Enhancing Air Monitoring Methods with Thermal Desorption

Liz Woolfenden, Director, Markes International, UK.
  • Accurately monitor down to ppb/ppt levels.
  • Use thermal desorption tubes for either passive or active sampling, without modification.
  • Compliant with air sampling methods.


Use of active sampling onto glass tubes packed with charcoal, followed by carbon disulfide (CS2) extraction and gas chromatography (GC) analysis, was developed as an air monitoring method for vapor-phase organic compounds (VOCs) in the 1970s. The approach is still used today for some personal exposure assessment (occupational hygiene) applications and stack emission testing, but is fundamentally limited with respect to detection limits. Thermal desorption (TD) is a complementary gas extraction technique whereby sorbent tubes (Figure 1) are heated in a flow of carrier gas. Trapped vapours desorb from the sample tubes into the gas stream and are transferred, via a refocusing device, into the GC/MS analyzer.

TD offers much better sensitivity than solvent extraction (see below) and has now almost universally superseded charcoal/CS2 for environmental (ambient and indoor) air monitoring. Steady reductions in exposure limit levels and new restrictions on chemicals such as CS2, in Europe and elsewhere, have also led to increased use of thermal desorption for occupational hygiene, i.e. for exposure assessment in the workplace [1]. The most recent international standard methods for thermal desorption include workplace air monitoring in the scope [2,3].

The trend towards thermal desorption for all air monitoring applications (workplace, indoor and ambient air) has been further encouraged by recent TD technical developments. The latest commercial thermal desorbers now allow quantitative recollection of split flow (both tube and trap desorption split flow) for repeat analysis. This overcomes the one-shot limitation of traditional TD methods and simplifies method/data validation.

The following article summarizes the key advantages of thermal desorption versus solvent extraction for monitoring organic vapors in air and explains why TD is preferred in most cases.

Figure 1  A selection of thermal desorption-compatible air sampling tubes from the new Restek range.

air sampling tubes


Solvent extraction of charcoal tubes requires at least 1 or 2ml of CS2 followed by injection of only 1-2µl of extract into the GC/MS. This results in a 103 dilution of the sample right at the start of the process. Other factors limiting sensitivity include, solvent artifacts, interference from the solvent itself (masking volatile target analytes) and low desorption efficiency (see below). Conversely, thermal desorption allows complete transfer of all target analytes to the analytical system, with no dilution or solvent interference. Detection limits offered by thermal desorption methods are typically at least 1000 times higher than equivalent solvent extraction methods, facilitating ambient monitoring at ppt/ppb levels as well as higher ppm (and %-level) concentrations. By comparison, charcoal/CS2 methods are invariably limited to concentrations above 0.1 ppm.

Desorption Efficiency

Thermal desorption efficiency is readily validated and is always above 95%, independent of ambient conditions and the nature of the target analytes (polar/nonpolar, volatile/semi-volatile, etc.). Conversely, the desorption efficiency of charcoal /CS2 extraction methods is only about 80%, even under best case conditions—i.e. with volatile, nonpolar target compounds collected from dry atmospheres. However, charcoal is hydrophyllic and adsorbs significant masses of water from humid air. The presence of water can reduce desorption efficiency (e.g. to 20-30%), especially in the case of polar compounds. The user may not even be aware that this problem has occurred, yet it can cause atmospheric concentrations to be under-reported by as much as a factor of 4. Desorption efficiencies for semivolatile compounds are also low—often below 50%.


As described above, the desorption efficiency of solvent extraction is significantly worse than that of TD and can vary from 20 to 80% depending on analyte volatility and atmospheric humidity. This significantly compromises reproducibility. Other factors contributing to analytical uncertainty/error include the evaporation of CS2 during sample preparation and its absorption into the rubber septa of GC autosampler vial caps.

