Analytical instruments

In short

Analytical instruments measure the composition of a sample - which species are present and how much - instead of a physical quantity like temperature or pressure. They work by exciting the sample with light, heat, an electric field or a carrier gas and measuring a response that depends on concentration: absorbance, electrode potential, conductivity, thermal conductivity or ion mass. In Indian industry they run refinery and fertiliser process control, continuous emission monitoring for CPCB/SPCB consent conditions, drinking-water testing to IS 10500, pharmaceutical QC under Schedule M, and food testing under FSSAI. The topic carries steady marks in Instrumentation and Measurement papers and in GATE IN (analytical instrumentation, spectroscopy, chromatography, pH and conductivity), and is asked at interviews for Thermax, Forbes Marshall, ABB, Emerson, Yokogawa, Siemens, Agilent, Shimadzu, Thermo Fisher, Chemtrols and the public-sector refineries.

Explained

Classification and the general measurement chain

• Classification by principle: (1) optical/spectroscopic - UV-visible, IR, NDIR, flame photometry, AAS, turbidimetry; (2) electrochemical - pH, ORP, conductivity, ion-selective electrodes, polarographic/galvanic oxygen, zirconia; (3) separation - gas and liquid chromatography, mass spectrometry; (4) thermal and physical - thermal conductivity analyser, paramagnetic oxygen, calorimeter, viscometer, density, humidity.
• Classification by use: laboratory (batch samples, highest accuracy, offline) versus on-line/process analysers (continuous, in a loop, ruggedised) versus portable/field kits.
• General chain: sample → sampling system (probe, filter, cooler, condensate trap, pressure/flow regulator) → detector/transducer → signal conditioning → readout, 4-20 mA / Modbus / HART to DCS.
• Sampling is the weakest link: a representative, unaltered sample at the right temperature, pressure, flow and dryness matters more than detector accuracy. Lag time = transport delay in the sample line, often tens of seconds, and it limits closed-loop control.
• Extractive versus in-situ: extractive analysers draw a sample out and condition it (easier maintenance, slower); in-situ analysers measure across the duct or in the pipe (fast, no sample loss, harder to calibrate).
• Performance terms: selectivity (freedom from interference), sensitivity (slope of output per unit concentration), detection limit (usually the concentration giving a signal of 3 × noise), linearity, drift, response time T90, resolution.
• Calibration: zero with a species-free gas or blank; span with a certified standard; two-point for linear instruments, multi-point for non-linear. Standards must be traceable - in India to NPL India, through NABL-accredited calibration laboratories (ISO/IEC 17025).
• Concentration units: ppm, ppb, % v/v, mg/Nm³ (emissions, normalised to 0 °C, 101.325 kPa, usually dry and at a reference O₂), mg/L, mol/L, pH units, NTU.

