related reading: Sample Preparation, Purification and Instrument Calibration.

Related reading: Method Validation, Measurement Uncertainty and Proficiency Testing.