Analytic Journal is growing – your feedback shapes the platform.Send feedback
Category: Gas Analytics | Process Analytical Technology | Laser Spectroscopy | PAT | Functional Safety
Author: Dusko Kadijevic, Publisher Analytic Journal
Last updated: May 2026
Reading time: approx. 10 minutes
Target audience: Process engineers, PAT specialists, instrumentation engineers, safety managers in the process industry
TDLS (Tunable Diode Laser Spectroscopy) — also referred to as Tunable Diode Laser Absorption Spectroscopy (TDLAS) — is an optical measurement principle in which a diode laser is tuned to an absorption line characteristic of the target molecule. The laser light traverses the measurement gas, is absorbed by the target molecules to a defined extent, and the gas concentration is calculated from the resulting attenuation of the signal. Measurement occurs without sample extraction, without media contact and without consumables.
TDLS is not the only laser-based gas measurement technique — but it is the most widely deployed for direct in-situ measurement in industrial processes. No other measurement principle combines high selectivity, low-maintenance operation and the ability to measure under the harshest process conditions as convincingly.
Over the past 30 years, the process industry's confidence in TDLS systems has grown to the point where they are increasingly used in safety-relevant applications — including SIL-classified measurement loops. This is not self-evident for a principle that was considered technically demanding and difficult to control only two decades ago.
A diode laser generates monochromatic light — light of a single, highly precise wavelength. This wavelength is selected to coincide exactly with an absorption line of the target molecule. The gas absorbs light at exactly this wavelength — the more target molecules in the light path, the more light is absorbed.
The laser is swept slightly across a narrow wavelength range — hence the term "tunable". On the receiver side, a detector measures how strongly the light has been attenuated. From this attenuation, the known path length and the process parameters of pressure and temperature, the gas concentration is calculated.
| Property | TDLS | NDIR | FTIR |
|---|---|---|---|
| Light source | Diode laser — narrowband, monochromatic | Broadband IR source | Broadband IR source |
| Selectivity | Very high — single absorption line | Medium — bandpass filter | High — full spectrum |
| Cross-sensitivity | Very low with correct line selection | Present | Matrix-dependent |
| Sample conditioning | Not required (in-situ) | Required | Often required |
| Measurement speed | Seconds | Seconds | Seconds to minutes |
| Pressure dependence | Present — compensable | Low | Low |
In some processes — particularly large pipe diameters, ducts or furnaces — transmitter and receiver can be installed directly opposite each other without optical windows separating the laser from the process gas. The laser beam traverses the gas directly. This configuration is mechanically simple and eliminates a significant failure source — window fouling.
In most process applications, optical windows are required — to separate the laser optics from aggressive process gas, to bridge pressure differentials or to meet explosion protection requirements. Several practical challenges arise:
Installation angle: Window panes must not be installed perpendicular (90°) to the laser beam. The reason is the etalon effect.
The Etalon Effect — a Critical Source of Error:
When the laser beam strikes a window pane oriented parallel to the beam axis, multiple reflections between the glass surfaces create an interference pattern — similar to an optical Fabry-Pérot etalon. These interference fringes superimpose on the actual absorption signal and can significantly distort the measurement. The problem is particularly insidious because the fringes shift with temperature and vibration changes and are therefore time-variable.
The solution: window panes are installed at a slight angle — typically a few degrees from perpendicular — so that reflected beams are directed out of the detection path.
Window Coatings:
Depending on the laser wavelength used, window panes may be fitted with anti-reflection coatings to minimise reflection losses and reduce the etalon effect. The coating must be matched to the specific wavelength of the laser in use.
In dust-laden or corrosive processes, optical windows are purged with an inert gas — typically nitrogen or instrument air — to prevent deposits. A critical point must be observed:
The Path Length Problem with Window Purging:
The path length — the distance the laser beam travels through the measurement gas — directly determines the measured concentration. If the window purge flow is set too high, the purge gas can enter the measurement path and displace the process gas. The result: the effective path length through the measurement gas shortens — the measured concentration appears lower than it actually is.
