Question
I am trying to desorb epichlorohydrin monomer from epoxy resin samples and am getting poor precision.  Sometimes the desorption works well and I get minimal carryover, but sometimes its like the sample has hardly been desorbed at all. Any suggestions??

Answer
There are two main possibilities for what could be going wrong in this case: 

It could be that too large a sample of resin is being taken such that the flow path through the sample tube is sometimes inadvertently blocked.  Thermal desorption is a dynamic process.  It requires both heat and a flow of inert gas.  If the sample of resin is too large (say greater than 50 mg in the case of a typical 1/4-inch metal tube), it may block the tube thus restricting or even stopping both the flow of gas and the thermal desorption process. 

Solution: Reduce sample size and check, by looking down the tube after every sample is introduced, to make sure that the carrier gas flow path is unrestricted. 

Another possible cause of the poor precision observed, is variable positioning of the resin sample inside the tube relative to the heated zone of the desorber.  On most commercial thermal desorbers, the tube oven does not extend to the very ends of the sample tube.  Consequently, samples placed at the ends of a tube will not be fully within the heated zone. Poor desorption efficiency and carryover will result and could cause the type of poor precision described. 

Solution: Ensure that the sample is positioned inside the tube such that it is well inside the heated zone of your desorber. 

Note that we recommend use of inert, sample tube liners for direct desorption of materials such as resins.  Liners simplify sampling procedures, help reduce errors and prevent desorbed samples from irreversibly contaminating sample tubes. PFTE liners are also available from Markes International in packs of 10 and 100 using part numbers C-PIO10 and C-PI100 respectively. 



Question
I'm interested in finding out more about thermal desorption compatible diffusive sampling for monitoring personal exposure. Please could you tell me where to get more information?

Answer
It would certainly be advisable to get yourself onto the distribution list of a journal entitled "The Diffusive Monitor".  This is published roughly once a year by HSE/CAR Working Group 5 and is a mine of useful information.  For further information please click here

The original 1981 paper describing a diffusive tube sampler for thermal desorption - "The development of an improved diffusive sampler." (Saunders KJ, Charlton J and Brown RH Am. Ind. Hyg. Assoc. J., 42, pp 865-869, (1981)) - is still regarded as essential reading and provides great background information and useful performance data.

There are a number of international standard methods which are also recommended reference material:

CEN Std EN 838: Workplace atmospheres - Diffusive samplers for the determination of gases and vapours - Requirements and test methods.
UK Health and Safety Executive Standard: Methods for the Determination of Hazardous substances (MDHS) 80: Volatile organic compounds in air. Laboratory method using diffusive solid sorbent tubes, thermal desorption and gas chromatography.
ISO DIS 16017 Part 2 Air Quality – Sampling and analysis of VOCs in ambient, indoor and workplace air by sorbent tube / thermal desorption / capillary GC. – Diffusive Sampling.
pr EN 13528 Ambient Air Quality – Diffusive samplers for the determination of concentrations of gases and vapours. 

Further information about these standards can be obtained from our technical note:

TDTS 03 - Relevant national and international standard methods



Question
The chromatograms I am obtaining from my thermal desorption analysis have deteriorated and I am now getting very poor peak shapes from my components of interest. What is going wrong?

Answer
There are a number of different reasons for poor peak shapes and some of them give quite different and distinctive symptoms. 

If the first peaks in your chromatogram have broadened or are beginning to tail, this can indicate that the cold trap sorbent needs changing.  On UNITY we recommend that the sorbent is changed every 1000 desorptions (remember to count "Heat Trap" operations as one desorption) or once every 12 months, whichever is the sooner.  If your cold trap sorbent is continually operated at or very close to the maximum temperature of the sorbent you may find that it will require replacing more frequently.  

For further information on maximum temperatures of sorbents see 

TDTS 05 - Advice on sorbent selection and conditioning sample tubes 

The desorption of tubes containing aggressive compounds can produce activity in the transfer line or in the analytical column itself and this will result in peak broadening or tailing, particularly of more active or polar compounds. A direct liquid injection onto the analytical column will show whether the column is at fault.

A poor connection between the transfer line and column will also distort peak shapes - check that the connecting ends of both transfer line and column have been cleanly cut. A fused silica column cutting tool and a small magnifying glass are essential tools. The union or connector assembly should be an inert, zero dead volume fitting recommended for butt connecting capillary tubing and the two ends of the fused silica tubing should touch inside the union. If the connection is found to be sound and peak tailing persists, replace the deactivated fused silica insert inside the transfer line ensuring that it is correctly installed to the thermal desorber.

Broad peaks are also the result of the selection of too strong an adsorbent in the cold trap (e.g. using a carbon sorbent such as spherocarb for the analysis of long chain hydrocarbons such as decane). Broad peaks will appear if the gas flow through the cold trap during secondary desorption (trap heat) is too low. This gas flow is the sum of the column flow and the outlet split flow and for UNITY this should exceed 2ml/min for optimum peak widths. Note however that may thermal desorption systems require a much higher minimum trap desorption flow - often as much as 10 to 20 ml/min.

Finally remember the loading capacity of your analytical column - overloading will cause band broadening and tailing. High resolution capillary columns work at optimum with analyte masses in the order of 20 to 200ng - use one or both splitters if necessary to ensure that the column is not overloaded. 


Question
I am using a GC-MS system with my thermal desorber and am seeing a high air/water background. Why is this?

Answer
UNITY uses high quality o-rings to create a strong seal around the sample tubes and cold trap and therefore the air/water background on a UNITY - GC-ECD or GC-MS system should be similar to that observed with conventional GC injector systems.  However, as with all split GC injectors it is recommended that a positive flow of at least 10ml/min is maintained through the split vent at all times to prevent back diffusion of air through the splitter. UNITY software allows the selection of "Split on in Standby" to facilitate this.  Note however that other thermal desorption systems may automatically switch the split flow off after analysis thus causing the problem. 




Question
I am monitoring benzene, toluene and xylene in humid ambient air and calculate that I need to sample at least 3 L of air to get sufficient sensitivity.  This volume of air contains tens of milligrams of water. Won't this saturate my analytical system?

Answer
The easiest solution in your case is to use only hydrophobic sorbents in the sampling tubes.  Do not use multi purpose tubes as these usually contain a backing bed of less hydrophobic, strong sorbent material. 

Tenax TA™ is one of the most hydrophobic of all sorbents. Its retention volume is barely affected by the presence of water and less than 1 mg is typically retained on a standard tube, however humid the air. Standard tubes packed with Tenax TA have a safe sampling volume of 6 L for benzene, more for toluene and xylene.  This is comfortably within the requirements of your monitoring program even when allowing for minor downward adjustment to account for 100% humidity levels.

By selecting a hydrophobic sorbent, you will essentially prevent water from reaching the analytical system in the first place.




Question
When should I use silanised and when unsilanised glass wool inside tubes - I have heard conflicting advice on this issue?

Answer
This issue can be confusing. If there is any requirement for desorption temperatures in excess of 250°C, it is best to use unsilanised, conditioned glass wool or quartz wool.  This is because residual silanising reagent can break down at high temperatures (>275°C) coating the instrument flow path with active degradation products and limiting its performance - especially with regard to high boilers. 

However, when monitoring extremely labile compounds such as mercaptans, there is no choice but to use silanised glass wool. In these cases desorption temperatures are invariably limited anyway to prevent thermal degradation of the target analytes. 


Further information on thermal desorption techniques and applications is available through our comprehensive library of technical publications

Back to Technical Zone home page