Electrochemical analysers - pH, ORP, conductivity, ISE, oxygen

• pH = −log₁₀ aH⁺, the negative logarithm of hydrogen-ion activity; the scale runs 0-14 at 25 °C, neutral = 7 (because Kw = 10⁻¹⁴).
• pH measuring cell: a glass electrode (thin pH-sensitive glass bulb enclosing a buffer of fixed pH and an internal Ag/AgCl wire) plus a reference electrode (Ag/AgCl in saturated KCl with a liquid junction). The potential developed across the glass membrane follows the Nernst equation.
• Nernst slope = 2.303RT/F ≈ 59.16 mV per pH unit at 25 °C, rising to about 64.1 mV/pH at 50 °C - hence automatic temperature compensation (ATC) with a built-in Pt100/Pt1000 or thermistor.
• Isopotential point ≈ pH 7, where output is near 0 mV and temperature has least effect; calibration uses standard buffers of pH 4.01, 7.00 and 9.18 (or 10.01).
• Because the glass membrane resistance is 10-1000 MΩ, the meter needs an electrometer amplifier with input impedance above 10¹² Ω, guarded/coaxial cable and short leads.
• Errors: alkaline (sodium) error - reads low above pH ~11 in high-Na⁺ solutions; acid error below about pH 1, where the electrode reads high; coating and junction clogging; asymmetry potential drift. ORP (redox) uses an inert Pt or Au electrode against the same reference and reads mV - used for chlorination, cyanide destruction and ETP control.
• Conductivity: measures total dissolved ions. A cell of two electrodes has cell constant K = L/A (cm⁻¹); conductivity κ = G × K. AC excitation (usually a few kHz) avoids polarisation and electrolysis; toroidal (inductive, electrodeless) cells suit dirty, corrosive liquids. Conductivity rises roughly 2% per °C, so readings are normalised to 25 °C. Ultrapure water at 25 °C has κ ≈ 0.055 µS/cm (resistivity 18.2 MΩ·cm) - the basis of boiler-feed and DM-plant monitoring.
• Ion-selective electrodes (ISE): membranes selective to F⁻, Cl⁻, NO₃⁻, Na⁺, NH₄⁺, Ca²⁺ etc.; same Nernstian behaviour, slope 59.16/z mV per decade; selectivity quantified by the Nikolsky-Eisenman selectivity coefficient.
• Dissolved oxygen: polarographic (Clark) cell - cathode held at about −0.8 V, O₂ diffuses through a PTFE membrane and is reduced, current ∝ pO₂; galvanic cell needs no polarising voltage; optical/luminescence (fluorescence quenching) sensors need no membrane or electrolyte and have largely replaced them in ETP/STP work.
• Zirconia oxygen analyser: a ZrO₂ (yttria-stabilised) tube at 600-850 °C becomes an oxygen-ion conductor; the Nernst potential between flue gas and reference air gives E = (RT/4F)·ln(p₁/p₂) ≈ 0.0496·T·log(p_ref/p_meas) mV, roughly 50 mV per decade of O₂ ratio at ~1100 K. In-situ, fast, used for combustion/excess-air control on boilers and furnaces; it is destroyed by unburnt hydrocarbons and reducing gases.
• Paramagnetic oxygen analyser: O₂ is strongly paramagnetic, so a dumb-bell suspended in a non-uniform magnetic field is deflected in proportion to O₂ - no consumables, used where a zirconia cell would be damaged.

Optical and spectroscopic analysers

• Electromagnetic spectrum used: UV 190-400 nm (electronic transitions), visible 400-800 nm, near-IR 0.8-2.5 µm (overtones), mid-IR 2.5-25 µm (molecular vibration - the "fingerprint" region).
• Beer-Lambert law: absorbance A = log₁₀(I₀/I) = εcl - linear only for dilute, monochromatic, non-scattering, non-associating samples.
• Spectrophotometer block diagram: source → monochromator (prism or diffraction grating, with entrance and exit slits) → sample cuvette → detector → amplifier → readout.
  • Sources: tungsten-halogen (visible/NIR), deuterium or hydrogen discharge lamp (UV), Nernst glower / globar / nichrome coil (IR).
  • Detectors: photomultiplier tube, silicon photodiode and photodiode array (UV-Vis); thermopile, bolometer, pyroelectric (DTGS) or MCT (IR).
  • Single-beam is simple but drifts with source and electronics; double-beam splits light between sample and reference and ratios them, cancelling drift - the standard for laboratory UV-Vis and IR.
• NDIR gas analyser: a broadband IR source, chopper, sample and reference cells and a filter or Luft-type pneumatic detector. Each gas absorbs at a characteristic band - CO₂ at 4.26 µm, CO at 4.6 µm, CH₄ at 3.3 µm, SO₂ at 7.3 µm, NO at 5.3 µm. Homonuclear diatomics (O₂, N₂, H₂, Cl₂) have no IR absorption and cannot be measured by IR. Water vapour interferes broadly - hence gas coolers and interference filters.
• FTIR replaces the monochromator with an interferometer and Fourier-transforms the interferogram: whole spectrum at once, high throughput and resolution; multi-component stack and ambient monitoring.
• UV absorption/fluorescence analysers: SO₂ by UV fluorescence, NOx by chemiluminescence (NO + O₃ → NO₂* → light at ~600 nm, with a converter to reduce NO₂ to NO for total NOx), O₃ by UV absorption at 254 nm - the reference methods in CPCB's CAAQMS stations.
• Flame photometry (emission): the sample is aspirated into a flame; alkali and alkaline-earth metals emit at characteristic wavelengths - Na 589 nm, K 766 nm, Li 670 nm, Ca 422 nm - intensity ∝ concentration. Cheap, limited to easily excited metals.
• Atomic absorption spectroscopy (AAS): a hollow-cathode lamp of the element shines through an air-acetylene flame or graphite furnace atomising the sample; absorption of the resonance line measures concentration to ppb. Used for heavy metals (Pb, Cd, Cr, As, Hg) in water and food testing; ICP-OES/ICP-MS are the higher-throughput modern alternatives.
• Turbidity: nephelometric measurement of light scattered at 90° to the beam, reported in NTU; IS 10500:2012 sets drinking water turbidity at 1 NTU (acceptable), relaxable to 5 NTU. Opacity/dust in stacks is measured by transmissometer or by light-scattering dust monitors.