In O₂ inertisation monitoring, this creates a particularly dangerous situation: excessive N₂ purge flow can cause the system to measure an O₂ concentration close to zero — not because inertisation is working, but because the nitrogen purge gas dominates the measurement path. A false low reading for O₂ in an inertisation application is safety-critical.
Practical note — costly project without a happy end: In a project involving three TDLS laser measurements (investment exceeding €500,000) for a safety-relevant measurement, a dedicated pipe section was specially designed and built into the plant to accommodate the TDLS infrastructure — a substantial mechanical effort in a running industrial facility. Before commissioning, the flow meters were tested for fouling tendency — with no finding. However, the optical windows themselves were not tested separately. This was the critical gap: after commissioning under real process conditions, exactly those windows fouled. The project concerned trace H₂O measurement in a critical process. The project ended without the intended operational readiness — despite the enormous structural and financial investment. The lesson: pre-commissioning tests must cover all relevant components of the optical path — not just the primary measurement component. And the effort invested in infrastructure makes failure all the more costly.
Practical note — O₂ measurement directly in combustion chambers: In another application, TDLS systems were installed directly in combustion chambers — under conditions of high temperature, dust and significant transmission losses due to particulate loading. Remarkably, the systems continued to operate reliably even at substantially reduced transmission. TDLS systems are more robust against transmission losses than often assumed — provided signal strength remains above the minimum detection threshold and self-diagnostics monitor the condition continuously.
The width of an absorption line depends on pressure — a physical effect known as pressure broadening. At higher pressure, the absorption line becomes broader and flatter; at lower pressure, it is narrower and steeper.
Why this matters in practice:
TDLS systems are calibrated for a defined pressure range. If the process pressure lies outside the calibration range or fluctuates significantly, this can influence the concentration calculation. Modern TDLS systems compensate for this effect through:
In practice, pressure dependence is well manageable in correctly designed systems — but it must be explicitly accounted for during system design.
| Component | Wavelength range | Typical application | Note |
|---|---|---|---|
| O₂ | 760 nm (visible) | Inertisation, combustion control, personnel protection | Strongest O₂ absorption band; in-situ preferred |
| H₂O | 1350–1950 nm | Dewpoint control, drying processes, Cl₂ electrolysis | Wherever conventional sensors fail |
| CO | 1560 nm, 2300 nm | Combustion monitoring, safety measurement | 2300 nm more sensitive but more demanding |
| NH₃ | 1500 nm, 2000 nm | SCR slip measurement, fertiliser production | Alternative to wet-chemical methods |
| HCl | 1740 nm, 3400 nm | Emissions monitoring, chlorine chemistry | TDLS increasingly type-approved for CEMS |
| HF | 1300 nm, 2500 nm | Glass industry, petrochemicals, aluminium smelters | Aggressive — in-situ avoids material compatibility issues |
| CH₄ | 1650 nm | Natural gas, biogas, leak detection | Good sensitivity in NIR |
| H₂ | 2122 nm (quadrupole transition) | Cl₂ electrolysis, hydrogen safety, electrolysis plants | Weak signal — specialised systems required, but in-situ possible |
| CO₂ | 1570 nm, 2000 nm | Fermentation, combustion monitoring, greenhouse gas reporting | High background concentration in many processes |
H₂ is traditionally considered "optically difficult to measure" — and for good physical reason. As a diatomic homonuclear molecule (both atoms identical), H₂ has no electric dipole moment. This means the strong dipole transitions that generate the absorption signal in most other gases simply do not exist for H₂.
What H₂ does have are weak electric quadrupole transitions — physically a completely different class of transitions, with a signal strength approximately one hundred times weaker than typical dipole transitions. The relevant absorption line is at approximately 2122 nm.
This extreme signal weakness made H₂ measurement by TDLS practically impossible with conventional system configurations for many years. Two developments made it viable for industrial applications:
Firstly, Wavelength Modulation Spectroscopy (WMS) — a detection technique that substantially improves signal-to-noise ratio and can extract the weak quadrupole signal from the noise.