Variability of solvent extraction data can also be caused by compounds coeluting with the solvent. CS2 is almost 'invisible' by FID detection. However, the FID response to compounds which coelute with CS2 will nevertheless be quenched by the relatively large concentration of solvent. Quenching effects of this sort are notoriously variable and will result in poor precision for the compounds affected. Solvent interference can have an even more dramatic effect on GC methods using mass spectrometer (MS) detection (see below).

Cost Saving

Thermal desorption offers enhanced automation and greatly reduced running costs. Tubes are reusable at least 100 times (typically >200 times). TD also eliminates solvent purchase and disposal concerns.

While automated thermal desorption systems typically have a higher initial price tag than liquid autosamplers for GC, most thermal desorbers also replace the need for a conventional GC injector. In addition, every GC system used for the analysis of CS2 extracts requires installation of sophisticated ventilation equipment to minimize the health and safety risk to laboratory personnel (see below). These costs vary, but can bring the total capital cost of automated liquid injection-GC into the same ball park as automated TD-GC once suitable ventilation is taken into consideration.

Manual thermal desorption systems offer affordable entry-level TD but without compromising performance or capability. The latest Markes UNITY thermal desorber for example (Figure 2), offers method-compliant, cryogen-free TD-GC/MS analysis of sorbent tubes together with quantitative recollection of split flow for repeat analysis and is roughly half the cost of a fully automated system.


Figure 2: UNITY single tube thermal desorber from Markes International, featuring repeat analysis.

Perhaps the major cost eliminated by thermal desorption is that of manual sample preparation. TD tubes arrive at the laboratory ready for analysis. In contrast, charcoal tubes require a relatively lengthy sample preparation procedure. First they have to be broken in order to tip the two separate beds of charcoal into individual vials. 1 or 2 ml of CS2, containing suitable internal standard(s), is then added to each vial and the vials are subsequently capped. These samples are typically agitated for at least 30 minutes before the supernatant liquid is decanted across to a second set of vials which again require capping before being placed on the autosampler. All of these procedures are manual, time-consuming, very difficult to automate, and a potential source of sample loss and error.

Passive Sampling Option

While thermal desorption tubes are used extensively for active air sampling, they are also compatible with low-cost passive (diffusive) sampling (Figure 3). Extensive data is now available for quantitative passive sampling using standard TD tubes and this can mean major cost savings for applicable studies [2,4,5]. Passive samplers eliminate the requirement for personal monitoring pumps making them much less heavy/intrusive. Instead of a pump, each tube is simply fitted with a diffusion cap at the sampling end.

Monitoring workers using lightweight passive samplers minimizes risk of individuals modifying their behavior and facilitates more representative measurement. The combination of low sampling costs and automated TD-GC/MS analysis also make passive sampling the method of choice for large scale ambient air monitoring campaigns, e.g. across an entire city [6].
[NOTE: Thermal desorption tubes may be used for either active or passive sampling without modification.]

air sampling tube in passive mode

Figure 3: Thermal desorption tube being used in passive (diffusive) sampling mode.

Solvent Interference

Originally, charcoal/CS2 methods were intended for use with packed column GC technology and FID detection. In this case, the limitations of carbon disulfideare minimized by its very low response on FID. However, even under these conditions, impurities in the solvent, solvent-related baseline disturbances, and the large dilution factor (see above) all contribute to limit method sensitivity—typically to ppm-level atmospheric concentrations. Solvent effects also increase analytical uncertainty/error. With the modern preference for GCs configured with mass spectrometer (MS) detectors, CS2 brings additional limitations. It generates a large response on the MS, often requiring deactivation of the detector ionizers until after the solvent has completely passed through the system. This makes it impossible to analyze target compounds which coelute with the solvent.

Environmental Health and Safety

CS2 is one of the most toxic common solvents. It is banned completely by some countries/companies and the latest international occupational exposure limit values have dropped to 5 ppm which effectively means average workplace air concentrations should not exceed 0.5 ppm to minimize exposure risk [1]. As described above, solvent extraction of charcoal requires multiple manual operations leading to significant risk of personal exposure to CS2 vapor. In many cases, laboratory staff carrying out CS2 extraction procedures may be at higher risk of hazardous chemical exposure than those workers whose samples they are trying to analyze. Personal monitoring of affected laboratory staff for CS2 is, therefore, recommended.