Chromatography and mass spectrometry

• Principle: a mixture carried by a mobile phase over a stationary phase separates because each component partitions differently; the time from injection to peak maximum is the retention time tR (qualitative), and peak area is quantitative.
• Gas chromatograph (GC) blocks: carrier gas (N₂, He, H₂ at controlled pressure) → injector (split/splitless, gas sampling valve on process GCs) → column in a temperature-programmed oven (packed, or capillary 0.1-0.5 mm × 15-60 m) → detector → integrator.
• GC detectors: TCD (thermal conductivity - universal, non-destructive, moderate sensitivity, uses a Wheatstone bridge of hot filaments); FID (flame ionisation - hydrocarbons, wide linear range, destructive); ECD (electron capture - halogens, pesticides, ppb level); FPD (sulphur/phosphorus); PID; MS (identification).
• Column efficiency: number of theoretical plates N = 16(tR/W)² for a Gaussian peak of base width W, or 5.54(tR/W½)²; HETP = L/N; the van Deemter equation HETP = A + B/u + Cu gives an optimum carrier velocity. Resolution R = 2(tR2 − tR1)/(W1 + W2); R ≥ 1.5 is baseline separation.
• HPLC: liquid mobile phase at 100-400 bar through a packed column (C18 reverse phase is commonest), with UV-Vis/DAD, fluorescence, RI or MS detection; isocratic versus gradient elution. The workhorse of Indian pharmaceutical QC and stability testing.
• Process (on-line) GC: automated sample valves, fixed oven programme, cycle time of minutes; used for refinery column control, natural-gas composition and calorific value.
• Mass spectrometer: inlet → ion source (electron impact, chemical ionisation, ESI for LC-MS) → analyser (magnetic sector, quadrupole, time-of-flight, ion trap) → detector (electron multiplier) → mass spectrum of abundance versus m/z. In a magnetic sector, r = (1/B)√(2Vm/q) - heavier ions travel a larger radius. Needs high vacuum (10⁻⁵-10⁻⁷ mbar). GC-MS and LC-MS give both separation and identification - residue, forensic and pollution work.
• Qualitative versus quantitative: retention time and mass spectrum identify; calibration by external standard, internal standard or standard addition quantifies.

Other process analysers and physical-property instruments

• Thermal conductivity (katharometer) analyser: measures a binary or pseudo-binary mixture from the gas's thermal conductivity - H₂ (very high k) in purge gas, CO₂ in flue gas, Ar/He purity.
• Humidity: psychrometer (dry- and wet-bulb), hair/electrical hygrometer, capacitive thin-film polymer sensor (standard in HVAC and labs), chilled-mirror dew-point meter (reference quality), aluminium-oxide and Karl Fischer methods for trace moisture in gases and oils.
• Gas density, specific gravity and calorific value: by vibrating-element densitometer, Wobbe-index meter and calorimeter - billing and burner control in city-gas distribution.
• Viscosity: capillary (Ostwald/Ubbelohde), falling-sphere, rotational (Brookfield) and on-line vibrating-rod viscometers; used in lube-oil blending, paints and polymer plants.
• Water-quality package (ETP/STP/CAAQMS): pH, conductivity/TDS, DO, turbidity, TSS, COD/BOD (COD by dichromate reflux or on-line UV-254 surrogate), ammoniacal nitrogen, residual chlorine, oil-in-water, TOC (by UV/persulphate or high-temperature catalytic oxidation with NDIR CO₂ detection).
• CEMS (continuous emission monitoring system): probe with heated filter and heated line, gas cooler, pump, NDIR/UV bench for SO₂, NOx, CO, CO₂, zirconia or paramagnetic O₂, dust monitor, flow and temperature, plus a data logger transmitting to the SPCB/CPCB server. Mandated by CPCB directions for 17 categories of highly polluting industries and CETPs; readings go on the OCEMS portal.
• Safety and area classification: flammable-sample analysers sit in purged analyser houses; gas detectors use catalytic bead (LEL), IR (hydrocarbons) and electrochemical cells (H₂S, CO), with enclosures certified per IS/IEC 60079 series and Petroleum and Explosives Safety Organisation (PESO) approval.