Secondly, the industry's willingness to work with longer path lengths and specifically optimised systems — which is practically significant for H₂ measurement in Cl₂ electrolysis and in electrolysis plants for green hydrogen production.
Why H₂ safety monitoring in Cl₂ is safety-critical:
In chlor-alkali electrolysis, H₂ is an unavoidable by-product. According to Eurochlor documentation (GEST 91/168), the lower explosive limit (LEL) for H₂/Cl₂ mixtures is 2.5 to 3 vol% H₂ — depending on pressure and temperature. H₂/Cl₂ mixtures are particularly hazardous: the ignition energy is extremely low (approximately 10⁻⁷ joules) and the auto-ignition temperature is approximately 207°C — considerably lower than H₂/air mixtures (approximately 400°C). Continuous H₂ monitoring in this application is therefore a safety measurement with a direct protective function.
Why conventional measurement methods reach their limits here:
| Method | Principle | Problem in H₂/Cl₂ application |
|---|---|---|
| Process GC | Separation and detection of individual components | Cycle time too long — several minutes; too slow for protective functions during rapid concentration changes |
| Converter + TCD | H₂ and Cl₂ are converted to HCl in a heated furnace; TCD compares signal before and after converter | TCD can only distinguish two components — with additional fluctuating background components, measurement becomes unreliable. And: the heated converter is itself a potential ignition source — it cannot be operated in all plant conditions, particularly during start-up and shutdown when gas composition is uncontrolled |
| IR measurement of HCl | Indirect measurement: HCl after converter by NDIR | More robust than TCD when other components fluctuate; but shares the fundamental problem of the heated converter as ignition source — and therefore the same limitation during start-up and shutdown operations |
| TDLS direct | Direct H₂ measurement via quadrupole absorption at 2122 nm | Direct, without converter, without ignition source, without media contact — and suitable for inline installation when process geometry permits. Can in principle be operated during start-up and shutdown |
The logic of method selection: GC is too slow for a protective function. The converter approach — whether with TCD or IR — has two fundamental limitations: first, matrix dependency for TCD; second and more fundamentally, the heated converter itself as a potential ignition source. This means: in critical operating states such as start-up and shutdown — precisely when gas composition is least controlled — the converter method often cannot be used. TDLS does not have this ignition source disadvantage.
Why TDLS despite the weak signal is valuable: Conventional H₂ sensors — catalytic sensors, electrochemical sensors, thermal conductivity detectors — are in direct contact with the measurement gas. In a chlorine gas stream, the aggressive Cl₂ rapidly destroys these sensors. TDLS measures without media contact — in corrosive, explosion-hazardous and high-purity gas streams this is a decisive advantage, even if sensitivity is lower than in other TDLS applications.
Oxygen monitoring in inerted plants and vessels is one of the most important TDLS applications in the chemical industry. TDLS measures O₂ without sample conditioning, without sample pump and without contact with the process gas — and therefore without the maintenance intensity of extractive paramagnetic systems.
Practical comparison from hazardous waste incineration: In a plant with extreme conditions — high particulate loading, aggressive flue gas, moisture and sludge — O₂ monitoring had been operated with extractive paramagnetic analysers. The sample conditioning system was a constant maintenance burden: filters blocked, pumps failed, condensable components created problems throughout the sample path. Operations and maintenance personnel worked under unpleasant and sometimes hazardous conditions. A switch to TDLS in-situ would have eliminated the sample conditioning path entirely. The example illustrates: in harsh process environments, the maintenance effort for the sample conditioning system is often the real operational problem — not the measurement accuracy of the analyser.
In combustion processes where organic compounds are present in the flue gas, the zirconia sensor (ZrO₂) fails — because combustible gases consume oxygen at the sensor surface and cause false low readings. TDLS measures O₂ via the optical absorption path without media contact and is therefore immune to this effect. For such applications, TDLS is often the only practical in-situ solution.