In addition to this, carbon disulfide has an extremely unpleasant odor, leading to increased discomfort in the working environment. Moreover, new environmental legislation means that solvent disposal is an increasingly costly and administration-intensive process. By eliminating the need for the solvent, TD completely overcomes all these issues.

Repeat Analysis

Historically, the only real limitation of thermal desorption, relative to charcoal/CS2 methods, was that it was one-shot. With early TD equipment, once the sample tube was desorbed (i.e. heated in a stream of carrier gas), it was gone. Therefore, if anything went wrong with the subsequent analysis there was no chance to repeat the run. However, since the advent of SecureTD-Q (quantitative re-collection of split flow during both tube and trap desorption), this is no longer the case.

TD-GC/MS methods for routine workplace, indoor and ambient air monitoring invariably employ a single or double split. Quantitative recollection of any and all split flow allows samples to be archived indefinitely (most compounds are stable on sorbent tubes for several months), used for third party validation or reanalyzed immediately to confirm results (Figure 4). In other words, SecureTD-Q allows repeat TD-GC/MS analysis of all but the lowest-concentration (ppt-level) samples—i.e. those which are analyzed splitless.

Figure 4  Use of SecureTD-Q to repeat the TD-GC/MS analysis of a sample of phthalate esters.

repeated TD-GCMS analysis of phthalate esters

The utility of quantitative sample re-collection for repeat TD-GC/MS analysis has recently been recognized in standard methods as an aid to TD method/data validation [3].
[NOTE: Repeat analysis was always a slightly dubious claim for charcoal/CS2 solvent extraction methods. CS2 is an extremely volatile solvent, which is rapidly lost through both evaporation and absorption into the rubber septa of the vial caps. If the first sample of a sequence is reanalyzed at the end, different results are obtained. Even refrigerated storage may not guarantee sample stability because of the absorptive losses of CS2.]

Method Compliance

Another historical advantage for solvent extraction was the number of applicable standard methods. Though equivalent thermal desorption methods were initially slow to be promulgated, this situation has now completely changed. Well-validated thermal desorption methods, describing both active and diffusive sampling and applicable to workplace monitoring, ambient air, indoor air and materials emissions testing, are now available from all the major international standards agencies. Key examples include: EN ISO 16017, ISO 16000-6, ASTM D-6196, US EPA Method TO-17, NIOSH 2549, MDHS 72, 80, etc. (UK) and EN 14662.


Thermal desorption technology offers several significant advantages over conventional solvent extraction. TD systems offer better sensitivity, desorption efficiency, and reproducibility compared to charcoal/CS2 systems. Additionally, tubes may be used for both passive and active sampling without modification. These benefits, in combination with SecureTD-Q technology, which allows repeat analysis, make thermal desorption an excellent choice for many air monitoring applications.


[1] 3rd Industrial Occupational Exposure Limit Value (IOELV) Directive Update, European Commission 2007
[2] EN ISO 16017
[3] ASTM D6196-03
[4] EN ISO 16017-2
[5] EN 14662-4
[6] Gonzalez-Flesca et al. Benzene Exposure Assessment at Indoor, Outdoor, and Personal Levels. The French Contribution to the Life MACBETH Programme. Environmental Monitoring and Assessment 65: 59-67 (2000).


SecureTD-Q is a trademark of Markes International, UK.

Restek Domestic Customer Service



Your Full Name

Your Email

Company Name


Spam Block (Please leave this blank)
Drag and Drop Attachments Here (optional)

all fields required

Thank you

Your message has been sent. We will be in touch shortly.

Message not sent

Sorry, your message could not be sent at this time. Please try again later, or contact Restek or your local Restek representative via phone.