Indian standards, statutes and practice

• Water: IS 10500:2012 (drinking water specification, with amendments), IS 3025 series (methods of test for water and wastewater), IS 1622 (bacteriological). Effluent discharge standards are notified under the Environment (Protection) Rules, 1986; consent is granted under the Water (Prevention and Control of Pollution) Act, 1974 and Air (Prevention and Control of Pollution) Act, 1981, administered by CPCB and the State Boards.
• Air: National Ambient Air Quality Standards (CPCB, 2009) - PM₁₀ 100 µg/m³ and PM₂.₅ 60 µg/m³ as 24-hour values, SO₂ 80 µg/m³, NO₂ 80 µg/m³; measured in CAAQMS/NAMP stations. Stack limits are industry-specific and expressed in mg/Nm³ at a stated reference O₂.
• Calibration and quality systems: ISO/IEC 17025 accreditation by NABL for test and calibration laboratories; ISO 9001; pharmaceutical analytical instruments are qualified (DQ-IQ-OQ-PQ) and validated under Schedule M of the Drugs and Cosmetics Rules and the Indian Pharmacopoeia; legal metrology instruments come under the Legal Metrology Act, 2009.
• Food and fuel: FSSAI laboratories use AAS, GC, HPLC and LC-MS for contaminants and adulterants; BIS fuel specifications (for example IS 1460 for diesel) are checked by distillation, GC and sulphur analysers.
• Workplace: occupational exposure monitoring and workplace safety duties now sit under the Occupational Safety, Health and Working Conditions Code, 2020, one of the four Labour Codes in force from 21 November 2025, which replaced the Factories Act, 1948 and the BOCW Act.
• Many older instrument-specific IS numbers have been withdrawn or superseded; where the number is in doubt, cite the subject (for example "BIS method for water turbidity in the IS 3025 series") rather than a number.

Key points

• Analytical instrument - measures composition (what and how much), not a physical variable.
• Sampling system - probe, filter, cooler and regulator; most analyser faults and all transport lag originate here.
• Beer-Lambert law - A = εcl; valid only for dilute, monochromatic, non-scattering samples.
• Double-beam - sample and reference beams ratioed, cancelling source and detector drift.
• NDIR - selective IR absorption for CO, CO₂, CH₄, SO₂; blind to O₂, N₂, H₂ (no IR-active dipole change).
• Nernst slope - 59.16 mV per pH unit at 25 °C; temperature compensation is mandatory.
• Glass electrode - very high impedance (MΩ), so it needs an electrometer amplifier above 10¹² Ω.
• Alkaline error - glass electrode reads low above about pH 11 in high-sodium solutions.
• Cell constant - K = L/A in cm⁻¹; conductivity κ = G × K, referenced to 25 °C (~2%/°C).
• Zirconia cell - in-situ, hot (600-850 °C) O₂ measurement for excess-air control; damaged by reducing gases.
• Paramagnetic analyser - exploits O₂'s paramagnetism; no consumables, no damage from hydrocarbons.
• Chromatography - retention time identifies, peak area quantifies.
• GC detectors - TCD universal and non-destructive; FID for hydrocarbons; ECD for halogens at ppb.
• Plate count - N = 16(tR/W)²; HETP = L/N; van Deemter gives the optimum carrier velocity.
• Resolution - R = 2ΔtR/(W1 + W2); R ≥ 1.5 means baseline separation.
• Mass spectrometer - separates ions by m/z under high vacuum; GC-MS adds identification to separation.
• Detection limit - concentration giving a signal of about three times the noise.
• CEMS/OCEMS - continuous stack monitoring with data to the Pollution Control Board, in mg/Nm³ at a reference O₂.
• Traceability - calibration standards traced to NPL India through NABL (ISO/IEC 17025) laboratories.

Formulas and examples

Example 1 - Beer-Lambert law and transmittance. A solution in a 1 cm cell transmits 25% of the incident light at 540 nm; ε = 6,000 L·mol⁻¹·cm⁻¹. Find the concentration, and the absorbance if the path is increased to 4 cm.

• A = −log₁₀(0.25) = log₁₀ 4 = 0.6021.
• c = A/(εl) = 0.6021/(6,000 × 1) = 1.0035 × 10⁻⁴ mol/L ≈ 100 µmol/L.
• At l = 4 cm: A = 0.6021 × 4 = 2.408, so T = 10⁻²·⁴⁰⁸ = 0.0039 (0.39%) - beyond the reliable range of most single-beam instruments, which is why a shorter path or dilution is preferred above A ≈ 2.