In chlor-alkali electrolysis, hydrogen is produced as a by-product. If H₂ enters the chlorine stream, an explosive mixture forms. TDLS measures H₂ via weak quadrupole transitions in the infrared — contactlessly, rapidly and without exposure to the aggressive chlorine gas that rapidly destroys conventional sensors.
After the drying tower in chlorine electrolysis, the moisture content of the chlorine gas must be monitored in the ppm range. Even small amounts of water (above approximately 30 ppm) cause corrosion in the downstream compressor and contaminate the product.
Conventional moisture sensors based on phosphorus pentoxide (P₂O₅) fail here for a simple reason: they are in direct contact with the chlorine gas and are attacked by its extreme aggressiveness — the sensor degrades, responds increasingly slowly and requires frequent replacement. Response times of up to five minutes are not uncommon for P₂O₅ sensors.
TDLS measures H₂O in the chlorine gas stream without media contact — with response times in the order of seconds and without sensor degradation from the aggressive process gas. This is a textbook example of an application where TDLS is not merely better than the alternative, but where the alternative fails to function sustainably.
TDLS systems for HCl and HF are increasingly receiving type approvals for CEMS use in Germany and across the EU — replacing wet-chemical methods. In the US, EPA Performance Specifications are beginning to recognise TDLS as an acceptable method for certain CEMS applications. In the UK, MCERTS certification is required for CEMS use. These systems are lower-maintenance, respond faster and measure continuously rather than discontinuously.
Confidence in TDLS systems built over 30 years of operational experience means they are now regularly used in SIL-classified measurement loops. Typical configurations:
1oo2 configuration (1 out of 2): Two TDLS systems — often from the same manufacturer — measure the same variable. Both systems must independently trigger for a protective action to be initiated. This configuration offers high availability combined with redundancy.
2oo3 configuration (2 out of 3): Three TDLS systems, with two needing to agree for activation. Particularly used for O₂ inertisation monitoring in highly critical facilities.
Why TDLS is suited for SIL applications:
| SIL-relevant property | TDLS advantage |
|---|---|
| No consumables | No failures from empty gas cylinders, worn electrodes or depleted reagents |
| Integrated self-diagnostics | Laser power and transmission signal continuously monitored — failure is detected |
| No sample conditioning system | Fewer components in the safety loop — fewer failure sources |
| Fast response time | Seconds rather than minutes — critical for rapid protective response |
| Selectivity | Low cross-sensitivity reduces risk of undetected false measurements |
Important note: With TDLS systems from the same manufacturer in redundant configurations, a residual risk of systematic failures exists — both systems could be affected by the same device-specific fault. At the highest safety requirements, diverse redundancy may be appropriate — for example TDLS combined with paramagnetic measurement for O₂. For US applications, SIL assessment should follow IEC 61511 as required by OSHA PSM (29 CFR 1910.119) RAGAGEP compliance.
Legend: ✅ TDLS preferred | ⚠️ TDLS possible, but check alternatives | ❌ TDLS not suitable
| Application | TDLS | Reason / Comment |
|---|---|---|
| O₂ inertisation monitoring | ✅ | No sample conditioning system, SIL-capable |
| O₂ combustion control | ✅ | Fast, in-situ, also with organic compounds present |
| O₂ with organic compounds in flue gas | ✅ | ZrO₂ fails — TDLS unaffected by combustible gases |
| H₂O in Cl₂ electrolysis | ✅ | P₂O₅ sensor fails — TDLS without media contact |
| H₂ in Cl₂ safety monitoring | ✅ | Contactless, fast, Cl₂-resistant |
| NH₃ slip measurement (SCR) | ✅ | Fast, in-situ, low maintenance |
| HCl emissions monitoring (CEMS) | ✅ | Increasingly type-approved for CEMS use |
| HF measurement | ✅ | Aggressive medium — no contact advantageous |
| CO in combustion monitoring | ✅ | In-situ possible, fast |
| Multicomponent gas analysis | ❌ | One component per laser — GC or FTIR more suitable |
| Liquid measurement | ❌ | TDLS is for gas measurement — not for liquids |
| Heavily particulate-laden gases | ⚠️ | Monitor transmission — dimension window purging critically |
In the US, TDLS is recognised as an acceptable measurement technique for a growing number of process safety and emissions applications. For CEMS under 40 CFR Part 60, acceptance depends on the specific Performance Specification (PS) applicable to the measured component. For safety instrumented systems, IEC 61511 compliance — as required under OSHA PSM RAGAGEP — provides the framework for SIL assessment of TDLS-based measurement loops.