Example 2 - pH electrode slope and temperature error. A pH meter calibrated at 25 °C (59.16 mV/pH, isopotential pH 7) is used without ATC on a sample at 80 °C whose true pH is 10.00.

• True slope at 80 °C: S = 0.1984 × 353.15 = 70.05 mV/pH.
• Electrode output = −70.05 × (10.00 − 7.00) = −210.2 mV.
• The meter divides by its 25 °C slope: indicated pH = 7 + 210.2/59.16 = 7 + 3.553 = 10.55.
• Error = +0.55 pH - uncompensated temperature error grows in proportion to the distance from the isopotential point.

Example 3 - Conductivity cell and cell constant. A cell has electrodes of area 0.80 cm² spaced 1.60 cm apart; immersed in a sample it reads 4.50 kΩ.

• K = L/A = 1.60/0.80 = 2.0 cm⁻¹.
• κ = K/R = 2.0/4,500 = 4.444 × 10⁻⁴ S/cm = 444.4 µS/cm.
• If the reading was taken at 30 °C, correcting to 25 °C with 2%/°C: κ₂₅ = 444.4/[1 + 0.02(30 − 25)] = 444.4/1.10 = 404 µS/cm.
• Approximate TDS at f = 0.6: 0.6 × 404 ≈ 242 mg/L - within the IS 10500:2012 acceptable limit of 500 mg/L for TDS.

Example 4 - GC column efficiency and resolution. On a 25 m capillary column, peak 1 has tR = 4.00 min with base width 0.20 min and peak 2 has tR = 4.50 min with base width 0.25 min.

• N₁ = 16(4.00/0.20)² = 16 × 400 = 6,400 plates; HETP = 25,000 mm/6,400 = 3.91 mm.
• R = 2(4.50 − 4.00)/(0.20 + 0.25) = 1.00/0.45 = 2.22 → well above 1.5, so the peaks are baseline-resolved.
• If the column is shortened to 10 m with the same HETP, N = 10,000/3.91 ≈ 2,560; resolution scales as √N, so R falls to 2.22 × √(2,560/6,400) = 1.40 - no longer baseline separation.

Example 5 - Stack SO₂ corrected to reference oxygen. A CEMS reads SO₂ = 420 mg/Nm³ (dry) at a measured 9.0% O₂; the consent condition is in mg/Nm³ at 6% O₂ reference.

• Corrected = 420 × (21 − 6)/(21 − 9.0) = 420 × 15/12 = 525 mg/Nm³ at 6% O₂.
• Reporting the uncorrected 420 would understate the emission by 20%; dilution air cannot be used to pass a limit expressed at a reference O₂.

Where it is used

• Refineries and petrochemicals (IOCL, BPCL, HPCL, RIL, GAIL): on-line process GCs for distillation-column and reformer control, H₂S and sulphur analysers, O₂ trim on furnaces, analyser houses maintained by instrumentation engineers.
• Power and cement (NTPC, Tata Power, UltraTech, ACC): zirconia O₂ for combustion efficiency, boiler-feedwater pH/conductivity/silica/dissolved-oxygen analysers, CEMS reporting to the SPCB on the OCEMS portal.
• Water and wastewater (Jal Jeevan Mission, municipal STPs, CETPs at Tarapur, Ankleshwar, Tiruppur): pH, DO, turbidity, residual chlorine, on-line COD/TOC and flow for consent compliance.
• Pharmaceuticals and biotech (Sun Pharma, Cipla, Dr Reddy's, Biocon, Serum Institute): HPLC, UV-Vis, FTIR, GC headspace for residual solvents, dissolution testers - all under qualification and data-integrity requirements.
• Food, agriculture and environment labs (FSSAI referral labs, NABL labs, CPCB CAAQMS network): AAS/ICP-MS for heavy metals, GC-MS for pesticide residues, continuous ambient PM, SO₂, NOx and ozone analysers.
• Job roles: process analyser engineer, instrumentation and control engineer in plant maintenance, applications/service engineer with ABB, Emerson, Yokogawa, Siemens, Thermax, Forbes Marshall, Agilent, Shimadzu, Thermo Fisher; QC analyst and calibration engineer in NABL laboratories.