For H₂ monitoring in chlor-alkali facilities, OSHA Process Safety Management (29 CFR 1910.119) and EPA Risk Management Program (40 CFR Part 68) requirements apply. TDLS provides the continuous, rapid-response monitoring that these regulations demand for highly hazardous chemicals.
In the UK, TDLS systems for CEMS use require MCERTS certification (Environment Agency Monitoring Certification Scheme). For SIL applications, IEC 61511 applies as the relevant functional safety standard, with HSE guidance providing additional context for process industry applications. ATEX certification requirements apply for hazardous area installations; post-Brexit, UKCA marking is required for the UK market.
| Standard / Regulation | Content |
|---|---|
| IEC 61508 / IEC 61511 | Functional safety — framework for SIL assessment of TDLS systems |
| EN 14181 | Quality assurance for automated measuring systems — relevant for TDLS in CEMS |
| EN 15267 | Certification of automated measuring systems — type approval for CEMS applications |
| NAMUR NE 107 | Self-monitoring and diagnostics of field devices — requirements for TDLS self-diagnostics |
| ATEX / IEC 60079 | Explosion protection requirements for TDLS systems in hazardous areas |
| UKCA | UK conformity assessment post-Brexit — replaces ATEX CE marking for UK market |
| 40 CFR Part 60 | US EPA — CEMS performance specifications applicable to TDLS |
| OSHA PSM (29 CFR 1910.119) | US process safety management — RAGAGEP framework for SIL systems |
| MCERTS | UK Environment Agency — certification scheme for CEMS including TDLS |
What does TDLS mean and how does it work?
TDLS stands for Tunable Diode Laser Spectroscopy. A diode laser is tuned to a characteristic absorption line of the target molecule and traverses the process gas. The more target molecules are present, the more light is absorbed. From the measured attenuation, the path length and process parameters, the concentration is calculated — without sample extraction and without media contact.
What is the etalon effect and why is it critical for TDLS?
The etalon effect occurs when the laser beam strikes parallel surfaces — typically window panes installed perpendicular to the beam. Multiple reflections between the glass surfaces generate interference fringes that superimpose on the absorption signal and can distort measurements. The solution: window panes are installed at a slight angle so that reflected beams are directed out of the detection path.
Why is window purging so critical in TDLS?
Window purging keeps optical windows clean — but also risks causing a false measurement if incorrectly dimensioned. Excessive purge flow can push the purge gas into the measurement path and displace the process gas. The measured path length through the process gas shortens, and the displayed concentration drops. In O₂ inertisation applications, a falsely low O₂ reading caused by excessive N₂ purge flow would be safety-critical.
Why is TDLS better suited than conventional sensors for H₂O measurement in Cl₂ electrolysis?
Conventional moisture sensors based on phosphorus pentoxide (P₂O₅) are in direct contact with the aggressive chlorine gas and are degraded by it — they wear rapidly, respond slowly and require frequent replacement. TDLS measures the H₂O content optically without any media contact — with response times in the order of seconds and without sensor degradation from the process gas.
Why is TDLS better than ZrO₂ for O₂ measurement with organic compounds in flue gas?
Zirconia sensors measure O₂ electrochemically at a hot ceramic sensor surface. Combustible by-products (CO, H₂, hydrocarbons) are oxidised at this surface and consume oxygen in the process — the sensor falsely indicates less O₂ than is actually present. TDLS measures via the optical absorption path without media contact and is therefore immune to this effect.
Which SIL configurations are used with TDLS?
In practice, 1oo2 and 2oo3 configurations are most common. For O₂ inertisation monitoring, 2oo3 is widespread. Systems from the same manufacturer are often combined — advantageous for maintenance and calibration but leaving a residual risk of systematic failures. At the highest requirements, diverse redundancy with two different measurement principles may be appropriate.
Can H₂ really be measured optically? I thought that was not possible.
H₂ as a homonuclear diatomic molecule has no electric dipole moment — which means the strong dipole transitions present in most gases do not exist for H₂. What H₂ does have are very weak electric quadrupole transitions at approximately 2122 nm. These transitions are roughly 100 times weaker than typical dipole transitions. Through wavelength modulation spectroscopy (WMS) and optimised longer path lengths, these weak signals can be detected reliably enough for industrial applications — contactlessly and without media exposure to corrosive gases.
Why can the converter method not be used during start-up and shutdown in Cl₂ electrolysis?
The heated converter furnace that converts H₂ and Cl₂ to HCl for measurement is itself a potential ignition source. During start-up and shutdown, the gas composition is least controlled and the risk is highest — precisely the conditions under which an ignition source cannot be tolerated. TDLS measures without any ignition source in the measurement path and can in principle be operated continuously across all plant states.
How robust is TDLS when transmission is reduced by dust or deposits?
TDLS systems monitor laser transmission continuously as part of self-diagnostics. When transmission falls below a defined threshold, an alarm is triggered. In practice, TDLS systems have continued to measure reliably even at substantially reduced transmission — for example in combustion chambers with particulate loading. There is, however, a physical limit: when transmission drops too low, valid measurement is no longer possible and the system must be treated as failed.
TDLS instrument types and system architectures
Different configurations — single-pass vs. multi-pass, extractive TDLS cells, cross-duct vs. probe configurations — as well as an overview of available system concepts are covered in a separate article.
Further reading: "NIR and Raman Analysers for PAT — Instrument Types, System Concepts and Selection Criteria" (forthcoming)
Detail physics: harmonic detection and wavelength modulation
The physical principles of wavelength modulation spectroscopy (WMS), 2f/1f detection and related techniques are a specialist subject not covered in this overview article.
TDLS has evolved over 30 years from a demanding laboratory technique into one of the most reliable and lowest-maintenance measurement principles in industrial process gas analytics. The combination of in-situ measurement without media contact, high selectivity, integrated self-diagnostics and a growing body of operational experience makes TDLS the first choice for a growing range of process tasks — from O₂ inertisation monitoring to H₂O measurement in aggressive chlorine gas streams.
The practical challenges — window fouling, etalon effect, path length variation from purging, pressure dependence — are solvable, but they must be explicitly addressed during system design. A TDLS system is only as good as its installation and the care with which the specific process boundary conditions have been analysed.
Analytic Journal — The Portal for PAT
© 2026 Analytic Journal / Dusko Kadijevic. All rights reserved.
This article is protected by copyright. Short excerpts with source attribution and an active link to analyticjournal.com are permitted. Full or substantial reproduction — including in digital form, by AI systems or automated processes — requires the prior written consent of the publisher. Enquiries: info@analyticjournal.de
Related topics on Analytic Journal:
Editorial Note
This article has been written to the best of our knowledge based on publicly available technical information, standards, and manufacturer data. Specifications may change — for binding information, consult current manufacturer datasheets and relevant standards bodies. Analytic Journal produces editorial content independently of manufacturers and advertising partners.
Author: Dusko Kadijevic — 20 years of experience as a PAT practitioner in the process industry
Dusko Kadijevic | Publisher Analytic Journal | analyticjournal.com
Written May 2026. All content formulated in our own words, created with AI assistance.
First published on 8 June 2026 on analyticjournal.com