Interview questions

• Q: What distinguishes an analytical instrument from other measuring instruments? A: It measures the chemical composition or a composition-dependent property of a sample, whereas conventional instruments measure physical variables such as temperature, pressure or flow.
• Q: State the Beer-Lambert law and two reasons it may fail. A: A = εcl. It fails at high concentration (solute-solute interaction and refractive-index change) and with polychromatic light, stray light, scattering or chemical association/dissociation of the absorbing species.
• Q: Why is a double-beam spectrophotometer preferred for precise work? A: It continuously compares sample and reference beams, so source intensity drift, detector drift and cell/solvent absorption cancel out in the ratio.
• Q: Why can an NDIR analyser not measure oxygen? A: O₂ is a homonuclear diatomic molecule with no change of dipole moment during vibration, so it has no infrared absorption band; O₂ needs a zirconia, paramagnetic or electrochemical analyser.
• Q: What is the Nernst slope of a pH electrode and why does temperature compensation matter? A: 2.303RT/F ≈ 59.16 mV per pH unit at 25 °C; the slope rises about 0.2 mV/pH per °C, so without compensation a hot or cold sample gives an error proportional to the distance from the isopotential point (about pH 7).
• Q: Why does a pH meter need an extremely high input impedance? A: The glass membrane has a resistance of 10 MΩ to 1,000 MΩ; any appreciable input current would load the cell and produce a large error, so the amplifier must present above 10¹² Ω.
• Q: What is alkaline error? A: Above about pH 11, in solutions high in sodium, the glass membrane responds partly to Na⁺ as well as H⁺, so the meter reads lower than the true pH; lithium-glass "high-pH" electrodes reduce it.
• Q: Why is AC excitation used in a conductivity cell? A: DC would polarise the electrodes and electrolyse the sample, producing a drifting, false reading; a few-kilohertz AC avoids both.
• Q: In chromatography, what do retention time and peak area tell you? A: Retention time identifies the component (qualitative) under fixed conditions; peak area, after calibration, gives its quantity (quantitative).
• Q: Compare TCD and FID. A: A TCD responds to almost any gas whose thermal conductivity differs from the carrier, is non-destructive but only moderately sensitive; an FID is far more sensitive to hydrocarbons, has a very wide linear range, but destroys the sample and does not respond to inert or fully oxidised gases.
• Q: A process analyser reads correctly on calibration gas but lags badly in the plant - where do you look? A: The sampling system: long or oversized sample line, low sample flow, a clogged filter or probe, condensation in the line, or a dead volume - transport lag and sample distortion usually dominate over detector response.
• Q: Why are stack emissions quoted at a reference oxygen or as mg/Nm³? A: Normalising to a reference O₂ and to standard temperature, pressure and dry basis prevents a plant from appearing compliant simply by diluting the flue gas with excess air.
• Q: How is a laboratory analyser's accuracy made legally defensible in India? A: By calibration with certified standards traceable to NPL India through NABL-accredited (ISO/IEC 17025) laboratories, with documented procedures, calibration records and, in pharma, DQ-IQ-OQ-PQ qualification.

Common mistakes

• Treating absorbance as linear at any concentration, or working above A ≈ 2 where stray light and noise destroy accuracy; confusing absorbance with transmittance and the sign of the log.
• Ignoring the cell constant, so conductance (S) is reported as conductivity (S/cm), or omitting the 25 °C temperature correction.
• Quoting 59.16 mV/pH as if it were fixed, and forgetting that the slope - not the intercept - is what temperature changes.
• Calibrating a pH electrode with buffers at one temperature and measuring at another without ATC, or storing the electrode dry so the gel layer dehydrates.
• Assuming any gas can be measured by IR - forgetting that O₂, N₂, H₂ and Cl₂ are IR-inactive; or putting a zirconia probe in a reducing, hydrocarbon-bearing stream.
• Using retention time alone to identify an unknown peak without a standard or mass spectrum, and ignoring N, HETP and resolution when a separation degrades.
• Blaming the analyser for a faulty sample: unrepresentative tapping point, condensation, filter bypass and leaks are the usual cause, and the sample lag is often mistaken for slow detector response.
• Reporting a concentration without its basis - wet or dry, at what reference O₂, at what temperature and pressure - which makes an emission figure meaningless.

Related topics

Transducers and sensors basics · Signal conditioning and data acquisition · Process control and final control elements · Industrial instrumentation and plant safety · Biomedical instrumentation · Environmental monitoring and pollution-control instrumentation